davanstrien/pytorchmodelmetadata · Datasets at Hugging Face

modelId stringlengths 4 112	sha stringlengths 40 40	lastModified stringlengths 24 24	tags list	pipeline_tag stringclasses 29 values	private bool 1 class	author stringlengths 2 38 ⌀	config null	id stringlengths 4 112	downloads float64 0 36.8M ⌀	likes float64 0 712 ⌀	library_name stringclasses 17 values	__index_level_0__ int64 0 38.5k	readme stringlengths 0 186k
bhavikardeshna/xlm-roberta-base-german	530e8c3dd543800078c3dfbfcde30c883480f258	2021-12-21T11:40:35.000Z	[ "pytorch", "xlm-roberta", "question-answering", "arxiv:2112.09866", "transformers", "autotrain_compatible" ]	question-answering	false	bhavikardeshna	null	bhavikardeshna/xlm-roberta-base-german	15	null	transformers	9,500	# BibTeX entry and citation info ``` @misc{pandya2021cascading, title={Cascading Adaptors to Leverage English Data to Improve Performance of Question Answering for Low-Resource Languages}, author={Hariom A. Pandya and Bhavik Ardeshna and Dr. Brijesh S. Bhatt}, year={2021}, eprint={2112.09866}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```
bs-modeling-metadata/website_metadata_exp_1_model_100k_checkpoint	875249e5fc9357a93f6eab4461688b3ac18d40dc	2021-10-07T13:32:07.000Z	[ "pytorch", "gpt2", "text-generation", "transformers" ]	text-generation	false	bs-modeling-metadata	null	bs-modeling-metadata/website_metadata_exp_1_model_100k_checkpoint	15	1	transformers	9,501	Entry not found
bsingh/roberta_goEmotion	af498dcbab4ef49f7163cac455aa0d34ae7d25d8	2021-10-11T00:26:09.000Z	[ "pytorch", "roberta", "text-classification", "en", "dataset:go_emotions", "transformers", "emotions", "license:mit" ]	text-classification	false	bsingh	null	bsingh/roberta_goEmotion	15	null	transformers	9,502	--- language: en tags: - text-classification - pytorch - roberta - emotions datasets: - go_emotions license: mit widget: - text: "I am not feeling well today." --- ## This model is trained for GoEmotions dataset which contains labeled 58k Reddit comments with 28 emotions - admiration, amusement, anger, annoyance, approval, caring, confusion, curiosity, desire, disappointment, disapproval, disgust, embarrassment, excitement, fear, gratitude, grief, joy, love, nervousness, optimism, pride, realization, relief, remorse, sadness, surprise + neutral ## Training details: - The training script is provided here: https://github.com/bsinghpratap/roberta_train_goEmotion - Please feel free to start an issue in the repo if you have trouble running the model and I would try to respond as soon as possible. - The model works well on most of the emotions except: 'desire', 'disgust', 'embarrassment', 'excitement', 'fear', 'grief', 'nervousness', 'pride', 'relief', 'remorse', 'surprise'] - I'll try to fine-tune the model further and update here if RoBERTa achieves a better performance. - Each text datapoint can have more than 1 label. Most of the training set had 1 label: Counter({1: 36308, 2: 6541, 3: 532, 4: 28, 5: 1}). So currently I just used the first label for each of the datapoint. Not ideal but it does a decent job. ## Model Performance ============================================================<br> Emotion: admiration<br> ============================================================<br> GoEmotions Paper: 0.65<br> RoBERTa: 0.62<br> Support: 504<br> ============================================================<br> Emotion: amusement<br> ============================================================<br> GoEmotions Paper: 0.80<br> RoBERTa: 0.78<br> Support: 252<br> ============================================================<br> Emotion: anger<br> ============================================================<br> GoEmotions Paper: 0.47<br> RoBERTa: 0.44<br> Support: 197<br> ============================================================<br> Emotion: annoyance<br> ============================================================<br> GoEmotions Paper: 0.34<br> RoBERTa: 0.22<br> Support: 286<br> ============================================================<br> Emotion: approval<br> ============================================================<br> GoEmotions Paper: 0.36<br> RoBERTa: 0.31<br> Support: 318<br> ============================================================<br> Emotion: caring<br> ============================================================<br> GoEmotions Paper: 0.39<br> RoBERTa: 0.24<br> Support: 114<br> ============================================================<br> Emotion: confusion<br> ============================================================<br> GoEmotions Paper: 0.37<br> RoBERTa: 0.29<br> Support: 139<br> ============================================================<br> Emotion: curiosity<br> ============================================================<br> GoEmotions Paper: 0.54<br> RoBERTa: 0.48<br> Support: 233<br> ============================================================<br> Emotion: disappointment<br> ============================================================<br> GoEmotions Paper: 0.28<br> RoBERTa: 0.18<br> Support: 127<br> ============================================================<br> Emotion: disapproval<br> ============================================================<br> GoEmotions Paper: 0.39<br> RoBERTa: 0.26<br> Support: 220<br> ============================================================<br> Emotion: gratitude<br> ============================================================<br> GoEmotions Paper: 0.86<br> RoBERTa: 0.84<br> Support: 288<br> ============================================================<br> Emotion: joy<br> ============================================================<br> GoEmotions Paper: 0.51<br> RoBERTa: 0.47<br> Support: 116<br> ============================================================<br> Emotion: love<br> ============================================================<br> GoEmotions Paper: 0.78<br> RoBERTa: 0.68<br> Support: 169<br> ============================================================<br> Emotion: neutral<br> ============================================================<br> GoEmotions Paper: 0.68<br> RoBERTa: 0.61<br> Support: 1606<br> ============================================================<br> Emotion: optimism<br> ============================================================<br> GoEmotions Paper: 0.51<br> RoBERTa: 0.52<br> Support: 120<br> ============================================================<br> Emotion: realization<br> ============================================================<br> GoEmotions Paper: 0.21<br> RoBERTa: 0.15<br> Support: 109<br> ============================================================<br> Emotion: sadness<br> ============================================================<br> GoEmotions Paper: 0.49<br> RoBERTa: 0.42<br> Support: 108
cardiffnlp/bertweet-base-stance-atheism	8a4275b426ee8d4136b36ed826bd3feb2dc41f3c	2021-05-20T14:53:17.000Z	[ "pytorch", "tf", "jax", "roberta", "text-classification", "transformers" ]	text-classification	false	cardiffnlp	null	cardiffnlp/bertweet-base-stance-atheism	15	null	transformers	9,503
chrommium/rubert-base-cased-sentence-finetuned-headlines_X	3ff3429c5539d43e2a02328421cf8204c67695e4	2021-09-16T00:34:06.000Z	[ "pytorch", "tensorboard", "bert", "text-classification", "transformers", "generated_from_trainer", "model-index" ]	text-classification	false	chrommium	null	chrommium/rubert-base-cased-sentence-finetuned-headlines_X	15	null	transformers	9,504	--- tags: - generated_from_trainer metrics: - accuracy model-index: - name: rubert-base-cased-sentence-finetuned-headlines_X results: - task: name: Text Classification type: text-classification metrics: - name: Accuracy type: accuracy value: 0.952 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # rubert-base-cased-sentence-finetuned-headlines_X This model is a fine-tuned version of [DeepPavlov/rubert-base-cased-sentence](https://huggingface.co/DeepPavlov/rubert-base-cased-sentence) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.2535 - Accuracy: 0.952 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\| \| No log \| 1.0 \| 157 \| 0.2759 \| 0.912 \| \| No log \| 2.0 \| 314 \| 0.2538 \| 0.936 \| \| No log \| 3.0 \| 471 \| 0.2556 \| 0.945 \| \| 0.1908 \| 4.0 \| 628 \| 0.2601 \| 0.95 \| \| 0.1908 \| 5.0 \| 785 \| 0.2535 \| 0.952 \| ### Framework versions - Transformers 4.10.2 - Pytorch 1.9.0+cu102 - Datasets 1.12.1 - Tokenizers 0.10.3
classla/bcms-bertic-frenk-hate	f41364e4a917e75feddf03e7525b9aa001650aca	2022-06-01T09:31:46.000Z	[ "pytorch", "bert", "text-classification", "hr", "arxiv:1906.02045", "transformers", "hate-speech" ]	text-classification	false	classla	null	classla/bcms-bertic-frenk-hate	15	null	transformers	9,505	--- language: "hr" tags: - text-classification - hate-speech widget: - text: "Potpredsjednik Vlade i ministar branitelja Tomo Medved komentirao je Vladine planove za zakonsku zabranu pozdrava 'za dom spremni'." --- # bcms-bertic-frenk-hate Text classification model based on [`classla/bcms-bertic`](https://huggingface.co/classla/bcms-bertic) and fine-tuned on the [FRENK dataset](https://www.clarin.si/repository/xmlui/handle/11356/1433) comprising of LGBT and migrant hatespeech. Only the Croatian subset of the data was used for fine-tuning and the dataset has been relabeled for binary classification (offensive or acceptable). ## Fine-tuning hyperparameters Fine-tuning was performed with `simpletransformers`. Beforehand a brief hyperparameter optimisation was performed and the presumed optimal hyperparameters are: ```python model_args = { "num_train_epochs": 12, "learning_rate": 1e-5, "train_batch_size": 74} ``` ## Performance The same pipeline was run with two other transformer models and `fasttext` for comparison. Accuracy and macro F1 score were recorded for each of the 6 fine-tuning sessions and post festum analyzed. \| model \| average accuracy \| average macro F1 \| \|----------------------------\|------------------\|------------------\| \| bcms-bertic-frenk-hate \| 0.8313 \| 0.8219 \| \| EMBEDDIA/crosloengual-bert \| 0.8054 \| 0.796 \| \| xlm-roberta-base \| 0.7175 \| 0.7049 \| \| fasttext \| 0.771 \| 0.754 \| From recorded accuracies and macro F1 scores p-values were also calculated: Comparison with `crosloengual-bert`: \| test \| accuracy p-value \| macro F1 p-value \| \|----------------\|------------------\|------------------\| \| Wilcoxon \| 0.00781 \| 0.00781 \| \| Mann Whithney \| 0.00108 \| 0.00108 \| \| Student t-test \| 2.43e-10 \| 1.27e-10 \| Comparison with `xlm-roberta-base`: \| test \| accuracy p-value \| macro F1 p-value \| \|----------------\|------------------\|------------------\| \| Wilcoxon \| 0.00781 \| 0.00781 \| \| Mann Whithney \| 0.00107 \| 0.00108 \| \| Student t-test \| 4.83e-11 \| 5.61e-11 \| ## Use examples ```python from simpletransformers.classification import ClassificationModel model = ClassificationModel( "bert", "5roop/bcms-bertic-frenk-hate", use_cuda=True, ) predictions, logit_output = model.predict(['Ne odbacujem da će RH primiti još migranata iz Afganistana, no neće biti novog vala', "Potpredsjednik Vlade i ministar branitelja Tomo Medved komentirao je Vladine planove za zakonsku zabranu pozdrava 'za dom spremni' "]) predictions ### Output: ### array([0, 0]) ``` ## Citation If you use the model, please cite the following paper on which the original model is based: ``` @inproceedings{ljubesic-lauc-2021-bertic, title = "{BERT}i{\'c} - The Transformer Language Model for {B}osnian, {C}roatian, {M}ontenegrin and {S}erbian", author = "Ljube{\v{s}}i{\'c}, Nikola and Lauc, Davor", booktitle = "Proceedings of the 8th Workshop on Balto-Slavic Natural Language Processing", month = apr, year = "2021", address = "Kiyv, Ukraine", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/2021.bsnlp-1.5", pages = "37--42", } ``` and the dataset used for fine-tuning: ``` @misc{ljubešić2019frenk, title={The FRENK Datasets of Socially Unacceptable Discourse in Slovene and English}, author={Nikola Ljubešić and Darja Fišer and Tomaž Erjavec}, year={2019}, eprint={1906.02045}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/1906.02045} } ```
crystina-z/monoELECTRA_LCE_nneg31	47923dbe2fb0dedea1c1572940b1289806838a92	2022-02-11T18:02:52.000Z	[ "pytorch", "tf", "electra", "text-classification", "transformers" ]	text-classification	false	crystina-z	null	crystina-z/monoELECTRA_LCE_nneg31	15	null	transformers	9,506	Entry not found
dbmdz/electra-base-german-europeana-cased-generator	195c6427c576e68a7c2a97de2e20421fc506c58c	2020-07-26T00:53:55.000Z	[ "pytorch", "tf", "electra", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	dbmdz	null	dbmdz/electra-base-german-europeana-cased-generator	15	null	transformers	9,507	Entry not found
fabriceyhc/bert-base-uncased-yahoo_answers_topics	968176fc24a2eb73cac26ab4312d8b22da98486a	2021-09-21T00:54:22.000Z	[ "pytorch", "bert", "text-classification", "dataset:yahoo_answers_topics", "transformers", "generated_from_trainer", "sibyl", "license:apache-2.0", "model-index" ]	text-classification	false	fabriceyhc	null	fabriceyhc/bert-base-uncased-yahoo_answers_topics	15	1	transformers	9,508	--- license: apache-2.0 tags: - generated_from_trainer - sibyl datasets: - yahoo_answers_topics metrics: - accuracy model-index: - name: bert-base-uncased-yahoo_answers_topics results: - task: name: Text Classification type: text-classification dataset: name: yahoo_answers_topics type: yahoo_answers_topics args: yahoo_answers_topics metrics: - name: Accuracy type: accuracy value: 0.7499166666666667 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # bert-base-uncased-yahoo_answers_topics This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the yahoo_answers_topics dataset. It achieves the following results on the evaluation set: - Loss: 0.8092 - Accuracy: 0.7499 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 86625 - training_steps: 866250 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:-----:\|:---------------:\|:--------:\| \| 2.162 \| 0.01 \| 2000 \| 1.7444 \| 0.5681 \| \| 1.3126 \| 0.02 \| 4000 \| 1.0081 \| 0.7054 \| \| 0.9592 \| 0.03 \| 6000 \| 0.9021 \| 0.7234 \| \| 0.8903 \| 0.05 \| 8000 \| 0.8827 \| 0.7276 \| \| 0.8685 \| 0.06 \| 10000 \| 0.8540 \| 0.7341 \| \| 0.8422 \| 0.07 \| 12000 \| 0.8547 \| 0.7365 \| \| 0.8535 \| 0.08 \| 14000 \| 0.8264 \| 0.7372 \| \| 0.8178 \| 0.09 \| 16000 \| 0.8331 \| 0.7389 \| \| 0.8325 \| 0.1 \| 18000 \| 0.8242 \| 0.7411 \| \| 0.8181 \| 0.12 \| 20000 \| 0.8356 \| 0.7437 \| \| 0.8171 \| 0.13 \| 22000 \| 0.8090 \| 0.7451 \| \| 0.8092 \| 0.14 \| 24000 \| 0.8469 \| 0.7392 \| \| 0.8057 \| 0.15 \| 26000 \| 0.8185 \| 0.7478 \| \| 0.8085 \| 0.16 \| 28000 \| 0.8090 \| 0.7467 \| \| 0.8229 \| 0.17 \| 30000 \| 0.8225 \| 0.7417 \| \| 0.8151 \| 0.18 \| 32000 \| 0.8262 \| 0.7419 \| \| 0.81 \| 0.2 \| 34000 \| 0.8149 \| 0.7383 \| \| 0.8073 \| 0.21 \| 36000 \| 0.8225 \| 0.7441 \| \| 0.816 \| 0.22 \| 38000 \| 0.8037 \| 0.744 \| \| 0.8217 \| 0.23 \| 40000 \| 0.8409 \| 0.743 \| \| 0.82 \| 0.24 \| 42000 \| 0.8286 \| 0.7385 \| \| 0.8101 \| 0.25 \| 44000 \| 0.8282 \| 0.7413 \| \| 0.8254 \| 0.27 \| 46000 \| 0.8170 \| 0.7414 \| ### Framework versions - Transformers 4.10.2 - Pytorch 1.7.1 - Datasets 1.6.1 - Tokenizers 0.10.3
fbaigt/procbert	20814e122765866e213447ebe2618d2f0b90cbf1	2021-11-08T15:08:01.000Z	[ "pytorch", "bert", "feature-extraction", "en", "dataset:pubmed", "dataset:chemical patent", "dataset:cooking recipe", "arxiv:2109.04711", "transformers" ]	feature-extraction	false	fbaigt	null	fbaigt/procbert	15	1	transformers	9,509	--- language: - en datasets: - pubmed - chemical patent - cooking recipe --- ## ProcBERT ProcBERT is a pre-trained language model specifically for procedural text. It was pre-trained on a large-scale procedural corpus (PubMed articles/chemical patents/cooking recipes) containing over 12B tokens and shows great performance on downstream tasks. More details can be found in the following [paper](https://arxiv.org/abs/2109.04711): ``` @inproceedings{bai-etal-2021-pre, title = "Pre-train or Annotate? Domain Adaptation with a Constrained Budget", author = "Bai, Fan and Ritter, Alan and Xu, Wei", booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing", month = nov, year = "2021", address = "Online and Punta Cana, Dominican Republic", publisher = "Association for Computational Linguistics", } ``` ## Usage ``` from transformers import * tokenizer = AutoTokenizer.from_pretrained("fbaigt/procbert") model = AutoModelForTokenClassification.from_pretrained("fbaigt/procbert") ``` More usage details can be found [here](https://github.com/bflashcp3f/ProcBERT).
federicopascual/finetune-sentiment-analysis-model-3000-samples	595ae6575f96bc971fd033b3560a98e80ede9517	2021-12-30T19:29:48.000Z	[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:imdb", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]	text-classification	false	federicopascual	null	federicopascual/finetune-sentiment-analysis-model-3000-samples	15	null	transformers	9,510	--- license: apache-2.0 tags: - generated_from_trainer datasets: - imdb metrics: - accuracy - f1 model-index: - name: finetune-sentiment-analysis-model-3000-samples results: - task: name: Text Classification type: text-classification dataset: name: imdb type: imdb args: plain_text metrics: - name: Accuracy type: accuracy value: 0.8866666666666667 - name: F1 type: f1 value: 0.8944099378881988 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # finetune-sentiment-analysis-model-3000-samples This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.4558 - Accuracy: 0.8867 - F1: 0.8944 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results ### Framework versions - Transformers 4.15.0 - Pytorch 1.10.0+cu111 - Datasets 1.17.0 - Tokenizers 0.10.3
figurative-nlp/t5-figurative-paraphrase	4c382b695540ace5fa8ce647e3fcd67a372a93f8	2022-02-17T12:21:53.000Z	[ "pytorch", "t5", "text2text-generation", "transformers", "autotrain_compatible" ]	text2text-generation	false	figurative-nlp	null	figurative-nlp/t5-figurative-paraphrase	15	2	transformers	9,511	This model can convert the figurative/metaphorical expression to the literal expression. Below is the usage of our model: from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("figurative-nlp/t5-figurative-paraphrase") model = AutoModelForSeq2SeqLM.from_pretrained("figurative-nlp/t5-figurative-paraphrase") input_ids = tokenizer( "paraphrase the sentence : i will talk this story to you from A to Z", return_tensors="pt" ).input_ids # Batch size 1 outputs = model.generate(input_ids,num_beams = 5) result = tokenizer.decode(outputs[0], skip_special_tokens=True) #result : i will talk this story to you from beginning to end.. For example: Input: He is always bang on when he makes a speech. Output: He is always presice when he makes a speech. Input: He always buy what he said. Output: He always agree with what he said. Input: Your team will be done like dinner if they play against the all-star team. Output: Your team will be defeated if they play against the all-star team. (the one is not particularly accurate) Note: the figurative language here includes metaphor, idiom and simile. We don't guarantee that the results generated results are satisfactory to you. We are trying to improve the effect of the model.
fnlp/elasticbert-large	0a5689cea93ed0bf88c87bcd623e0de0f98516e2	2021-10-28T11:05:49.000Z	[ "pytorch", "elasticbert", "fill-mask", "arxiv:2110.07038", "transformers", "autotrain_compatible" ]	fill-mask	false	fnlp	null	fnlp/elasticbert-large	15	2	transformers	9,512	# ElasticBERT-LARGE ## Model description This is an implementation of the `large` version of ElasticBERT. [Towards Efficient NLP: A Standard Evaluation and A Strong Baseline](https://arxiv.org/pdf/2110.07038.pdf) Xiangyang Liu, Tianxiang Sun, Junliang He, Lingling Wu, Xinyu Zhang, Hao Jiang, Zhao Cao, Xuanjing Huang, Xipeng Qiu ## Code link [fastnlp/elasticbert](https://github.com/fastnlp/ElasticBERT) ## Usage ```python >>> from transformers import BertTokenizer as ElasticBertTokenizer >>> from models.configuration_elasticbert import ElasticBertConfig >>> from models.modeling_elasticbert import ElasticBertForSequenceClassification >>> num_output_layers = 1 >>> config = ElasticBertConfig.from_pretrained('fnlp/elasticbert-large', num_output_layers=num_output_layers ) >>> tokenizer = ElasticBertTokenizer.from_pretrained('fnlp/elasticbert-large') >>> model = ElasticBertForSequenceClassification.from_pretrained('fnlp/elasticbert-large', config=config) >>> input_ids = tokenizer.encode('The actors are fantastic .', return_tensors='pt') >>> outputs = model(input_ids) ``` ## Citation ```bibtex @article{liu2021elasticbert, author = {Xiangyang Liu and Tianxiang Sun and Junliang He and Lingling Wu and Xinyu Zhang and Hao Jiang and Zhao Cao and Xuanjing Huang and Xipeng Qiu}, title = {Towards Efficient {NLP:} {A} Standard Evaluation and {A} Strong Baseline}, journal = {CoRR}, volume = {abs/2110.07038}, year = {2021}, url = {https://arxiv.org/abs/2110.07038}, eprinttype = {arXiv}, eprint = {2110.07038}, timestamp = {Fri, 22 Oct 2021 13:33:09 +0200}, biburl = {https://dblp.org/rec/journals/corr/abs-2110-07038.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ```
geekfeed/gpt2_ja	6c297f7e58fc6e7c75d654941380620cd3710660	2021-05-21T16:11:52.000Z	[ "pytorch", "jax", "gpt2", "feature-extraction", "transformers" ]	feature-extraction	false	geekfeed	null	geekfeed/gpt2_ja	15	null	transformers	9,513	hello
ghadeermobasher/BC2GM-Gene-Modified_scibert_scivocab_cased	cf11c70aceb0994ec32ebedfa0a2e878043b12f9	2022-01-23T19:55:04.000Z	[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC2GM-Gene-Modified_scibert_scivocab_cased	15	null	transformers	9,514	Entry not found
ghadeermobasher/BC2GM-Gene_ImbalancedPubMedBERT	2574cb8b0ee0fe4af0e9b272c51f855bfa3c1b01	2022-01-22T01:44:25.000Z	[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC2GM-Gene_ImbalancedPubMedBERT	15	null	transformers	9,515	Entry not found
ghadeermobasher/BC4-Original-biobert-v1.1	ab10a139dbc7b0c0c9bc6a4558a206e5d73fb3e5	2022-02-24T14:45:29.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-Original-biobert-v1.1	15	null	transformers	9,516	Entry not found
ghadeermobasher/BC4-Original-bluebert_pubmed_uncased_L-12_H-768_A-12	cbceb41840ab0b1b4d9cd8d4cc7a19ac477fae8d	2022-02-24T14:22:37.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-Original-bluebert_pubmed_uncased_L-12_H-768_A-12	15	null	transformers	9,517	Entry not found
ghadeermobasher/BC4-Original-scibert_scivocab_uncased	1bb6e36961e4a831a996c806637ac891282bf3e9	2022-02-24T14:28:32.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-Original-scibert_scivocab_uncased	15	null	transformers	9,518	Entry not found
ghadeermobasher/BC5CDR-Chemical-imbalanced-bluebert_pubmed_uncased_L-12_H-768_A-12_latest	17f9443fd2d369763bf7c64b4eb9206803f0df27	2022-02-21T23:07:06.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC5CDR-Chemical-imbalanced-bluebert_pubmed_uncased_L-12_H-768_A-12_latest	15	null	transformers	9,519	Entry not found
glasses/resnet18	a15b2ef76c4e01cc6b3f4518f56c1c98722d6793	2021-11-30T20:06:28.000Z	[ "pytorch", "dataset:imagenet", "arxiv:1512.03385", "arxiv:1812.01187", "transformers", "image-classification", "license:apache-2.0" ]	image-classification	false	glasses	null	glasses/resnet18	15	null	transformers	9,520	--- license: apache-2.0 tags: - image-classification datasets: - imagenet --- # resnet18 Implementation of ResNet proposed in [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) ``` python ResNet.resnet18() ResNet.resnet26() ResNet.resnet34() ResNet.resnet50() ResNet.resnet101() ResNet.resnet152() ResNet.resnet200() Variants (d) proposed in `Bag of Tricks for Image Classification with Convolutional Neural Networks <https://arxiv.org/pdf/1812.01187.pdf`_ ResNet.resnet26d() ResNet.resnet34d() ResNet.resnet50d() # You can construct your own one by chaning `stem` and `block` resnet101d = ResNet.resnet101(stem=ResNetStemC, block=partial(ResNetBottleneckBlock, shortcut=ResNetShorcutD)) ``` Examples: ``` python # change activation ResNet.resnet18(activation = nn.SELU) # change number of classes (default is 1000 ) ResNet.resnet18(n_classes=100) # pass a different block ResNet.resnet18(block=SENetBasicBlock) # change the steam model = ResNet.resnet18(stem=ResNetStemC) change shortcut model = ResNet.resnet18(block=partial(ResNetBasicBlock, shortcut=ResNetShorcutD)) # store each feature x = torch.rand((1, 3, 224, 224)) # get features model = ResNet.resnet18() # first call .features, this will activate the forward hooks and tells the model you'll like to get the features model.encoder.features model(torch.randn((1,3,224,224))) # get the features from the encoder features = model.encoder.features print([x.shape for x in features]) #[torch.Size([1, 64, 112, 112]), torch.Size([1, 64, 56, 56]), torch.Size([1, 128, 28, 28]), torch.Size([1, 256, 14, 14])] ```
gpssohi/distilbart-qgen-3-3	66e9d4f41a1c4bf55bdab0a9eb904476542d5d06	2022-01-12T08:29:26.000Z	[ "pytorch", "bart", "text2text-generation", "en", "dataset:squad", "transformers", "question-generation", "summarization", "license:apache-2.0", "autotrain_compatible" ]	summarization	false	gpssohi	null	gpssohi/distilbart-qgen-3-3	15	2	transformers	9,521	--- language: en tags: - question-generation - summarization license: apache-2.0 datasets: - squad --- # Introduction This model checkpoint is obtained by first fine-tuning the sshleifer/distilbart-cnn-6-6 summarization checkpoint on the SQuAD dataset. After this, the 6-6 fine-tuned model is distilled down to a 3-3 model which gives us the final checkpoint. [GitHub Link for training scripts.](https://github.com/darth-c0d3r/bart-question-generation) # Usage The input format is as follows: `[answer] <s> [passage]`. The model will predict the question that corresponds to the answer from the passage. # Plot ![Distillation Run](distill_run_21.png) # Dataset The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Hence, the input to the model will be a passage context and an answer, and the output / target will be the question for the given answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chat-bots to lead a conversation. The final dataset is created by taking the union of the following Question Answering Datasets. The dataset must have the following three columns: context, question, answer. ## [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowd-workers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable. We use the SQuAD 1.1 variant which does not have unanswerable questions. So, every question will have a corresponding answer and vice-versa. ### Preprocessing The first step is to remove questions which don't have answers. After that, we split the train set into Train and Eval sets and treat the dev set as the test set. ### Stats Original Dataset \| Split \| Num Docs \| Num Contexts \| Ques w/ Ans \| Ques w/o Ans \| Num Unique Ans \| \| ----- \| -------- \| ------------ \| ----------- \| ------------ \| -------------- \| \| Train \| 442 \| 19035 \| 86821 \| 43498 \| 86821 \| \| Dev \| 35 \| 1204 \| 5928 \| 5945 \| 10279 \| After Preprocessing \| Split \| Num Rows \| Context \| Answer \| Question \| \| ----- \| -------- \| ---------- \| ------ \| -------- \| \| Train \| 80995 \| 653,120,20 \| 43,3,1 \| 40,10,1 \| \| Eval \| 5826 \| 445,123,67 \| 28,3,1 \| 29,10,3 \| \| Test \| 10297 \| 629,129,25 \| 29,4,1 \| 31,10,3 \| The numbers in the columns indicate max, avg, min number of words.
gpssohi/distilbart-qgen-6-6	18d85ee5d5482d7af7b5f719048f4bc641c3a5ff	2022-01-12T08:29:13.000Z	[ "pytorch", "bart", "text2text-generation", "en", "dataset:squad", "transformers", "summarization", "question-generation", "license:apache-2.0", "autotrain_compatible" ]	summarization	false	gpssohi	null	gpssohi/distilbart-qgen-6-6	15	1	transformers	9,522	--- language: en tags: - summarization - question-generation license: apache-2.0 datasets: - squad --- # Introduction This model checkpoint is obtained by fine-tuning the `sshleifer/distilbart-cnn-6-6` summarization checkpoint on the SQuAD dataset. [GitHub Link for training scripts.](https://github.com/darth-c0d3r/bart-question-generation) # Usage The input format is as follows: `[answer] <s> [passage]`. The model will predict the question that corresponds to the answer from the passage. # Plot ![Training Run](train_run_6.png) # Dataset The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Hence, the input to the model will be a passage context and an answer, and the output / target will be the question for the given answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chat-bots to lead a conversation. The final dataset is created by taking the union of the following Question Answering Datasets. The dataset must have the following three columns: context, question, answer. ## [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowd-workers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable. We use the SQuAD 1.1 variant which does not have unanswerable questions. So, every question will have a corresponding answer and vice-versa. ### Preprocessing The first step is to remove questions that don't have answers. After that, we split the train set into Train and Eval sets and treat the dev set as the test set. ### Stats Original Dataset \| Split \| Num Docs \| Num Contexts \| Ques w/ Ans \| Ques w/o Ans \| Num Unique Ans \| \| ----- \| -------- \| ------------ \| ----------- \| ------------ \| -------------- \| \| Train \| 442 \| 19035 \| 86821 \| 43498 \| 86821 \| \| Dev \| 35 \| 1204 \| 5928 \| 5945 \| 10279 \| After Preprocessing \| Split \| Num Rows \| Context \| Answer \| Question \| \| ----- \| -------- \| ---------- \| ------ \| -------- \| \| Train \| 80995 \| 653,120,20 \| 43,3,1 \| 40,10,1 \| \| Eval \| 5826 \| 445,123,67 \| 28,3,1 \| 29,10,3 \| \| Test \| 10297 \| 629,129,25 \| 29,4,1 \| 31,10,3 \| The numbers in the columns indicate max, avg, min number of words.
gwkim22/domain_base2_disc	253c926654ffcae55fe363bca751474d03b90ec7	2021-07-19T01:56:14.000Z	[ "pytorch", "electra", "pretraining", "transformers" ]	null	false	gwkim22	null	gwkim22/domain_base2_disc	15	1	transformers	9,523	"domain_base2_disc_0719"
hamzaMM/questionClassifier	d1d326c6965fb4f91070df4356562dade3b37364	2021-12-02T20:08:26.000Z	[ "pytorch", "bert", "text-classification", "transformers" ]	text-classification	false	hamzaMM	null	hamzaMM/questionClassifier	15	2	transformers	9,524	Entry not found
hrdipto/wav2vec2-xls-r-300m-bangla-command-generated-data-finetune	6361d8dfe95c21a0fe20389a195e9de9aab1de02	2022-02-14T08:58:20.000Z	[ "pytorch", "tensorboard", "wav2vec2", "automatic-speech-recognition", "transformers", "generated_from_trainer", "model-index" ]	automatic-speech-recognition	false	hrdipto	null	hrdipto/wav2vec2-xls-r-300m-bangla-command-generated-data-finetune	15	null	transformers	9,525	--- tags: - generated_from_trainer model-index: - name: wav2vec2-xls-r-300m-bangla-command-generated-data-finetune results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # wav2vec2-xls-r-300m-bangla-command-generated-data-finetune This model is a fine-tuned version of [hrdipto/wav2vec2-xls-r-300m-bangla-command-data](https://huggingface.co/hrdipto/wav2vec2-xls-r-300m-bangla-command-data) on the None dataset. It achieves the following results on the evaluation set: - eval_loss: 0.0099 - eval_wer: 0.0208 - eval_runtime: 2.5526 - eval_samples_per_second: 75.217 - eval_steps_per_second: 9.402 - epoch: 71.43 - step: 2000 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 100 - mixed_precision_training: Native AMP ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.0
huggingartists/kanye-west	ef0b90df2f597af17783d7ae3477a01de520f35c	2022-05-05T00:27:21.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "en", "dataset:huggingartists/kanye-west", "transformers", "huggingartists", "lyrics", "lm-head", "causal-lm" ]	text-generation	false	huggingartists	null	huggingartists/kanye-west	15	null	transformers	9,526	--- language: en datasets: - huggingartists/kanye-west tags: - huggingartists - lyrics - lm-head - causal-lm widget: - text: "I am" --- <div class="inline-flex flex-col" style="line-height: 1.5;"> <div class="flex"> <div style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://images.genius.com/54520386ec39aca6408c7e2c156ae10a.399x399x1.png')"> </div> </div> <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div> <div style="text-align: center; font-size: 16px; font-weight: 800">Kanye West</div> <a href="https://genius.com/artists/kanye-west"> <div style="text-align: center; font-size: 14px;">@kanye-west</div> </a> </div> I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists). Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)! ## How does it work? To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist). ## Training data The model was trained on lyrics from Kanye West. Dataset is available [here](https://huggingface.co/datasets/huggingartists/kanye-west). And can be used with: ```python from datasets import load_dataset dataset = load_dataset("huggingartists/kanye-west") ``` [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/hl7afoso/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Kanye West's lyrics. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/28dw8m5v) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/28dw8m5v/artifacts) is logged and versioned. ## How to use You can use this model directly with a pipeline for text generation: ```python from transformers import pipeline generator = pipeline('text-generation', model='huggingartists/kanye-west') generator("I am", num_return_sequences=5) ``` Or with Transformers library: ```python from transformers import AutoTokenizer, AutoModelWithLMHead tokenizer = AutoTokenizer.from_pretrained("huggingartists/kanye-west") model = AutoModelWithLMHead.from_pretrained("huggingartists/kanye-west") ``` ## Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About Built by Aleksey Korshuk [![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk) [![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk) [![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky) For more details, visit the project repository. [![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)
huggingtweets/amazon	7bc08510372ae3b213814f77f110ae1f3138dd3a	2021-05-21T18:33:06.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]	text-generation	false	huggingtweets	null	huggingtweets/amazon	15	null	transformers	9,527	--- language: en thumbnail: https://www.huggingtweets.com/amazon/1609713999453/predictions.png tags: - huggingtweets widget: - text: "My dream is" --- <link rel="stylesheet" href="https://unpkg.com/@tailwindcss/[email protected]/dist/typography.min.css"> <style> @media (prefers-color-scheme: dark) { .prose { color: #E2E8F0 !important; } .prose h2, .prose h3, .prose a, .prose thead { color: #F7FAFC !important; } } </style> <section class='prose'> <div> <div style="width: 132px; height:132px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/949070360103698432/kXSiPeTk_400x400.jpg')"> </div> <div style="margin-top: 8px; font-size: 19px; font-weight: 800">Amazon 🤖 AI Bot </div> <div style="font-size: 15px; color: #657786">@amazon bot</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://app.wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-model-to-generate-tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on [@amazon's tweets](https://twitter.com/amazon). <table style='border-width:0'> <thead style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #CBD5E0'> <th style='border-width:0'>Data</th> <th style='border-width:0'>Quantity</th> </tr> </thead> <tbody style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Tweets downloaded</td> <td style='border-width:0'>3242</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Retweets</td> <td style='border-width:0'>40</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Short tweets</td> <td style='border-width:0'>60</td> </tr> <tr style='border-width:0'> <td style='border-width:0'>Tweets kept</td> <td style='border-width:0'>3142</td> </tr> </tbody> </table> [Explore the data](https://wandb.ai/wandb/huggingtweets/runs/1fd78mc2/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @amazon's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/76pxw0n0) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/76pxw0n0/artifacts) is logged and versioned. ## Intended uses & limitations ### How to use You can use this model directly with a pipeline for text generation: <pre><code><span style="color:#03A9F4">from</span> transformers <span style="color:#03A9F4">import</span> pipeline generator = pipeline(<span style="color:#FF9800">'text-generation'</span>, model=<span style="color:#FF9800">'huggingtweets/amazon'</span>) generator(<span style="color:#FF9800">"My dream is"</span>, num_return_sequences=<span style="color:#8BC34A">5</span>)</code></pre> ### Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About Built by Boris Dayma </section> [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) <section class='prose'> For more details, visit the project repository. </section> [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets)
huggingtweets/deepleffen	9f6a7a1f1bb73ef33f7cead6ee0b72ba37411d4f	2022-06-03T17:34:54.000Z	[ "pytorch", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]	text-generation	false	huggingtweets	null	huggingtweets/deepleffen	15	null	transformers	9,528	--- language: en thumbnail: http://www.huggingtweets.com/deepleffen/1654277690184/predictions.png tags: - huggingtweets widget: - text: "My dream is" --- <div class="inline-flex flex-col" style="line-height: 1.5;"> <div class="flex"> <div style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1241879678455078914/e2EdZIrr_400x400.jpg')"> </div> <div style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')"> </div> <div style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')"> </div> </div> <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI BOT 🤖</div> <div style="text-align: center; font-size: 16px; font-weight: 800">Deep Leffen Bot</div> <div style="text-align: center; font-size: 14px;">@deepleffen</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on tweets from Deep Leffen Bot. \| Data \| Deep Leffen Bot \| \| --- \| --- \| \| Tweets downloaded \| 589 \| \| Retweets \| 14 \| \| Short tweets \| 27 \| \| Tweets kept \| 548 \| [Explore the data](https://wandb.ai/wandb/huggingtweets/runs/1p32tock/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @deepleffen's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/imjjixah) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/imjjixah/artifacts) is logged and versioned. ## How to use You can use this model directly with a pipeline for text generation: ```python from transformers import pipeline generator = pipeline('text-generation', model='huggingtweets/deepleffen') generator("My dream is", num_return_sequences=5) ``` ## Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About Built by Boris Dayma [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) For more details, visit the project repository. [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets)
huggingtweets/nvidia	cceee848844258789df63f96b72b676fade2a4aa	2021-05-22T17:00:36.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]	text-generation	false	huggingtweets	null	huggingtweets/nvidia	15	null	transformers	9,529	--- language: en thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true tags: - huggingtweets widget: - text: "My dream is" --- <link rel="stylesheet" href="https://unpkg.com/@tailwindcss/[email protected]/dist/typography.min.css"> <style> @media (prefers-color-scheme: dark) { .prose { color: #E2E8F0 !important; } .prose h2, .prose h3, .prose a, .prose thead { color: #F7FAFC !important; } } </style> <section class='prose'> <div> <div style="width: 132px; height:132px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1145524454170062848/U4lxVYEw_400x400.png')"> </div> <div style="margin-top: 8px; font-size: 19px; font-weight: 800">NVIDIA 🤖 AI Bot </div> <div style="font-size: 15px; color: #657786">@nvidia bot</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://app.wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-model-to-generate-tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on [@nvidia's tweets](https://twitter.com/nvidia). <table style='border-width:0'> <thead style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #CBD5E0'> <th style='border-width:0'>Data</th> <th style='border-width:0'>Quantity</th> </tr> </thead> <tbody style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Tweets downloaded</td> <td style='border-width:0'>3222</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Retweets</td> <td style='border-width:0'>1876</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Short tweets</td> <td style='border-width:0'>18</td> </tr> <tr style='border-width:0'> <td style='border-width:0'>Tweets kept</td> <td style='border-width:0'>1328</td> </tr> </tbody> </table> [Explore the data](https://app.wandb.ai/wandb/huggingtweets/runs/35e4bboc/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @nvidia's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://app.wandb.ai/wandb/huggingtweets/runs/2t7z7a45) for full transparency and reproducibility. At the end of training, [the final model](https://app.wandb.ai/wandb/huggingtweets/runs/2t7z7a45/artifacts) is logged and versioned. ## Intended uses & limitations ### How to use You can use this model directly with a pipeline for text generation: <pre><code><span style="color:#03A9F4">from</span> transformers <span style="color:#03A9F4">import</span> pipeline generator = pipeline(<span style="color:#FF9800">'text-generation'</span>, model=<span style="color:#FF9800">'huggingtweets/nvidia'</span>) generator(<span style="color:#FF9800">"My dream is"</span>, num_return_sequences=<span style="color:#8BC34A">5</span>)</code></pre> ### Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About Built by Boris Dayma </section> [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) <section class='prose'> For more details, visit the project repository. </section> [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets) <!--- random size file -->
imzachjohnson/autonlp-spinner-check-16492731	3e96dbddaf4b1b2c760fbf196391ac77ecfc7890	2021-10-11T00:02:11.000Z	[ "pytorch", "bert", "text-classification", "en", "dataset:imzachjohnson/autonlp-data-spinner-check", "transformers", "autonlp" ]	text-classification	false	imzachjohnson	null	imzachjohnson/autonlp-spinner-check-16492731	15	null	transformers	9,530	--- tags: autonlp language: en widget: - text: "I love AutoNLP 🤗" datasets: - imzachjohnson/autonlp-data-spinner-check --- # Model Trained Using AutoNLP - Problem type: Binary Classification - Model ID: 16492731 ## Validation Metrics - Loss: 0.21610039472579956 - Accuracy: 0.9155366722657816 - Precision: 0.9530714194995978 - Recall: 0.944871149164778 - AUC: 0.9553238723676906 - F1: 0.9489535692456846 ## Usage You can use cURL to access this model: ``` $ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoNLP"}' https://api-inference.huggingface.co/models/imzachjohnson/autonlp-spinner-check-16492731 ``` Or Python API: ``` from transformers import AutoModelForSequenceClassification, AutoTokenizer model = AutoModelForSequenceClassification.from_pretrained("imzachjohnson/autonlp-spinner-check-16492731", use_auth_token=True) tokenizer = AutoTokenizer.from_pretrained("imzachjohnson/autonlp-spinner-check-16492731", use_auth_token=True) inputs = tokenizer("I love AutoNLP", return_tensors="pt") outputs = model(**inputs) ```
jaimin/plagiarism_checker	63fdab21cd1ee8fa220533eeb00c77238156728f	2021-08-20T05:44:24.000Z	[ "pytorch", "longformer", "text-classification", "transformers" ]	text-classification	false	jaimin	null	jaimin/plagiarism_checker	15	null	transformers	9,531	"hello"
kanishka/GlossBERT	0cc3b83af5496e27ebcc95ef0cf37ea0a9281a7a	2021-09-22T08:54:41.000Z	[ "pytorch", "bert", "en", "dataset:SemCor3.0", "arxiv:1908.07245", "transformers", "glossbert", "license:mit" ]	null	false	kanishka	null	kanishka/GlossBERT	15	null	transformers	9,532	--- language: en tags: - glossbert license: mit datasets: - SemCor3.0 --- ## GlossBERT A BERT-based model fine-tuned on SemCor 3.0 to perform word-sense-disambiguation by leveraging gloss information. This model is the research output of the paper titled: '[GlossBERT: BERT for Word Sense Disambiguation with Gloss Knowledge](https://arxiv.org/pdf/1908.07245.pdf)' Disclaimer: This model was built and trained by a group of researchers different than the repository's author. The original model code can be found on github: https://github.com/HSLCY/GlossBERT ## Usage The following code loads GlossBERT: ```py from transformers import AutoTokenizer, BertForSequenceClassification tokenizer = AutoTokenizer.from_pretrained('kanishka/GlossBERT') model = BertForSequenceClassification.from_pretrained('kanishka/GlossBERT') ``` ## Citation If you use this model in any of your projects, please cite the original authors using the following bibtex: ``` @inproceedings{huang-etal-2019-glossbert, title = "{G}loss{BERT}: {BERT} for Word Sense Disambiguation with Gloss Knowledge", author = "Huang, Luyao and Sun, Chi and Qiu, Xipeng and Huang, Xuanjing", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)", month = nov, year = "2019", address = "Hong Kong, China", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/D19-1355", doi = "10.18653/v1/D19-1355", pages = "3507--3512" } ```
keshan/sinhala-roberta-oscar	655873a1b237c7e09c424d0a55bb9fb05456248e	2021-07-14T06:28:47.000Z	[ "pytorch", "jax", "tensorboard", "roberta", "fill-mask", "si", "dataset:oscar", "arxiv:1907.11692", "transformers", "oscar", "Sinhala", "autotrain_compatible" ]	fill-mask	false	keshan	null	keshan/sinhala-roberta-oscar	15	null	transformers	9,533	--- language: si tags: - oscar - Sinhala - roberta - fill-mask widget: - text: "මම සිංහල භාෂාව <mask>" datasets: - oscar --- ### Overview This is a slightly smaller model trained on [OSCAR](https://oscar-corpus.com/) Sinhala dedup dataset. As Sinhala is one of those low resource languages, there are only a handful of models been trained. So, this would be a great place to start training for more downstream tasks. ## Model Specification The model chosen for training is [Roberta](https://arxiv.org/abs/1907.11692) with the following specifications: 1. vocab_size=50265 2. max_position_embeddings=514 3. num_attention_heads=12 4. num_hidden_layers=12 5. type_vocab_size=1 ## How to Use You can use this model directly with a pipeline for masked language modeling: ```py from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline model = AutoModelWithLMHead.from_pretrained("keshan/sinhala-roberta-oscar") tokenizer = AutoTokenizer.from_pretrained("keshan/sinhala-roberta-oscar") fill_mask = pipeline('fill-mask', model=model, tokenizer=tokenizer) fill_mask("මම ගෙදර <mask>.") ```
kurianbenoy/distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb	465c2a001ccd95ff2faee805ffe909ef79fdf366	2022-02-21T11:55:41.000Z	[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:imdb", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]	text-classification	false	kurianbenoy	null	kurianbenoy/distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb	15	null	transformers	9,534	--- license: apache-2.0 tags: - generated_from_trainer datasets: - imdb metrics: - accuracy model-index: - name: distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb results: - task: name: Text Classification type: text-classification dataset: name: imdb type: imdb args: plain_text metrics: - name: Accuracy type: accuracy value: 0.93032 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb This model is a fine-tuned version of [distilbert-base-uncased-finetuned-sst-2-english](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.2165 - Accuracy: 0.9303 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\| \| 0.2749 \| 1.0 \| 3125 \| 0.2165 \| 0.9303 \| ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.0
leonardvorbeck/wav2vec2-large-robust-SB300	e630e8662f2d52c6be25ccd3e95ba417f7c8b21b	2021-08-26T12:22:18.000Z	[ "pytorch", "wav2vec2", "automatic-speech-recognition", "en", "dataset:libri_light", "dataset:common_voice", "dataset:switchboard", "dataset:fisher", "arxiv:2104.01027", "transformers", "speech", "CTC", "Attention", "license:apache-2.0" ]	automatic-speech-recognition	false	leonardvorbeck	null	leonardvorbeck/wav2vec2-large-robust-SB300	15	1	transformers	9,535	--- language: en datasets: - libri_light - common_voice - switchboard - fisher tags: - speech - automatic-speech-recognition - CTC - Attention - wav2vec2 license: apache-2.0 --- # Wav2Vec2-Large-Robust - Finetuned on Switchboard (300 hours) ## Note : Model has not been initialized. If you want to use it without further finetuning, do a forward pass first to recalculate the normalized weights of the positional convolutional layer : ```ipython with torch.no_grad(): model(torch.randn((1,300_000))) ``` [Facebook's Wav2Vec2](https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/) The base model pretrained on 16kHz sampled speech audio. Speech datasets from multiple domains were used to pretrain the model: - [Libri-Light](https://github.com/facebookresearch/libri-light): open-source audio books from the LibriVox project; clean, read-out audio data - [CommonVoice](https://huggingface.co/datasets/common_voice): crowd-source collected audio data; read-out text snippets - [Switchboard](https://catalog.ldc.upenn.edu/LDC97S62): telephone speech corpus; noisy telephone data - [Fisher](https://catalog.ldc.upenn.edu/LDC2004T19): conversational telephone speech; noisy telephone data When using the model make sure that your speech input is also sampled at 16Khz. Check out [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information. [Paper Robust Wav2Vec2](https://arxiv.org/abs/2104.01027) Authors: Wei-Ning Hsu, Anuroop Sriram, Alexei Baevski, Tatiana Likhomanenko, Qiantong Xu, Vineel Pratap, Jacob Kahn, Ann Lee, Ronan Collobert, Gabriel Synnaeve, Michael Auli Abstract Self-supervised learning of speech representations has been a very active research area but most work is focused on a single domain such as read audio books for which there exist large quantities of labeled and unlabeled data. In this paper, we explore more general setups where the domain of the unlabeled data for pre-training data differs from the domain of the labeled data for fine-tuning, which in turn may differ from the test data domain. Our experiments show that using target domain data during pre-training leads to large performance improvements across a variety of setups. On a large-scale competitive setup, we show that pre-training on unlabeled in-domain data reduces the gap between models trained on in-domain and out-of-domain labeled data by 66%-73%. This has obvious practical implications since it is much easier to obtain unlabeled target domain data than labeled data. Moreover, we find that pre-training on multiple domains improves generalization performance on domains not seen during training. Code and models will be made available at this https URL. The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20. # Usage See [this notebook](https://colab.research.google.com/drive/1FjTsqbYKphl9kL-eILgUc-bl4zVThL8F?usp=sharing) for more information on how to fine-tune the model.
luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5	45fa2fa42a97b0478a98380d53a7a50ad0177cec	2021-10-26T11:37:15.000Z	[ "pytorch", "roberta", "transformers" ]	null	false	luffycodes	null	luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5	15	null	transformers	9,536	Entry not found
luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5_norm_bb_mlm_loss	ef1df072f813dcd203c30e97b7b273ed1f4e33ad	2021-10-26T04:03:29.000Z	[ "pytorch", "roberta", "transformers" ]	null	false	luffycodes	null	luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5_norm_bb_mlm_loss	15	null	transformers	9,537	Entry not found
m3hrdadfi/zabanshenas-roberta-base-mix	a82c58d3632aaedec457182ea6e65d523fc960b0	2021-06-24T19:40:27.000Z	[ "pytorch", "tf", "roberta", "text-classification", "multilingual", "dataset:wili_2018", "transformers", "license:apache-2.0" ]	text-classification	false	m3hrdadfi	null	m3hrdadfi/zabanshenas-roberta-base-mix	15	1	transformers	9,538	--- language: multilingual license: apache-2.0 datasets: - wili_2018 --- # Zabanshenas - Language Detector Zabanshenas is a Transformer-based solution for identifying the most likely language of a written document/text. Zabanshenas is a Persian word that has two meanings: - A person who studies linguistics. - A way to identify the type of written language. ## How to use Follow [Zabanshenas repo](https://github.com/m3hrdadfi/zabanshenas) for more information! ## Evaluation The following tables summarize the scores obtained by model overall and per each class. ### By Paragraph \| language \| precision \| recall \| f1-score \| \|:--------------------------------------:\|:---------:\|:--------:\|:--------:\| \| Achinese (ace) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Afrikaans (afr) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Alemannic German (als) \| 1.000000 \| 0.946429 \| 0.972477 \| \| Amharic (amh) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Old English (ang) \| 0.981818 \| 0.964286 \| 0.972973 \| \| Arabic (ara) \| 0.846154 \| 0.982143 \| 0.909091 \| \| Aragonese (arg) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Egyptian Arabic (arz) \| 0.979592 \| 0.857143 \| 0.914286 \| \| Assamese (asm) \| 0.981818 \| 0.964286 \| 0.972973 \| \| Asturian (ast) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Avar (ava) \| 0.941176 \| 0.905660 \| 0.923077 \| \| Aymara (aym) \| 0.964912 \| 0.982143 \| 0.973451 \| \| South Azerbaijani (azb) \| 0.965517 \| 1.000000 \| 0.982456 \| \| Azerbaijani (aze) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Bashkir (bak) \| 1.000000 \| 0.978261 \| 0.989011 \| \| Bavarian (bar) \| 0.843750 \| 0.964286 \| 0.900000 \| \| Central Bikol (bcl) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Belarusian (Taraschkewiza) (be-tarask) \| 1.000000 \| 0.875000 \| 0.933333 \| \| Belarusian (bel) \| 0.870968 \| 0.964286 \| 0.915254 \| \| Bengali (ben) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Bhojpuri (bho) \| 1.000000 \| 0.928571 \| 0.962963 \| \| Banjar (bjn) \| 0.981132 \| 0.945455 \| 0.962963 \| \| Tibetan (bod) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Bosnian (bos) \| 0.552632 \| 0.375000 \| 0.446809 \| \| Bishnupriya (bpy) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Breton (bre) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Bulgarian (bul) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Buryat (bxr) \| 0.946429 \| 0.946429 \| 0.946429 \| \| Catalan (cat) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Chavacano (cbk) \| 0.914894 \| 0.767857 \| 0.834951 \| \| Min Dong (cdo) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Cebuano (ceb) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Czech (ces) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Chechen (che) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Cherokee (chr) \| 1.000000 \| 0.963636 \| 0.981481 \| \| Chuvash (chv) \| 0.938776 \| 0.958333 \| 0.948454 \| \| Central Kurdish (ckb) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Cornish (cor) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Corsican (cos) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Crimean Tatar (crh) \| 1.000000 \| 0.946429 \| 0.972477 \| \| Kashubian (csb) \| 1.000000 \| 0.963636 \| 0.981481 \| \| Welsh (cym) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Danish (dan) \| 1.000000 \| 1.000000 \| 1.000000 \| \| German (deu) \| 0.828125 \| 0.946429 \| 0.883333 \| \| Dimli (diq) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Dhivehi (div) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Lower Sorbian (dsb) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Doteli (dty) \| 0.940000 \| 0.854545 \| 0.895238 \| \| Emilian (egl) \| 1.000000 \| 0.928571 \| 0.962963 \| \| Modern Greek (ell) \| 1.000000 \| 1.000000 \| 1.000000 \| \| English (eng) \| 0.588889 \| 0.946429 \| 0.726027 \| \| Esperanto (epo) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Estonian (est) \| 0.963636 \| 0.946429 \| 0.954955 \| \| Basque (eus) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Extremaduran (ext) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Faroese (fao) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Persian (fas) \| 0.948276 \| 0.982143 \| 0.964912 \| \| Finnish (fin) \| 1.000000 \| 1.000000 \| 1.000000 \| \| French (fra) \| 0.710145 \| 0.875000 \| 0.784000 \| \| Arpitan (frp) \| 1.000000 \| 0.946429 \| 0.972477 \| \| Western Frisian (fry) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Friulian (fur) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Gagauz (gag) \| 0.981132 \| 0.945455 \| 0.962963 \| \| Scottish Gaelic (gla) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Irish (gle) \| 0.949153 \| 1.000000 \| 0.973913 \| \| Galician (glg) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Gilaki (glk) \| 0.981132 \| 0.945455 \| 0.962963 \| \| Manx (glv) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Guarani (grn) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Gujarati (guj) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Hakka Chinese (hak) \| 0.981818 \| 0.964286 \| 0.972973 \| \| Haitian Creole (hat) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Hausa (hau) \| 1.000000 \| 0.945455 \| 0.971963 \| \| Serbo-Croatian (hbs) \| 0.448276 \| 0.464286 \| 0.456140 \| \| Hebrew (heb) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Fiji Hindi (hif) \| 0.890909 \| 0.890909 \| 0.890909 \| \| Hindi (hin) \| 0.981481 \| 0.946429 \| 0.963636 \| \| Croatian (hrv) \| 0.500000 \| 0.636364 \| 0.560000 \| \| Upper Sorbian (hsb) \| 0.955556 \| 1.000000 \| 0.977273 \| \| Hungarian (hun) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Armenian (hye) \| 1.000000 \| 0.981818 \| 0.990826 \| \| Igbo (ibo) \| 0.918033 \| 1.000000 \| 0.957265 \| \| Ido (ido) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Interlingue (ile) \| 1.000000 \| 0.962264 \| 0.980769 \| \| Iloko (ilo) \| 0.947368 \| 0.964286 \| 0.955752 \| \| Interlingua (ina) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Indonesian (ind) \| 0.761905 \| 0.872727 \| 0.813559 \| \| Icelandic (isl) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Italian (ita) \| 0.861538 \| 1.000000 \| 0.925620 \| \| Jamaican Patois (jam) \| 1.000000 \| 0.946429 \| 0.972477 \| \| Javanese (jav) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Lojban (jbo) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Japanese (jpn) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Karakalpak (kaa) \| 0.965517 \| 1.000000 \| 0.982456 \| \| Kabyle (kab) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Kannada (kan) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Georgian (kat) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Kazakh (kaz) \| 0.980769 \| 0.980769 \| 0.980769 \| \| Kabardian (kbd) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Central Khmer (khm) \| 0.960784 \| 0.875000 \| 0.915888 \| \| Kinyarwanda (kin) \| 0.981132 \| 0.928571 \| 0.954128 \| \| Kirghiz (kir) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Komi-Permyak (koi) \| 0.962264 \| 0.910714 \| 0.935780 \| \| Konkani (kok) \| 0.964286 \| 0.981818 \| 0.972973 \| \| Komi (kom) \| 1.000000 \| 0.962264 \| 0.980769 \| \| Korean (kor) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Karachay-Balkar (krc) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Ripuarisch (ksh) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Kurdish (kur) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Ladino (lad) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Lao (lao) \| 0.961538 \| 0.909091 \| 0.934579 \| \| Latin (lat) \| 0.877193 \| 0.943396 \| 0.909091 \| \| Latvian (lav) \| 0.963636 \| 0.946429 \| 0.954955 \| \| Lezghian (lez) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Ligurian (lij) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Limburgan (lim) \| 0.938776 \| 1.000000 \| 0.968421 \| \| Lingala (lin) \| 0.980769 \| 0.927273 \| 0.953271 \| \| Lithuanian (lit) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Lombard (lmo) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Northern Luri (lrc) \| 1.000000 \| 0.928571 \| 0.962963 \| \| Latgalian (ltg) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Luxembourgish (ltz) \| 0.949153 \| 1.000000 \| 0.973913 \| \| Luganda (lug) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Literary Chinese (lzh) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Maithili (mai) \| 0.931034 \| 0.964286 \| 0.947368 \| \| Malayalam (mal) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Banyumasan (map-bms) \| 0.977778 \| 0.785714 \| 0.871287 \| \| Marathi (mar) \| 0.949153 \| 1.000000 \| 0.973913 \| \| Moksha (mdf) \| 0.980000 \| 0.890909 \| 0.933333 \| \| Eastern Mari (mhr) \| 0.981818 \| 0.964286 \| 0.972973 \| \| Minangkabau (min) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Macedonian (mkd) \| 1.000000 \| 0.981818 \| 0.990826 \| \| Malagasy (mlg) \| 0.981132 \| 1.000000 \| 0.990476 \| \| Maltese (mlt) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Min Nan Chinese (nan) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Mongolian (mon) \| 1.000000 \| 0.981818 \| 0.990826 \| \| Maori (mri) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Western Mari (mrj) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Malay (msa) \| 0.862069 \| 0.892857 \| 0.877193 \| \| Mirandese (mwl) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Burmese (mya) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Erzya (myv) \| 0.818182 \| 0.964286 \| 0.885246 \| \| Mazanderani (mzn) \| 0.981481 \| 1.000000 \| 0.990654 \| \| Neapolitan (nap) \| 1.000000 \| 0.981818 \| 0.990826 \| \| Navajo (nav) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Classical Nahuatl (nci) \| 0.981481 \| 0.946429 \| 0.963636 \| \| Low German (nds) \| 0.982143 \| 0.982143 \| 0.982143 \| \| West Low German (nds-nl) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Nepali (macrolanguage) (nep) \| 0.881356 \| 0.928571 \| 0.904348 \| \| Newari (new) \| 1.000000 \| 0.909091 \| 0.952381 \| \| Dutch (nld) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Norwegian Nynorsk (nno) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Bokmål (nob) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Narom (nrm) \| 0.981818 \| 0.964286 \| 0.972973 \| \| Northern Sotho (nso) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Occitan (oci) \| 0.903846 \| 0.839286 \| 0.870370 \| \| Livvi-Karelian (olo) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Oriya (ori) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Oromo (orm) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Ossetian (oss) \| 0.982143 \| 1.000000 \| 0.990991 \| \| Pangasinan (pag) \| 0.980000 \| 0.875000 \| 0.924528 \| \| Pampanga (pam) \| 0.928571 \| 0.896552 \| 0.912281 \| \| Panjabi (pan) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Papiamento (pap) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Picard (pcd) \| 0.849057 \| 0.849057 \| 0.849057 \| \| Pennsylvania German (pdc) \| 0.854839 \| 0.946429 \| 0.898305 \| \| Palatine German (pfl) \| 0.946429 \| 0.946429 \| 0.946429 \| \| Western Panjabi (pnb) \| 0.981132 \| 0.962963 \| 0.971963 \| \| Polish (pol) \| 0.933333 \| 1.000000 \| 0.965517 \| \| Portuguese (por) \| 0.774648 \| 0.982143 \| 0.866142 \| \| Pushto (pus) \| 1.000000 \| 0.910714 \| 0.953271 \| \| Quechua (que) \| 0.962963 \| 0.928571 \| 0.945455 \| \| Tarantino dialect (roa-tara) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Romansh (roh) \| 1.000000 \| 0.928571 \| 0.962963 \| \| Romanian (ron) \| 0.965517 \| 1.000000 \| 0.982456 \| \| Rusyn (rue) \| 0.946429 \| 0.946429 \| 0.946429 \| \| Aromanian (rup) \| 0.962963 \| 0.928571 \| 0.945455 \| \| Russian (rus) \| 0.859375 \| 0.982143 \| 0.916667 \| \| Yakut (sah) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Sanskrit (san) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Sicilian (scn) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Scots (sco) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Samogitian (sgs) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Sinhala (sin) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Slovak (slk) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Slovene (slv) \| 1.000000 \| 0.981818 \| 0.990826 \| \| Northern Sami (sme) \| 0.962264 \| 0.962264 \| 0.962264 \| \| Shona (sna) \| 0.933333 \| 1.000000 \| 0.965517 \| \| Sindhi (snd) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Somali (som) \| 0.948276 \| 1.000000 \| 0.973451 \| \| Spanish (spa) \| 0.739130 \| 0.910714 \| 0.816000 \| \| Albanian (sqi) \| 0.982143 \| 0.982143 \| 0.982143 \| \| Sardinian (srd) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Sranan (srn) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Serbian (srp) \| 1.000000 \| 0.946429 \| 0.972477 \| \| Saterfriesisch (stq) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Sundanese (sun) \| 1.000000 \| 0.977273 \| 0.988506 \| \| Swahili (macrolanguage) (swa) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Swedish (swe) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Silesian (szl) \| 1.000000 \| 0.981481 \| 0.990654 \| \| Tamil (tam) \| 0.982143 \| 1.000000 \| 0.990991 \| \| Tatar (tat) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Tulu (tcy) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Telugu (tel) \| 1.000000 \| 0.920000 \| 0.958333 \| \| Tetum (tet) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Tajik (tgk) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Tagalog (tgl) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Thai (tha) \| 0.932203 \| 0.982143 \| 0.956522 \| \| Tongan (ton) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Tswana (tsn) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Turkmen (tuk) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Turkish (tur) \| 0.901639 \| 0.982143 \| 0.940171 \| \| Tuvan (tyv) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Udmurt (udm) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Uighur (uig) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Ukrainian (ukr) \| 0.963636 \| 0.946429 \| 0.954955 \| \| Urdu (urd) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Uzbek (uzb) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Venetian (vec) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Veps (vep) \| 0.982456 \| 1.000000 \| 0.991150 \| \| Vietnamese (vie) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Vlaams (vls) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Volapük (vol) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Võro (vro) \| 0.964286 \| 0.964286 \| 0.964286 \| \| Waray (war) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Walloon (wln) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Wolof (wol) \| 0.981481 \| 0.963636 \| 0.972477 \| \| Wu Chinese (wuu) \| 0.981481 \| 0.946429 \| 0.963636 \| \| Xhosa (xho) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Mingrelian (xmf) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Yiddish (yid) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Yoruba (yor) \| 0.964912 \| 0.982143 \| 0.973451 \| \| Zeeuws (zea) \| 1.000000 \| 0.982143 \| 0.990991 \| \| Cantonese (zh-yue) \| 0.981481 \| 0.946429 \| 0.963636 \| \| Standard Chinese (zho) \| 0.932203 \| 0.982143 \| 0.956522 \| \| accuracy \| 0.963055 \| 0.963055 \| 0.963055 \| \| macro avg \| 0.966424 \| 0.963216 \| 0.963891 \| \| weighted avg \| 0.966040 \| 0.963055 \| 0.963606 \| ### By Sentence \| language \| precision \| recall \| f1-score \| \|:--------------------------------------:\|:---------:\|:--------:\|:--------:\| \| Achinese (ace) \| 0.754545 \| 0.873684 \| 0.809756 \| \| Afrikaans (afr) \| 0.708955 \| 0.940594 \| 0.808511 \| \| Alemannic German (als) \| 0.870130 \| 0.752809 \| 0.807229 \| \| Amharic (amh) \| 1.000000 \| 0.820000 \| 0.901099 \| \| Old English (ang) \| 0.966667 \| 0.906250 \| 0.935484 \| \| Arabic (ara) \| 0.907692 \| 0.967213 \| 0.936508 \| \| Aragonese (arg) \| 0.921569 \| 0.959184 \| 0.940000 \| \| Egyptian Arabic (arz) \| 0.964286 \| 0.843750 \| 0.900000 \| \| Assamese (asm) \| 0.964286 \| 0.870968 \| 0.915254 \| \| Asturian (ast) \| 0.880000 \| 0.795181 \| 0.835443 \| \| Avar (ava) \| 0.864198 \| 0.843373 \| 0.853659 \| \| Aymara (aym) \| 1.000000 \| 0.901961 \| 0.948454 \| \| South Azerbaijani (azb) \| 0.979381 \| 0.989583 \| 0.984456 \| \| Azerbaijani (aze) \| 0.989899 \| 0.960784 \| 0.975124 \| \| Bashkir (bak) \| 0.837209 \| 0.857143 \| 0.847059 \| \| Bavarian (bar) \| 0.741935 \| 0.766667 \| 0.754098 \| \| Central Bikol (bcl) \| 0.962963 \| 0.928571 \| 0.945455 \| \| Belarusian (Taraschkewiza) (be-tarask) \| 0.857143 \| 0.733333 \| 0.790419 \| \| Belarusian (bel) \| 0.775510 \| 0.752475 \| 0.763819 \| \| Bengali (ben) \| 0.861111 \| 0.911765 \| 0.885714 \| \| Bhojpuri (bho) \| 0.965517 \| 0.933333 \| 0.949153 \| \| Banjar (bjn) \| 0.891566 \| 0.880952 \| 0.886228 \| \| Tibetan (bod) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Bosnian (bos) \| 0.375000 \| 0.323077 \| 0.347107 \| \| Bishnupriya (bpy) \| 0.986301 \| 1.000000 \| 0.993103 \| \| Breton (bre) \| 0.951613 \| 0.893939 \| 0.921875 \| \| Bulgarian (bul) \| 0.945055 \| 0.877551 \| 0.910053 \| \| Buryat (bxr) \| 0.955556 \| 0.843137 \| 0.895833 \| \| Catalan (cat) \| 0.692308 \| 0.750000 \| 0.720000 \| \| Chavacano (cbk) \| 0.842857 \| 0.641304 \| 0.728395 \| \| Min Dong (cdo) \| 0.972973 \| 1.000000 \| 0.986301 \| \| Cebuano (ceb) \| 0.981308 \| 0.954545 \| 0.967742 \| \| Czech (ces) \| 0.944444 \| 0.915385 \| 0.929687 \| \| Chechen (che) \| 0.875000 \| 0.700000 \| 0.777778 \| \| Cherokee (chr) \| 1.000000 \| 0.970588 \| 0.985075 \| \| Chuvash (chv) \| 0.875000 \| 0.836957 \| 0.855556 \| \| Central Kurdish (ckb) \| 1.000000 \| 0.983051 \| 0.991453 \| \| Cornish (cor) \| 0.979592 \| 0.969697 \| 0.974619 \| \| Corsican (cos) \| 0.986842 \| 0.925926 \| 0.955414 \| \| Crimean Tatar (crh) \| 0.958333 \| 0.907895 \| 0.932432 \| \| Kashubian (csb) \| 0.920354 \| 0.904348 \| 0.912281 \| \| Welsh (cym) \| 0.971014 \| 0.943662 \| 0.957143 \| \| Danish (dan) \| 0.865169 \| 0.777778 \| 0.819149 \| \| German (deu) \| 0.721311 \| 0.822430 \| 0.768559 \| \| Dimli (diq) \| 0.915966 \| 0.923729 \| 0.919831 \| \| Dhivehi (div) \| 1.000000 \| 0.991228 \| 0.995595 \| \| Lower Sorbian (dsb) \| 0.898876 \| 0.879121 \| 0.888889 \| \| Doteli (dty) \| 0.821429 \| 0.638889 \| 0.718750 \| \| Emilian (egl) \| 0.988095 \| 0.922222 \| 0.954023 \| \| Modern Greek (ell) \| 0.988636 \| 0.966667 \| 0.977528 \| \| English (eng) \| 0.522727 \| 0.784091 \| 0.627273 \| \| Esperanto (epo) \| 0.963855 \| 0.930233 \| 0.946746 \| \| Estonian (est) \| 0.922222 \| 0.873684 \| 0.897297 \| \| Basque (eus) \| 1.000000 \| 0.941176 \| 0.969697 \| \| Extremaduran (ext) \| 0.925373 \| 0.885714 \| 0.905109 \| \| Faroese (fao) \| 0.855072 \| 0.887218 \| 0.870849 \| \| Persian (fas) \| 0.879630 \| 0.979381 \| 0.926829 \| \| Finnish (fin) \| 0.952830 \| 0.943925 \| 0.948357 \| \| French (fra) \| 0.676768 \| 0.943662 \| 0.788235 \| \| Arpitan (frp) \| 0.867925 \| 0.807018 \| 0.836364 \| \| Western Frisian (fry) \| 0.956989 \| 0.890000 \| 0.922280 \| \| Friulian (fur) \| 1.000000 \| 0.857143 \| 0.923077 \| \| Gagauz (gag) \| 0.939024 \| 0.802083 \| 0.865169 \| \| Scottish Gaelic (gla) \| 1.000000 \| 0.879121 \| 0.935673 \| \| Irish (gle) \| 0.989247 \| 0.958333 \| 0.973545 \| \| Galician (glg) \| 0.910256 \| 0.922078 \| 0.916129 \| \| Gilaki (glk) \| 0.964706 \| 0.872340 \| 0.916201 \| \| Manx (glv) \| 1.000000 \| 0.965517 \| 0.982456 \| \| Guarani (grn) \| 0.983333 \| 1.000000 \| 0.991597 \| \| Gujarati (guj) \| 1.000000 \| 0.991525 \| 0.995745 \| \| Hakka Chinese (hak) \| 0.955224 \| 0.955224 \| 0.955224 \| \| Haitian Creole (hat) \| 0.833333 \| 0.666667 \| 0.740741 \| \| Hausa (hau) \| 0.936709 \| 0.913580 \| 0.925000 \| \| Serbo-Croatian (hbs) \| 0.452830 \| 0.410256 \| 0.430493 \| \| Hebrew (heb) \| 0.988235 \| 0.976744 \| 0.982456 \| \| Fiji Hindi (hif) \| 0.936709 \| 0.840909 \| 0.886228 \| \| Hindi (hin) \| 0.965517 \| 0.756757 \| 0.848485 \| \| Croatian (hrv) \| 0.443820 \| 0.537415 \| 0.486154 \| \| Upper Sorbian (hsb) \| 0.951613 \| 0.830986 \| 0.887218 \| \| Hungarian (hun) \| 0.854701 \| 0.909091 \| 0.881057 \| \| Armenian (hye) \| 1.000000 \| 0.816327 \| 0.898876 \| \| Igbo (ibo) \| 0.974359 \| 0.926829 \| 0.950000 \| \| Ido (ido) \| 0.975000 \| 0.987342 \| 0.981132 \| \| Interlingue (ile) \| 0.880597 \| 0.921875 \| 0.900763 \| \| Iloko (ilo) \| 0.882353 \| 0.821918 \| 0.851064 \| \| Interlingua (ina) \| 0.952381 \| 0.895522 \| 0.923077 \| \| Indonesian (ind) \| 0.606383 \| 0.695122 \| 0.647727 \| \| Icelandic (isl) \| 0.978261 \| 0.882353 \| 0.927835 \| \| Italian (ita) \| 0.910448 \| 0.910448 \| 0.910448 \| \| Jamaican Patois (jam) \| 0.988764 \| 0.967033 \| 0.977778 \| \| Javanese (jav) \| 0.903614 \| 0.862069 \| 0.882353 \| \| Lojban (jbo) \| 0.943878 \| 0.929648 \| 0.936709 \| \| Japanese (jpn) \| 1.000000 \| 0.764706 \| 0.866667 \| \| Karakalpak (kaa) \| 0.940171 \| 0.901639 \| 0.920502 \| \| Kabyle (kab) \| 0.985294 \| 0.837500 \| 0.905405 \| \| Kannada (kan) \| 0.975806 \| 0.975806 \| 0.975806 \| \| Georgian (kat) \| 0.953704 \| 0.903509 \| 0.927928 \| \| Kazakh (kaz) \| 0.934579 \| 0.877193 \| 0.904977 \| \| Kabardian (kbd) \| 0.987952 \| 0.953488 \| 0.970414 \| \| Central Khmer (khm) \| 0.928571 \| 0.829787 \| 0.876404 \| \| Kinyarwanda (kin) \| 0.953125 \| 0.938462 \| 0.945736 \| \| Kirghiz (kir) \| 0.927632 \| 0.881250 \| 0.903846 \| \| Komi-Permyak (koi) \| 0.750000 \| 0.776786 \| 0.763158 \| \| Konkani (kok) \| 0.893491 \| 0.872832 \| 0.883041 \| \| Komi (kom) \| 0.734177 \| 0.690476 \| 0.711656 \| \| Korean (kor) \| 0.989899 \| 0.989899 \| 0.989899 \| \| Karachay-Balkar (krc) \| 0.928571 \| 0.917647 \| 0.923077 \| \| Ripuarisch (ksh) \| 0.915789 \| 0.896907 \| 0.906250 \| \| Kurdish (kur) \| 0.977528 \| 0.935484 \| 0.956044 \| \| Ladino (lad) \| 0.985075 \| 0.904110 \| 0.942857 \| \| Lao (lao) \| 0.896552 \| 0.812500 \| 0.852459 \| \| Latin (lat) \| 0.741935 \| 0.831325 \| 0.784091 \| \| Latvian (lav) \| 0.710526 \| 0.878049 \| 0.785455 \| \| Lezghian (lez) \| 0.975309 \| 0.877778 \| 0.923977 \| \| Ligurian (lij) \| 0.951807 \| 0.897727 \| 0.923977 \| \| Limburgan (lim) \| 0.909091 \| 0.921053 \| 0.915033 \| \| Lingala (lin) \| 0.942857 \| 0.814815 \| 0.874172 \| \| Lithuanian (lit) \| 0.892857 \| 0.925926 \| 0.909091 \| \| Lombard (lmo) \| 0.766234 \| 0.951613 \| 0.848921 \| \| Northern Luri (lrc) \| 0.972222 \| 0.875000 \| 0.921053 \| \| Latgalian (ltg) \| 0.895349 \| 0.865169 \| 0.880000 \| \| Luxembourgish (ltz) \| 0.882353 \| 0.750000 \| 0.810811 \| \| Luganda (lug) \| 0.946429 \| 0.883333 \| 0.913793 \| \| Literary Chinese (lzh) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Maithili (mai) \| 0.893617 \| 0.823529 \| 0.857143 \| \| Malayalam (mal) \| 1.000000 \| 0.975000 \| 0.987342 \| \| Banyumasan (map-bms) \| 0.924242 \| 0.772152 \| 0.841379 \| \| Marathi (mar) \| 0.874126 \| 0.919118 \| 0.896057 \| \| Moksha (mdf) \| 0.771242 \| 0.830986 \| 0.800000 \| \| Eastern Mari (mhr) \| 0.820000 \| 0.860140 \| 0.839590 \| \| Minangkabau (min) \| 0.973684 \| 0.973684 \| 0.973684 \| \| Macedonian (mkd) \| 0.895652 \| 0.953704 \| 0.923767 \| \| Malagasy (mlg) \| 1.000000 \| 0.966102 \| 0.982759 \| \| Maltese (mlt) \| 0.987952 \| 0.964706 \| 0.976190 \| \| Min Nan Chinese (nan) \| 0.975000 \| 1.000000 \| 0.987342 \| \| Mongolian (mon) \| 0.954545 \| 0.933333 \| 0.943820 \| \| Maori (mri) \| 0.985294 \| 1.000000 \| 0.992593 \| \| Western Mari (mrj) \| 0.966292 \| 0.914894 \| 0.939891 \| \| Malay (msa) \| 0.770270 \| 0.695122 \| 0.730769 \| \| Mirandese (mwl) \| 0.970588 \| 0.891892 \| 0.929577 \| \| Burmese (mya) \| 1.000000 \| 0.964286 \| 0.981818 \| \| Erzya (myv) \| 0.535714 \| 0.681818 \| 0.600000 \| \| Mazanderani (mzn) \| 0.968750 \| 0.898551 \| 0.932331 \| \| Neapolitan (nap) \| 0.892308 \| 0.865672 \| 0.878788 \| \| Navajo (nav) \| 0.984375 \| 0.984375 \| 0.984375 \| \| Classical Nahuatl (nci) \| 0.901408 \| 0.761905 \| 0.825806 \| \| Low German (nds) \| 0.896226 \| 0.913462 \| 0.904762 \| \| West Low German (nds-nl) \| 0.873563 \| 0.835165 \| 0.853933 \| \| Nepali (macrolanguage) (nep) \| 0.704545 \| 0.861111 \| 0.775000 \| \| Newari (new) \| 0.920000 \| 0.741935 \| 0.821429 \| \| Dutch (nld) \| 0.925926 \| 0.872093 \| 0.898204 \| \| Norwegian Nynorsk (nno) \| 0.847059 \| 0.808989 \| 0.827586 \| \| Bokmål (nob) \| 0.861386 \| 0.852941 \| 0.857143 \| \| Narom (nrm) \| 0.966667 \| 0.983051 \| 0.974790 \| \| Northern Sotho (nso) \| 0.897436 \| 0.921053 \| 0.909091 \| \| Occitan (oci) \| 0.958333 \| 0.696970 \| 0.807018 \| \| Livvi-Karelian (olo) \| 0.967742 \| 0.937500 \| 0.952381 \| \| Oriya (ori) \| 0.933333 \| 1.000000 \| 0.965517 \| \| Oromo (orm) \| 0.977528 \| 0.915789 \| 0.945652 \| \| Ossetian (oss) \| 0.958333 \| 0.841463 \| 0.896104 \| \| Pangasinan (pag) \| 0.847328 \| 0.909836 \| 0.877470 \| \| Pampanga (pam) \| 0.969697 \| 0.780488 \| 0.864865 \| \| Panjabi (pan) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Papiamento (pap) \| 0.876190 \| 0.920000 \| 0.897561 \| \| Picard (pcd) \| 0.707317 \| 0.568627 \| 0.630435 \| \| Pennsylvania German (pdc) \| 0.827273 \| 0.827273 \| 0.827273 \| \| Palatine German (pfl) \| 0.882353 \| 0.914634 \| 0.898204 \| \| Western Panjabi (pnb) \| 0.964286 \| 0.931034 \| 0.947368 \| \| Polish (pol) \| 0.859813 \| 0.910891 \| 0.884615 \| \| Portuguese (por) \| 0.535714 \| 0.833333 \| 0.652174 \| \| Pushto (pus) \| 0.989362 \| 0.902913 \| 0.944162 \| \| Quechua (que) \| 0.979167 \| 0.903846 \| 0.940000 \| \| Tarantino dialect (roa-tara) \| 0.964912 \| 0.901639 \| 0.932203 \| \| Romansh (roh) \| 0.914894 \| 0.895833 \| 0.905263 \| \| Romanian (ron) \| 0.880597 \| 0.880597 \| 0.880597 \| \| Rusyn (rue) \| 0.932584 \| 0.805825 \| 0.864583 \| \| Aromanian (rup) \| 0.783333 \| 0.758065 \| 0.770492 \| \| Russian (rus) \| 0.517986 \| 0.765957 \| 0.618026 \| \| Yakut (sah) \| 0.954023 \| 0.922222 \| 0.937853 \| \| Sanskrit (san) \| 0.866667 \| 0.951220 \| 0.906977 \| \| Sicilian (scn) \| 0.984375 \| 0.940299 \| 0.961832 \| \| Scots (sco) \| 0.851351 \| 0.900000 \| 0.875000 \| \| Samogitian (sgs) \| 0.977011 \| 0.876289 \| 0.923913 \| \| Sinhala (sin) \| 0.406154 \| 0.985075 \| 0.575163 \| \| Slovak (slk) \| 0.956989 \| 0.872549 \| 0.912821 \| \| Slovene (slv) \| 0.907216 \| 0.854369 \| 0.880000 \| \| Northern Sami (sme) \| 0.949367 \| 0.892857 \| 0.920245 \| \| Shona (sna) \| 0.936508 \| 0.855072 \| 0.893939 \| \| Sindhi (snd) \| 0.984962 \| 0.992424 \| 0.988679 \| \| Somali (som) \| 0.949153 \| 0.848485 \| 0.896000 \| \| Spanish (spa) \| 0.584158 \| 0.746835 \| 0.655556 \| \| Albanian (sqi) \| 0.988095 \| 0.912088 \| 0.948571 \| \| Sardinian (srd) \| 0.957746 \| 0.931507 \| 0.944444 \| \| Sranan (srn) \| 0.985714 \| 0.945205 \| 0.965035 \| \| Serbian (srp) \| 0.950980 \| 0.889908 \| 0.919431 \| \| Saterfriesisch (stq) \| 0.962500 \| 0.875000 \| 0.916667 \| \| Sundanese (sun) \| 0.778846 \| 0.910112 \| 0.839378 \| \| Swahili (macrolanguage) (swa) \| 0.915493 \| 0.878378 \| 0.896552 \| \| Swedish (swe) \| 0.989247 \| 0.958333 \| 0.973545 \| \| Silesian (szl) \| 0.944444 \| 0.904255 \| 0.923913 \| \| Tamil (tam) \| 0.990000 \| 0.970588 \| 0.980198 \| \| Tatar (tat) \| 0.942029 \| 0.902778 \| 0.921986 \| \| Tulu (tcy) \| 0.980519 \| 0.967949 \| 0.974194 \| \| Telugu (tel) \| 0.965986 \| 0.965986 \| 0.965986 \| \| Tetum (tet) \| 0.898734 \| 0.855422 \| 0.876543 \| \| Tajik (tgk) \| 0.974684 \| 0.939024 \| 0.956522 \| \| Tagalog (tgl) \| 0.965909 \| 0.934066 \| 0.949721 \| \| Thai (tha) \| 0.923077 \| 0.882353 \| 0.902256 \| \| Tongan (ton) \| 0.970149 \| 0.890411 \| 0.928571 \| \| Tswana (tsn) \| 0.888889 \| 0.926316 \| 0.907216 \| \| Turkmen (tuk) \| 0.968000 \| 0.889706 \| 0.927203 \| \| Turkish (tur) \| 0.871287 \| 0.926316 \| 0.897959 \| \| Tuvan (tyv) \| 0.948454 \| 0.859813 \| 0.901961 \| \| Udmurt (udm) \| 0.989362 \| 0.894231 \| 0.939394 \| \| Uighur (uig) \| 1.000000 \| 0.953333 \| 0.976109 \| \| Ukrainian (ukr) \| 0.893617 \| 0.875000 \| 0.884211 \| \| Urdu (urd) \| 1.000000 \| 1.000000 \| 1.000000 \| \| Uzbek (uzb) \| 0.636042 \| 0.886700 \| 0.740741 \| \| Venetian (vec) \| 1.000000 \| 0.941176 \| 0.969697 \| \| Veps (vep) \| 0.858586 \| 0.965909 \| 0.909091 \| \| Vietnamese (vie) \| 1.000000 \| 0.940476 \| 0.969325 \| \| Vlaams (vls) \| 0.885714 \| 0.898551 \| 0.892086 \| \| Volapük (vol) \| 0.975309 \| 0.975309 \| 0.975309 \| \| Võro (vro) \| 0.855670 \| 0.864583 \| 0.860104 \| \| Waray (war) \| 0.972222 \| 0.909091 \| 0.939597 \| \| Walloon (wln) \| 0.742138 \| 0.893939 \| 0.810997 \| \| Wolof (wol) \| 0.882979 \| 0.954023 \| 0.917127 \| \| Wu Chinese (wuu) \| 0.961538 \| 0.833333 \| 0.892857 \| \| Xhosa (xho) \| 0.934066 \| 0.867347 \| 0.899471 \| \| Mingrelian (xmf) \| 0.958333 \| 0.929293 \| 0.943590 \| \| Yiddish (yid) \| 0.984375 \| 0.875000 \| 0.926471 \| \| Yoruba (yor) \| 0.868421 \| 0.857143 \| 0.862745 \| \| Zeeuws (zea) \| 0.879518 \| 0.793478 \| 0.834286 \| \| Cantonese (zh-yue) \| 0.896552 \| 0.812500 \| 0.852459 \| \| Standard Chinese (zho) \| 0.906250 \| 0.935484 \| 0.920635 \| \| accuracy \| 0.881051 \| 0.881051 \| 0.881051 \| \| macro avg \| 0.903245 \| 0.880618 \| 0.888996 \| \| weighted avg \| 0.894174 \| 0.881051 \| 0.884520 \| ### By Token (3 to 5) \| language \| precision \| recall \| f1-score \| \|:--------------------------------------:\|:---------:\|:--------:\|:--------:\| \| Achinese (ace) \| 0.873846 \| 0.827988 \| 0.850299 \| \| Afrikaans (afr) \| 0.638060 \| 0.732334 \| 0.681954 \| \| Alemannic German (als) \| 0.673780 \| 0.547030 \| 0.603825 \| \| Amharic (amh) \| 0.997743 \| 0.954644 \| 0.975717 \| \| Old English (ang) \| 0.840816 \| 0.693603 \| 0.760148 \| \| Arabic (ara) \| 0.768737 \| 0.840749 \| 0.803132 \| \| Aragonese (arg) \| 0.493671 \| 0.505181 \| 0.499360 \| \| Egyptian Arabic (arz) \| 0.823529 \| 0.741935 \| 0.780606 \| \| Assamese (asm) \| 0.948454 \| 0.893204 \| 0.920000 \| \| Asturian (ast) \| 0.490000 \| 0.508299 \| 0.498982 \| \| Avar (ava) \| 0.813636 \| 0.655678 \| 0.726166 \| \| Aymara (aym) \| 0.795833 \| 0.779592 \| 0.787629 \| \| South Azerbaijani (azb) \| 0.832836 \| 0.863777 \| 0.848024 \| \| Azerbaijani (aze) \| 0.867470 \| 0.800000 \| 0.832370 \| \| Bashkir (bak) \| 0.851852 \| 0.750000 \| 0.797688 \| \| Bavarian (bar) \| 0.560897 \| 0.522388 \| 0.540958 \| \| Central Bikol (bcl) \| 0.708229 \| 0.668235 \| 0.687651 \| \| Belarusian (Taraschkewiza) (be-tarask) \| 0.615635 \| 0.526462 \| 0.567568 \| \| Belarusian (bel) \| 0.539952 \| 0.597855 \| 0.567430 \| \| Bengali (ben) \| 0.830275 \| 0.885086 \| 0.856805 \| \| Bhojpuri (bho) \| 0.723118 \| 0.691517 \| 0.706965 \| \| Banjar (bjn) \| 0.619586 \| 0.726269 \| 0.668699 \| \| Tibetan (bod) \| 0.999537 \| 0.991728 \| 0.995617 \| \| Bosnian (bos) \| 0.330849 \| 0.403636 \| 0.363636 \| \| Bishnupriya (bpy) \| 0.941634 \| 0.949020 \| 0.945312 \| \| Breton (bre) \| 0.772222 \| 0.745308 \| 0.758527 \| \| Bulgarian (bul) \| 0.771505 \| 0.706897 \| 0.737789 \| \| Buryat (bxr) \| 0.741935 \| 0.753149 \| 0.747500 \| \| Catalan (cat) \| 0.528716 \| 0.610136 \| 0.566516 \| \| Chavacano (cbk) \| 0.409449 \| 0.312625 \| 0.354545 \| \| Min Dong (cdo) \| 0.951264 \| 0.936057 \| 0.943599 \| \| Cebuano (ceb) \| 0.888298 \| 0.876640 \| 0.882431 \| \| Czech (ces) \| 0.806045 \| 0.758294 \| 0.781441 \| \| Chechen (che) \| 0.857143 \| 0.600000 \| 0.705882 \| \| Cherokee (chr) \| 0.997840 \| 0.952577 \| 0.974684 \| \| Chuvash (chv) \| 0.874346 \| 0.776744 \| 0.822660 \| \| Central Kurdish (ckb) \| 0.984848 \| 0.953545 \| 0.968944 \| \| Cornish (cor) \| 0.747596 \| 0.807792 \| 0.776529 \| \| Corsican (cos) \| 0.673913 \| 0.708571 \| 0.690808 \| \| Crimean Tatar (crh) \| 0.498801 \| 0.700337 \| 0.582633 \| \| Kashubian (csb) \| 0.797059 \| 0.794721 \| 0.795888 \| \| Welsh (cym) \| 0.829609 \| 0.841360 \| 0.835443 \| \| Danish (dan) \| 0.649789 \| 0.622222 \| 0.635707 \| \| German (deu) \| 0.559406 \| 0.763514 \| 0.645714 \| \| Dimli (diq) \| 0.835580 \| 0.763547 \| 0.797941 \| \| Dhivehi (div) \| 1.000000 \| 0.980645 \| 0.990228 \| \| Lower Sorbian (dsb) \| 0.740484 \| 0.694805 \| 0.716918 \| \| Doteli (dty) \| 0.616314 \| 0.527132 \| 0.568245 \| \| Emilian (egl) \| 0.822993 \| 0.769625 \| 0.795414 \| \| Modern Greek (ell) \| 0.972043 \| 0.963753 \| 0.967880 \| \| English (eng) \| 0.260492 \| 0.724346 \| 0.383183 \| \| Esperanto (epo) \| 0.766764 \| 0.716621 \| 0.740845 \| \| Estonian (est) \| 0.698885 \| 0.673835 \| 0.686131 \| \| Basque (eus) \| 0.882716 \| 0.841176 \| 0.861446 \| \| Extremaduran (ext) \| 0.570605 \| 0.511628 \| 0.539510 \| \| Faroese (fao) \| 0.773987 \| 0.784017 \| 0.778970 \| \| Persian (fas) \| 0.709836 \| 0.809346 \| 0.756332 \| \| Finnish (fin) \| 0.866261 \| 0.796089 \| 0.829694 \| \| French (fra) \| 0.496263 \| 0.700422 \| 0.580927 \| \| Arpitan (frp) \| 0.663366 \| 0.584302 \| 0.621329 \| \| Western Frisian (fry) \| 0.750000 \| 0.756148 \| 0.753061 \| \| Friulian (fur) \| 0.713555 \| 0.675545 \| 0.694030 \| \| Gagauz (gag) \| 0.728125 \| 0.677326 \| 0.701807 \| \| Scottish Gaelic (gla) \| 0.831601 \| 0.817996 \| 0.824742 \| \| Irish (gle) \| 0.868852 \| 0.801296 \| 0.833708 \| \| Galician (glg) \| 0.469816 \| 0.454315 \| 0.461935 \| \| Gilaki (glk) \| 0.703883 \| 0.687204 \| 0.695444 \| \| Manx (glv) \| 0.873047 \| 0.886905 \| 0.879921 \| \| Guarani (grn) \| 0.848580 \| 0.793510 \| 0.820122 \| \| Gujarati (guj) \| 0.995643 \| 0.926978 \| 0.960084 \| \| Hakka Chinese (hak) \| 0.898403 \| 0.904971 \| 0.901675 \| \| Haitian Creole (hat) \| 0.719298 \| 0.518987 \| 0.602941 \| \| Hausa (hau) \| 0.815353 \| 0.829114 \| 0.822176 \| \| Serbo-Croatian (hbs) \| 0.343465 \| 0.244589 \| 0.285714 \| \| Hebrew (heb) \| 0.891304 \| 0.933941 \| 0.912125 \| \| Fiji Hindi (hif) \| 0.662577 \| 0.664615 \| 0.663594 \| \| Hindi (hin) \| 0.782301 \| 0.778169 \| 0.780229 \| \| Croatian (hrv) \| 0.360308 \| 0.374000 \| 0.367026 \| \| Upper Sorbian (hsb) \| 0.745763 \| 0.611111 \| 0.671756 \| \| Hungarian (hun) \| 0.876812 \| 0.846154 \| 0.861210 \| \| Armenian (hye) \| 0.988201 \| 0.917808 \| 0.951705 \| \| Igbo (ibo) \| 0.825397 \| 0.696429 \| 0.755448 \| \| Ido (ido) \| 0.760479 \| 0.814103 \| 0.786378 \| \| Interlingue (ile) \| 0.701299 \| 0.580645 \| 0.635294 \| \| Iloko (ilo) \| 0.688356 \| 0.844538 \| 0.758491 \| \| Interlingua (ina) \| 0.577889 \| 0.588235 \| 0.583016 \| \| Indonesian (ind) \| 0.415879 \| 0.514019 \| 0.459770 \| \| Icelandic (isl) \| 0.855263 \| 0.790754 \| 0.821745 \| \| Italian (ita) \| 0.474576 \| 0.561247 \| 0.514286 \| \| Jamaican Patois (jam) \| 0.826087 \| 0.791667 \| 0.808511 \| \| Javanese (jav) \| 0.670130 \| 0.658163 \| 0.664093 \| \| Lojban (jbo) \| 0.896861 \| 0.917431 \| 0.907029 \| \| Japanese (jpn) \| 0.931373 \| 0.848214 \| 0.887850 \| \| Karakalpak (kaa) \| 0.790393 \| 0.827744 \| 0.808637 \| \| Kabyle (kab) \| 0.828571 \| 0.759162 \| 0.792350 \| \| Kannada (kan) \| 0.879357 \| 0.847545 \| 0.863158 \| \| Georgian (kat) \| 0.916399 \| 0.907643 \| 0.912000 \| \| Kazakh (kaz) \| 0.900901 \| 0.819672 \| 0.858369 \| \| Kabardian (kbd) \| 0.923345 \| 0.892256 \| 0.907534 \| \| Central Khmer (khm) \| 0.976667 \| 0.816156 \| 0.889226 \| \| Kinyarwanda (kin) \| 0.824324 \| 0.726190 \| 0.772152 \| \| Kirghiz (kir) \| 0.674766 \| 0.779698 \| 0.723447 \| \| Komi-Permyak (koi) \| 0.652830 \| 0.633700 \| 0.643123 \| \| Konkani (kok) \| 0.778865 \| 0.728938 \| 0.753075 \| \| Komi (kom) \| 0.737374 \| 0.572549 \| 0.644592 \| \| Korean (kor) \| 0.984615 \| 0.967603 \| 0.976035 \| \| Karachay-Balkar (krc) \| 0.869416 \| 0.857627 \| 0.863481 \| \| Ripuarisch (ksh) \| 0.709859 \| 0.649485 \| 0.678331 \| \| Kurdish (kur) \| 0.883777 \| 0.862884 \| 0.873206 \| \| Ladino (lad) \| 0.660920 \| 0.576441 \| 0.615797 \| \| Lao (lao) \| 0.986175 \| 0.918455 \| 0.951111 \| \| Latin (lat) \| 0.581250 \| 0.636986 \| 0.607843 \| \| Latvian (lav) \| 0.824513 \| 0.797844 \| 0.810959 \| \| Lezghian (lez) \| 0.898955 \| 0.793846 \| 0.843137 \| \| Ligurian (lij) \| 0.662903 \| 0.677100 \| 0.669927 \| \| Limburgan (lim) \| 0.615385 \| 0.581818 \| 0.598131 \| \| Lingala (lin) \| 0.836207 \| 0.763780 \| 0.798354 \| \| Lithuanian (lit) \| 0.756329 \| 0.804714 \| 0.779772 \| \| Lombard (lmo) \| 0.556818 \| 0.536986 \| 0.546722 \| \| Northern Luri (lrc) \| 0.838574 \| 0.753296 \| 0.793651 \| \| Latgalian (ltg) \| 0.759531 \| 0.755102 \| 0.757310 \| \| Luxembourgish (ltz) \| 0.645062 \| 0.614706 \| 0.629518 \| \| Luganda (lug) \| 0.787535 \| 0.805797 \| 0.796562 \| \| Literary Chinese (lzh) \| 0.921951 \| 0.949749 \| 0.935644 \| \| Maithili (mai) \| 0.777778 \| 0.761658 \| 0.769634 \| \| Malayalam (mal) \| 0.993377 \| 0.949367 \| 0.970874 \| \| Banyumasan (map-bms) \| 0.531429 \| 0.453659 \| 0.489474 \| \| Marathi (mar) \| 0.748744 \| 0.818681 \| 0.782152 \| \| Moksha (mdf) \| 0.728745 \| 0.800000 \| 0.762712 \| \| Eastern Mari (mhr) \| 0.790323 \| 0.760870 \| 0.775316 \| \| Minangkabau (min) \| 0.953271 \| 0.886957 \| 0.918919 \| \| Macedonian (mkd) \| 0.816399 \| 0.849722 \| 0.832727 \| \| Malagasy (mlg) \| 0.925187 \| 0.918317 \| 0.921739 \| \| Maltese (mlt) \| 0.869421 \| 0.890017 \| 0.879599 \| \| Min Nan Chinese (nan) \| 0.743707 \| 0.820707 \| 0.780312 \| \| Mongolian (mon) \| 0.852194 \| 0.838636 \| 0.845361 \| \| Maori (mri) \| 0.934726 \| 0.937173 \| 0.935948 \| \| Western Mari (mrj) \| 0.818792 \| 0.827119 \| 0.822934 \| \| Malay (msa) \| 0.508065 \| 0.376119 \| 0.432247 \| \| Mirandese (mwl) \| 0.650407 \| 0.685225 \| 0.667362 \| \| Burmese (mya) \| 0.995968 \| 0.972441 \| 0.984064 \| \| Erzya (myv) \| 0.475783 \| 0.503012 \| 0.489019 \| \| Mazanderani (mzn) \| 0.775362 \| 0.701639 \| 0.736661 \| \| Neapolitan (nap) \| 0.628993 \| 0.595349 \| 0.611708 \| \| Navajo (nav) \| 0.955882 \| 0.937500 \| 0.946602 \| \| Classical Nahuatl (nci) \| 0.679758 \| 0.589005 \| 0.631136 \| \| Low German (nds) \| 0.669789 \| 0.690821 \| 0.680143 \| \| West Low German (nds-nl) \| 0.513889 \| 0.504545 \| 0.509174 \| \| Nepali (macrolanguage) (nep) \| 0.640476 \| 0.649758 \| 0.645084 \| \| Newari (new) \| 0.928571 \| 0.745902 \| 0.827273 \| \| Dutch (nld) \| 0.553763 \| 0.553763 \| 0.553763 \| \| Norwegian Nynorsk (nno) \| 0.569277 \| 0.519231 \| 0.543103 \| \| Bokmål (nob) \| 0.519856 \| 0.562500 \| 0.540338 \| \| Narom (nrm) \| 0.691275 \| 0.605882 \| 0.645768 \| \| Northern Sotho (nso) \| 0.950276 \| 0.815166 \| 0.877551 \| \| Occitan (oci) \| 0.483444 \| 0.366834 \| 0.417143 \| \| Livvi-Karelian (olo) \| 0.816850 \| 0.790780 \| 0.803604 \| \| Oriya (ori) \| 0.981481 \| 0.963636 \| 0.972477 \| \| Oromo (orm) \| 0.885714 \| 0.829218 \| 0.856536 \| \| Ossetian (oss) \| 0.822006 \| 0.855219 \| 0.838284 \| \| Pangasinan (pag) \| 0.842105 \| 0.715655 \| 0.773748 \| \| Pampanga (pam) \| 0.770000 \| 0.435028 \| 0.555957 \| \| Panjabi (pan) \| 0.996154 \| 0.984791 \| 0.990440 \| \| Papiamento (pap) \| 0.674672 \| 0.661670 \| 0.668108 \| \| Picard (pcd) \| 0.407895 \| 0.356322 \| 0.380368 \| \| Pennsylvania German (pdc) \| 0.487047 \| 0.509485 \| 0.498013 \| \| Palatine German (pfl) \| 0.614173 \| 0.570732 \| 0.591656 \| \| Western Panjabi (pnb) \| 0.926267 \| 0.887417 \| 0.906426 \| \| Polish (pol) \| 0.797059 \| 0.734417 \| 0.764457 \| \| Portuguese (por) \| 0.500914 \| 0.586724 \| 0.540434 \| \| Pushto (pus) \| 0.941489 \| 0.898477 \| 0.919481 \| \| Quechua (que) \| 0.854167 \| 0.797665 \| 0.824950 \| \| Tarantino dialect (roa-tara) \| 0.669794 \| 0.724138 \| 0.695906 \| \| Romansh (roh) \| 0.745527 \| 0.760649 \| 0.753012 \| \| Romanian (ron) \| 0.805486 \| 0.769048 \| 0.786845 \| \| Rusyn (rue) \| 0.718543 \| 0.645833 \| 0.680251 \| \| Aromanian (rup) \| 0.288482 \| 0.730245 \| 0.413580 \| \| Russian (rus) \| 0.530120 \| 0.690583 \| 0.599805 \| \| Yakut (sah) \| 0.853521 \| 0.865714 \| 0.859574 \| \| Sanskrit (san) \| 0.931343 \| 0.896552 \| 0.913616 \| \| Sicilian (scn) \| 0.734139 \| 0.618321 \| 0.671271 \| \| Scots (sco) \| 0.571429 \| 0.540816 \| 0.555701 \| \| Samogitian (sgs) \| 0.829167 \| 0.748120 \| 0.786561 \| \| Sinhala (sin) \| 0.909474 \| 0.935065 \| 0.922092 \| \| Slovak (slk) \| 0.738235 \| 0.665782 \| 0.700139 \| \| Slovene (slv) \| 0.671123 \| 0.662269 \| 0.666667 \| \| Northern Sami (sme) \| 0.800676 \| 0.825784 \| 0.813036 \| \| Shona (sna) \| 0.761702 \| 0.724696 \| 0.742739 \| \| Sindhi (snd) \| 0.950172 \| 0.946918 \| 0.948542 \| \| Somali (som) \| 0.849462 \| 0.802030 \| 0.825065 \| \| Spanish (spa) \| 0.325234 \| 0.413302 \| 0.364017 \| \| Albanian (sqi) \| 0.875899 \| 0.832479 \| 0.853637 \| \| Sardinian (srd) \| 0.750000 \| 0.711061 \| 0.730012 \| \| Sranan (srn) \| 0.888889 \| 0.771084 \| 0.825806 \| \| Serbian (srp) \| 0.824561 \| 0.814356 \| 0.819427 \| \| Saterfriesisch (stq) \| 0.790087 \| 0.734417 \| 0.761236 \| \| Sundanese (sun) \| 0.764192 \| 0.631769 \| 0.691700 \| \| Swahili (macrolanguage) (swa) \| 0.763496 \| 0.796247 \| 0.779528 \| \| Swedish (swe) \| 0.838284 \| 0.723647 \| 0.776758 \| \| Silesian (szl) \| 0.819788 \| 0.750809 \| 0.783784 \| \| Tamil (tam) \| 0.985765 \| 0.955172 \| 0.970228 \| \| Tatar (tat) \| 0.469780 \| 0.795349 \| 0.590674 \| \| Tulu (tcy) \| 0.893300 \| 0.873786 \| 0.883436 \| \| Telugu (tel) \| 1.000000 \| 0.913690 \| 0.954899 \| \| Tetum (tet) \| 0.765116 \| 0.744344 \| 0.754587 \| \| Tajik (tgk) \| 0.828418 \| 0.813158 \| 0.820717 \| \| Tagalog (tgl) \| 0.751468 \| 0.757396 \| 0.754420 \| \| Thai (tha) \| 0.933884 \| 0.807143 \| 0.865900 \| \| Tongan (ton) \| 0.920245 \| 0.923077 \| 0.921659 \| \| Tswana (tsn) \| 0.873397 \| 0.889070 \| 0.881164 \| \| Turkmen (tuk) \| 0.898438 \| 0.837887 \| 0.867107 \| \| Turkish (tur) \| 0.666667 \| 0.716981 \| 0.690909 \| \| Tuvan (tyv) \| 0.857143 \| 0.805063 \| 0.830287 \| \| Udmurt (udm) \| 0.865517 \| 0.756024 \| 0.807074 \| \| Uighur (uig) \| 0.991597 \| 0.967213 \| 0.979253 \| \| Ukrainian (ukr) \| 0.771341 \| 0.702778 \| 0.735465 \| \| Urdu (urd) \| 0.877647 \| 0.855505 \| 0.866434 \| \| Uzbek (uzb) \| 0.655652 \| 0.797040 \| 0.719466 \| \| Venetian (vec) \| 0.611111 \| 0.527233 \| 0.566082 \| \| Veps (vep) \| 0.672862 \| 0.688213 \| 0.680451 \| \| Vietnamese (vie) \| 0.932406 \| 0.914230 \| 0.923228 \| \| Vlaams (vls) \| 0.594427 \| 0.501305 \| 0.543909 \| \| Volapük (vol) \| 0.765625 \| 0.942308 \| 0.844828 \| \| Võro (vro) \| 0.797203 \| 0.740260 \| 0.767677 \| \| Waray (war) \| 0.930876 \| 0.930876 \| 0.930876 \| \| Walloon (wln) \| 0.636804 \| 0.693931 \| 0.664141 \| \| Wolof (wol) \| 0.864220 \| 0.845601 \| 0.854809 \| \| Wu Chinese (wuu) \| 0.848921 \| 0.830986 \| 0.839858 \| \| Xhosa (xho) \| 0.837398 \| 0.759214 \| 0.796392 \| \| Mingrelian (xmf) \| 0.943396 \| 0.874126 \| 0.907441 \| \| Yiddish (yid) \| 0.955729 \| 0.897311 \| 0.925599 \| \| Yoruba (yor) \| 0.812010 \| 0.719907 \| 0.763190 \| \| Zeeuws (zea) \| 0.617737 \| 0.550409 \| 0.582133 \| \| Cantonese (zh-yue) \| 0.859649 \| 0.649007 \| 0.739623 \| \| Standard Chinese (zho) \| 0.845528 \| 0.781955 \| 0.812500 \| \| accuracy \| 0.749527 \| 0.749527 \| 0.749527 \| \| macro avg \| 0.762866 \| 0.742101 \| 0.749261 \| \| weighted avg \| 0.762006 \| 0.749527 \| 0.752910 \| ## Questions? Post a Github issue from [HERE](https://github.com/m3hrdadfi/zabanshenas/issues).
manishiitg/output	5ddc5ec9f5bf098c3e9de99c27773112ce34c510	2021-05-20T17:44:39.000Z	[ "pytorch", "jax", "roberta", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	manishiitg	null	manishiitg/output	15	null	transformers	9,539	Entry not found
mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili	ab8ad3c070d1fea9b697c2870129bca2d4a6f760	2021-11-25T09:04:07.000Z	[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]	token-classification	false	mbeukman	null	mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili	15	null	transformers	9,540	--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-kinyarwanda](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation\|Description -\|- O\|Outside of a named entity B-DATE \|Beginning of a DATE entity right after another DATE entity I-DATE \|DATE entity B-PER \|Beginning of a person’s name right after another person’s name I-PER \|Person’s name B-ORG \|Beginning of an organisation right after another organisation I-ORG \|Organisation B-LOC \|Beginning of a location right after another location I-LOC \|Location \| Model Name \| Staring point \| Evaluation / Fine-tune Language \| F1 \| Precision \| Recall \| F1 (DATE) \| F1 (LOC) \| F1 (ORG) \| F1 (PER) \| \| -------------------------------------------------- \| -------------------- \| -------------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| \| [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) (This model) \| [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) \| [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) \| swa \| 88.36 \| 86.95 \| 89.82 \| 86.00 \| 91.00 \| 77.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) \| [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) \| swa \| 87.75 \| 86.55 \| 88.97 \| 85.00 \| 92.00 \| 77.00 \| 91.00 \| \| [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) \| [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) \| swa \| 88.93 \| 87.64 \| 90.25 \| 83.00 \| 92.00 \| 79.00 \| 95.00 \| \| [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) \| [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) \| swa \| 87.93 \| 86.91 \| 88.97 \| 83.00 \| 91.00 \| 76.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) \| [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) \| [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) \| swa \| 90.36 \| 88.59 \| 92.20 \| 86.00 \| 93.00 \| 79.00 \| 96.00 \| \| [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) \| [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) \| swa \| 87.80 \| 86.50 \| 89.14 \| 86.00 \| 90.00 \| 78.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) \| [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) \| swa \| 87.73 \| 86.67 \| 88.80 \| 85.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) \| [base](https://huggingface.co/xlm-roberta-base) \| swa \| 88.71 \| 86.84 \| 90.67 \| 83.00 \| 91.00 \| 79.00 \| 95.00 \| ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```
mbeukman/xlm-roberta-base-finetuned-ner-swahili	d09c7418ffdc293886fb957cc014eadacd60718b	2021-11-25T09:04:40.000Z	[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]	token-classification	false	mbeukman	null	mbeukman/xlm-roberta-base-finetuned-ner-swahili	15	1	transformers	9,541	--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation\|Description -\|- O\|Outside of a named entity B-DATE \|Beginning of a DATE entity right after another DATE entity I-DATE \|DATE entity B-PER \|Beginning of a person’s name right after another person’s name I-PER \|Person’s name B-ORG \|Beginning of an organisation right after another organisation I-ORG \|Organisation B-LOC \|Beginning of a location right after another location I-LOC \|Location \| Model Name \| Staring point \| Evaluation / Fine-tune Language \| F1 \| Precision \| Recall \| F1 (DATE) \| F1 (LOC) \| F1 (ORG) \| F1 (PER) \| \| -------------------------------------------------- \| -------------------- \| -------------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| \| [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) (This model) \| [base](https://huggingface.co/xlm-roberta-base) \| swa \| 88.71 \| 86.84 \| 90.67 \| 83.00 \| 91.00 \| 79.00 \| 95.00 \| \| [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) \| [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) \| swa \| 88.36 \| 86.95 \| 89.82 \| 86.00 \| 91.00 \| 77.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) \| [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) \| swa \| 87.75 \| 86.55 \| 88.97 \| 85.00 \| 92.00 \| 77.00 \| 91.00 \| \| [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) \| [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) \| [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) \| swa \| 88.93 \| 87.64 \| 90.25 \| 83.00 \| 92.00 \| 79.00 \| 95.00 \| \| [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) \| [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) \| swa \| 87.93 \| 86.91 \| 88.97 \| 83.00 \| 91.00 \| 76.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) \| [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) \| [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) \| swa \| 90.36 \| 88.59 \| 92.20 \| 86.00 \| 93.00 \| 79.00 \| 96.00 \| \| [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) \| [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) \| swa \| 87.80 \| 86.50 \| 89.14 \| 86.00 \| 90.00 \| 78.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) \| [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) \| swa \| 87.73 \| 86.67 \| 88.80 \| 85.00 \| 91.00 \| 75.00 \| 93.00 \| ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```
mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo	728511ffd5d6856f5d75db779b2719d53ab75f07	2021-11-25T09:04:58.000Z	[ "pytorch", "xlm-roberta", "token-classification", "luo", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]	token-classification	false	mbeukman	null	mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo	15	null	transformers	9,542	--- language: - luo tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Jii 2 moko jowito ngimagi ka machielo 1 to ohinyore marach mokalo e masira makoch mar apaya mane otimore e apaya mawuok Oyugis kochimo Chabera e sub county ma Rachuonyo East e County ma Homa Bay ewii odhiambo makawuononi" --- # xlm-roberta-base-finetuned-swahili-finetuned-ner-luo This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-swahili](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Luo part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation\|Description -\|- O\|Outside of a named entity B-DATE \|Beginning of a DATE entity right after another DATE entity I-DATE \|DATE entity B-PER \|Beginning of a person’s name right after another person’s name I-PER \|Person’s name B-ORG \|Beginning of an organisation right after another organisation I-ORG \|Organisation B-LOC \|Beginning of a location right after another location I-LOC \|Location \| Model Name \| Staring point \| Evaluation / Fine-tune Language \| F1 \| Precision \| Recall \| F1 (DATE) \| F1 (LOC) \| F1 (ORG) \| F1 (PER) \| \| -------------------------------------------------- \| -------------------- \| -------------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| \| [xlm-roberta-base-finetuned-swahili-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo) (This model) \| [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) \| luo \| 78.13 \| 77.75 \| 78.52 \| 65.00 \| 82.00 \| 61.00 \| 89.00 \| \| [xlm-roberta-base-finetuned-luo-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-luo) \| [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) \| luo \| 78.71 \| 78.91 \| 78.52 \| 72.00 \| 84.00 \| 59.00 \| 87.00 \| \| [xlm-roberta-base-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-luo) \| [base](https://huggingface.co/xlm-roberta-base) \| luo \| 75.99 \| 76.18 \| 75.80 \| 71.00 \| 76.00 \| 62.00 \| 85.00 \| ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Jii 2 moko jowito ngimagi ka machielo 1 to ohinyore marach mokalo e masira makoch mar apaya mane otimore e apaya mawuok Oyugis kochimo Chabera e sub county ma Rachuonyo East e County ma Homa Bay ewii odhiambo makawuononi" ner_results = nlp(example) print(ner_results) ```
mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili	80a0b4230b7b706e50ce5038e9f18b21f44c1198	2021-11-25T09:05:15.000Z	[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]	token-classification	false	mbeukman	null	mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili	15	null	transformers	9,543	--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-yoruba](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation\|Description -\|- O\|Outside of a named entity B-DATE \|Beginning of a DATE entity right after another DATE entity I-DATE \|DATE entity B-PER \|Beginning of a person’s name right after another person’s name I-PER \|Person’s name B-ORG \|Beginning of an organisation right after another organisation I-ORG \|Organisation B-LOC \|Beginning of a location right after another location I-LOC \|Location \| Model Name \| Staring point \| Evaluation / Fine-tune Language \| F1 \| Precision \| Recall \| F1 (DATE) \| F1 (LOC) \| F1 (ORG) \| F1 (PER) \| \| -------------------------------------------------- \| -------------------- \| -------------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| -------------- \| \| [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) (This model) \| [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) \| swa \| 87.73 \| 86.67 \| 88.80 \| 85.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) \| [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) \| swa \| 88.36 \| 86.95 \| 89.82 \| 86.00 \| 91.00 \| 77.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) \| [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) \| swa \| 87.75 \| 86.55 \| 88.97 \| 85.00 \| 92.00 \| 77.00 \| 91.00 \| \| [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) \| [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) \| [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) \| swa \| 88.93 \| 87.64 \| 90.25 \| 83.00 \| 92.00 \| 79.00 \| 95.00 \| \| [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) \| [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) \| swa \| 87.93 \| 86.91 \| 88.97 \| 83.00 \| 91.00 \| 76.00 \| 94.00 \| \| [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) \| [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) \| swa \| 87.26 \| 85.15 \| 89.48 \| 83.00 \| 91.00 \| 75.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) \| [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) \| swa \| 90.36 \| 88.59 \| 92.20 \| 86.00 \| 93.00 \| 79.00 \| 96.00 \| \| [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) \| [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) \| swa \| 87.80 \| 86.50 \| 89.14 \| 86.00 \| 90.00 \| 78.00 \| 93.00 \| \| [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) \| [base](https://huggingface.co/xlm-roberta-base) \| swa \| 88.71 \| 86.84 \| 90.67 \| 83.00 \| 91.00 \| 79.00 \| 95.00 \| ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```
meghanabhange/hinglish-sentence-bert	f7e2387d18a11062ab0ba7eb2919ee70d41d3795	2021-05-19T23:17:18.000Z	[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	meghanabhange	null	meghanabhange/hinglish-sentence-bert	15	null	transformers	9,544	Entry not found
midas/gupshup_e2e_pegasus	4ddf0da7354c31ae27cdce2436ba6b87c6d21537	2021-11-14T02:07:37.000Z	[ "pytorch", "pegasus", "text2text-generation", "arxiv:1910.04073", "transformers", "autotrain_compatible" ]	text2text-generation	false	midas	null	midas/gupshup_e2e_pegasus	15	null	transformers	9,545	# Gupshup GupShup: Summarizing Open-Domain Code-Switched Conversations EMNLP 2021 Paper: [https://aclanthology.org/2021.emnlp-main.499.pdf](https://aclanthology.org/2021.emnlp-main.499.pdf) Github: [https://github.com/midas-research/gupshup](https://github.com/midas-research/gupshup) ### Dataset Please request for the Gupshup data using [this Google form](https://docs.google.com/forms/d/1zvUk7WcldVF3RCoHdWzQPzPprtSJClrnHoIOYbzaJEI/edit?ts=61381ec0). Dataset is available for `Hinglish Dilaogues to English Summarization`(h2e) and `English Dialogues to English Summarization`(e2e). For each task, Dialogues/conversastion have `.source`(train.source) as file extension whereas Summary has `.target`(train.target) file extension. ".source" file need to be provided to `input_path` and ".target" file to `reference_path` argument in the scripts. ## Models All model weights are available on the Huggingface model hub. Users can either directly download these weights in their local and provide this path to `model_name` argument in the scripts or use the provided alias (to `model_name` argument) in scripts directly; this will lead to download weights automatically by scripts. Model names were aliased in "gupshup_TASK_MODEL" sense, where "TASK" can be h2e,e2e and MODEL can be mbart, pegasus, etc., as listed below. 1. Hinglish Dialogues to English Summary (h2e) \| Model \| Huggingface Alias \| \|---------\|-------------------------------------------------------------------------------\| \| mBART \| [midas/gupshup_h2e_mbart](https://huggingface.co/midas/gupshup_h2e_mbart) \| \| PEGASUS \| [midas/gupshup_h2e_pegasus](https://huggingface.co/midas/gupshup_h2e_pegasus) \| \| T5 MTL \| [midas/gupshup_h2e_t5_mtl](https://huggingface.co/midas/gupshup_h2e_t5_mtl) \| \| T5 \| [midas/gupshup_h2e_t5](https://huggingface.co/midas/gupshup_h2e_t5) \| \| BART \| [midas/gupshup_h2e_bart](https://huggingface.co/midas/gupshup_h2e_bart) \| \| GPT-2 \| [midas/gupshup_h2e_gpt](https://huggingface.co/midas/gupshup_h2e_gpt) \| 2. English Dialogues to English Summary (e2e) \| Model \| Huggingface Alias \| \|---------\|-------------------------------------------------------------------------------\| \| mBART \| [midas/gupshup_e2e_mbart](https://huggingface.co/midas/gupshup_e2e_mbart) \| \| PEGASUS \| [midas/gupshup_e2e_pegasus](https://huggingface.co/midas/gupshup_e2e_pegasus) \| \| T5 MTL \| [midas/gupshup_e2e_t5_mtl](https://huggingface.co/midas/gupshup_e2e_t5_mtl) \| \| T5 \| [midas/gupshup_e2e_t5](https://huggingface.co/midas/gupshup_e2e_t5) \| \| BART \| [midas/gupshup_e2e_bart](https://huggingface.co/midas/gupshup_e2e_bart) \| \| GPT-2 \| [midas/gupshup_e2e_gpt](https://huggingface.co/midas/gupshup_e2e_gpt) \| ## Inference ### Using command line 1. Clone this repo and create a python virtual environment (https://docs.python.org/3/library/venv.html). Install the required packages using ``` git clone https://github.com/midas-research/gupshup.git pip install -r requirements.txt ``` 2. run_eval script has the following arguments. * model_name : Path or alias to one of our models available on Huggingface as listed above. * input_path : Source file or path to file containing conversations, which will be summarized. * save_path : File path where to save summaries generated by the model. * reference_path : Target file or path to file containing summaries, used to calculate matrices. * score_path : File path where to save scores. * bs : Batch size * device: Cuda devices to use. Please make sure you have downloaded the Gupshup dataset using the above google form and provide the correct path to these files in the argument's `input_path` and `refrence_path.` Or you can simply put `test.source` and `test.target` in `data/h2e/`(hinglish to english) or `data/e2e/`(english to english) folder. For example, to generate English summaries from Hinglish dialogues using the mbart model, run the following command ``` python run_eval.py \ --model_name midas/gupshup_h2e_mbart \ --input_path data/h2e/test.source \ --save_path generated_summary.txt \ --reference_path data/h2e/test.target \ --score_path scores.txt \ --bs 8 ``` Another example, to generate English summaries from English dialogues using the Pegasus model ``` python run_eval.py \ --model_name midas/gupshup_e2e_pegasus \ --input_path data/e2e/test.source \ --save_path generated_summary.txt \ --reference_path data/e2e/test.target \ --score_path scores.txt \ --bs 8 ``` Please create an issue if you are facing any difficulties in replicating the results. ### References Please cite [[1]](https://arxiv.org/abs/1910.04073) if you found the resources in this repository useful. [1] Mehnaz, Laiba, Debanjan Mahata, Rakesh Gosangi, Uma Sushmitha Gunturi, Riya Jain, Gauri Gupta, Amardeep Kumar, Isabelle G. Lee, Anish Acharya, and Rajiv Shah. [GupShup: Summarizing Open-Domain Code-Switched Conversations](https://aclanthology.org/2021.emnlp-main.499.pdf) ``` @inproceedings{mehnaz2021gupshup, title={GupShup: Summarizing Open-Domain Code-Switched Conversations}, author={Mehnaz, Laiba and Mahata, Debanjan and Gosangi, Rakesh and Gunturi, Uma Sushmitha and Jain, Riya and Gupta, Gauri and Kumar, Amardeep and Lee, Isabelle G and Acharya, Anish and Shah, Rajiv}, booktitle={Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing}, pages={6177--6192}, year={2021} } ```
mmcquade11/reviews-sentiment-analysis	312c90a7ffe06f57619b45485e136e2c22c973b1	2021-12-01T18:52:49.000Z	[ "pytorch", "distilbert", "text-classification", "transformers" ]	text-classification	false	mmcquade11	null	mmcquade11/reviews-sentiment-analysis	15	null	transformers	9,546	Entry not found
mofawzy/bert-ajgt	2b31eb1e999581c5ccc7cf3c7d47ca3d2f9f50a0	2022-02-17T19:56:26.000Z	[ "pytorch", "bert", "text-classification", "ar", "dataset:AJGT", "transformers", "AJGT" ]	text-classification	false	mofawzy	null	mofawzy/bert-ajgt	15	null	transformers	9,547	--- language: - ar datasets: - AJGT tags: - AJGT widget: - text: "يهدي الله من يشاء" - text: "الاسلوب قذر وقمامه" --- # BERT-AJGT Arabic version bert model fine tuned on AJGT dataset ## Data The model were fine-tuned on ~1800 sentence from twitter for Jordanian dialect. ## Results \| class \| precision \| recall \| f1-score \| Support \| \|----------\|-----------\|--------\|----------\|---------\| \| 0 \| 0.9462 \| 0.9778 \| 0.9617 \| 90 \| \| 1 \| 0.9399 \| 0.9689 \| 0.9542 \| 90 \| \| Accuracy \| \| \| 0.9611 \| 180 \| ## How to use You can use these models by installing `torch` or `tensorflow` and Huggingface library `transformers`. And you can use it directly by initializing it like this: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer model_name="mofawzy/bert-ajgt" model = AutoModelForSequenceClassification.from_pretrained(model_name,num_labels=2) tokenizer = AutoTokenizer.from_pretrained(model_name) ```
monologg/koelectra-small-generator	a7d050043fdb98b63ebcb747df7004b5b94dc3b8	2020-12-26T16:23:42.000Z	[ "pytorch", "electra", "fill-mask", "ko", "transformers", "autotrain_compatible" ]	fill-mask	false	monologg	null	monologg/koelectra-small-generator	15	null	transformers	9,548	--- language: ko --- # KoELECTRA (Small Generator) Pretrained ELECTRA Language Model for Korean (`koelectra-small-generator`) For more detail, please see [original repository](https://github.com/monologg/KoELECTRA/blob/master/README_EN.md). ## Usage ### Load model and tokenizer ```python >>> from transformers import ElectraModel, ElectraTokenizer >>> model = ElectraModel.from_pretrained("monologg/koelectra-small-generator") >>> tokenizer = ElectraTokenizer.from_pretrained("monologg/koelectra-small-generator") ``` ### Tokenizer example ```python >>> from transformers import ElectraTokenizer >>> tokenizer = ElectraTokenizer.from_pretrained("monologg/koelectra-small-generator") >>> tokenizer.tokenize("[CLS] 한국어 ELECTRA를 공유합니다. [SEP]") ['[CLS]', '한국어', 'E', '##L', '##EC', '##T', '##RA', '##를', '공유', '##합니다', '.', '[SEP]'] >>> tokenizer.convert_tokens_to_ids(['[CLS]', '한국어', 'E', '##L', '##EC', '##T', '##RA', '##를', '공유', '##합니다', '.', '[SEP]']) [2, 18429, 41, 6240, 15229, 6204, 20894, 5689, 12622, 10690, 18, 3] ``` ## Example using ElectraForMaskedLM ```python from transformers import pipeline fill_mask = pipeline( "fill-mask", model="monologg/koelectra-small-generator", tokenizer="monologg/koelectra-small-generator" ) print(fill_mask("나는 {} 밥을 먹었다.".format(fill_mask.tokenizer.mask_token))) ```
monsoon-nlp/es-seq2seq-gender-decoder	e271163d2c94e74e2ebba5315c4c0d1e7e598ac2	2021-05-20T00:09:13.000Z	[ "pytorch", "bert", "text-generation", "es", "transformers" ]	text-generation	false	monsoon-nlp	null	monsoon-nlp/es-seq2seq-gender-decoder	15	null	transformers	9,549	--- language: es --- # es-seq2seq-gender (decoder) This is a seq2seq model (decoder half) to "flip" gender in Spanish sentences. The model can augment your existing Spanish data, or generate counterfactuals to test a model's decisions (would changing the gender of the subject or speaker change output?). Intended Examples: - el profesor viejo => la profesora vieja (article, noun, adjective all flip) - una actriz => un actor (irregular noun) - el lingüista => la lingüista (irregular noun) - la biblioteca => la biblioteca (no person, no flip) People's names are unchanged in this version, but you can use packages such as https://pypi.org/project/gender-guesser/ ## Sample code https://colab.research.google.com/drive/1Ta_YkXx93FyxqEu_zJ-W23PjPumMNHe5 ``` import torch from transformers import AutoTokenizer, EncoderDecoderModel model = EncoderDecoderModel.from_encoder_decoder_pretrained("monsoon-nlp/es-seq2seq-gender-encoder", "monsoon-nlp/es-seq2seq-gender-decoder") tokenizer = AutoTokenizer.from_pretrained('monsoon-nlp/es-seq2seq-gender-decoder') # all are same as BETO uncased original input_ids = torch.tensor(tokenizer.encode("la profesora vieja")).unsqueeze(0) generated = model.generate(input_ids, decoder_start_token_id=model.config.decoder.pad_token_id) tokenizer.decode(generated.tolist()[0]) > '[PAD] el profesor viejo profesor viejo profesor...' ``` ## Training I originally developed <a href="https://github.com/MonsoonNLP/el-la">a gender flip Python script</a> with <a href="https://huggingface.co/dccuchile/bert-base-spanish-wwm-uncased">BETO</a>, the Spanish-language BERT from Universidad de Chile, and spaCy to parse dependencies in sentences. More about this project: https://medium.com/ai-in-plain-english/gender-bias-in-spanish-bert-1f4d76780617 The seq2seq model is trained on gender-flipped text from that script run on the <a href="https://huggingface.co/datasets/muchocine">muchocine dataset</a>, and the first 6,853 lines from the <a href="https://oscar-corpus.com/">OSCAR corpus</a> (Spanish ded-duped). The encoder and decoder started with weights and vocabulary from BETO (uncased). ## Non-binary gender This model is useful to generate male and female text samples, but falls short of capturing gender diversity in the world and in the Spanish language. Some communities prefer the plural -@s to represent -os and -as, or -e and -es for gender-neutral or mixed-gender plural, or use fewer gendered professional nouns (la juez and not jueza). This is not yet embraced by the Royal Spanish Academy and is not represented in the corpora and tokenizers used to build this project. This seq2seq project and script could, in the future, help generate more text samples and prepare NLP models to understand us all better. #### Sources - https://www.nytimes.com/2020/04/15/world/americas/argentina-gender-language.html - https://www.washingtonpost.com/dc-md-va/2019/12/05/teens-argentina-are-leading-charge-gender-neutral-language/?arc404=true - https://www.theguardian.com/world/2020/jan/19/gender-neutral-language-battle-spain - https://es.wikipedia.org/wiki/Lenguaje_no_sexista - https://remezcla.com/culture/argentine-company-re-imagines-little-prince-gender-neutral-language/
monsoon-nlp/gpt-nyc	b49baf5fe2c2c03d309cbf681fd630ff15c564b9	2021-05-23T10:03:21.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "transformers" ]	text-generation	false	monsoon-nlp	null	monsoon-nlp/gpt-nyc	15	1	transformers	9,550	# GPT-NYC ## About GPT2-Medium fine-tuned on questions and responses from https://reddit.com/r/asknyc I filtered comments to ones with scores >= 3, and responding directly to the original post ( = ignoring responses to other commenters). I added tokens to match NYC neighborhoods, subway stations, foods, and other common terms in the original batches of questions and comments. You would be surprised what is missing from GPT tokens! Try prompting with ```question? %% ``` or ```question? - more info %%``` ## Status I would like to continue by: - fine-tuning GPT2-Large with a larger dataset of questions - examining bias and toxicity - examining memorization vs. original responses - releasing a reusable benchmark ## Blog https://mapmeld.medium.com/gpt-nyc-part-1-9cb698b2e3d ## Notebooks ### Data processing / new tokens https://colab.research.google.com/drive/13BOw0uekoAYB4jjQtaXTn6J_VHatiRLu ### Fine-tuning GPT2 (small) https://colab.research.google.com/drive/1FnXcAh4H-k8dAzixkV5ieygV96ePh3lR ### Fine-tuning GPT2-Medium Same code as small, but on Google Cloud to use an A100 GPU ### Predictive text and probabilities Scroll to end of https://colab.research.google.com/drive/1FnXcAh4H-k8dAzixkV5ieygV96ePh3lR to see how to install git-lfs and trick ecco into loading this.
mrm8488/GPT-2-finetuned-CRD3	ce6e1457142a7aa61c564e5e32364f40f8cd3201	2021-05-23T10:10:58.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "transformers" ]	text-generation	false	mrm8488	null	mrm8488/GPT-2-finetuned-CRD3	15	null	transformers	9,551	Entry not found
mrm8488/spanbert-base-finetuned-squadv1	54aff17ce703bd24116984d9abbd019c06253159	2021-05-20T00:49:33.000Z	[ "pytorch", "jax", "bert", "en", "arxiv:1907.10529", "transformers" ]	null	false	mrm8488	null	mrm8488/spanbert-base-finetuned-squadv1	15	null	transformers	9,552	--- language: en thumbnail: --- # SpanBERT base fine-tuned on SQuAD v1 [SpanBERT](https://github.com/facebookresearch/SpanBERT) created by [Facebook Research](https://github.com/facebookresearch) and fine-tuned on [SQuAD 1.1](https://rajpurkar.github.io/SQuAD-explorer/explore/1.1/dev/) for Q&A downstream task ([by them](https://github.com/facebookresearch/SpanBERT#finetuned-models-squad-1120-relation-extraction-coreference-resolution)). ## Details of SpanBERT [SpanBERT: Improving Pre-training by Representing and Predicting Spans](https://arxiv.org/abs/1907.10529) ## Details of the downstream task (Q&A) - Dataset 📚 🧐 ❓ [SQuAD1.1](https://rajpurkar.github.io/SQuAD-explorer/) ## Model fine-tuning 🏋️‍ You can get the fine-tuning script [here](https://github.com/facebookresearch/SpanBERT) ```bash python code/run_squad.py \ --do_train \ --do_eval \ --model spanbert-base-cased \ --train_file train-v1.1.json \ --dev_file dev-v1.1.json \ --train_batch_size 32 \ --eval_batch_size 32 \ --learning_rate 2e-5 \ --num_train_epochs 4 \ --max_seq_length 512 \ --doc_stride 128 \ --eval_metric f1 \ --output_dir squad_output \ --fp16 ``` ## Results Comparison 📝 \| \| SQuAD 1.1 \| SQuAD 2.0 \| Coref \| TACRED \| \| ---------------------- \| ------------- \| --------- \| ------- \| ------ \| \| \| F1 \| F1 \| avg. F1 \| F1 \| \| BERT (base) \| 88.5 \| 76.5 \| 73.1 \| 67.7 \| \| SpanBERT (base) \| 92.4 (this one) \| [83.6](https://huggingface.co/mrm8488/spanbert-base-finetuned-squadv2) \| 77.4 \| [68.2](https://huggingface.co/mrm8488/spanbert-base-finetuned-tacred) \| \| BERT (large) \| 91.3 \| 83.3 \| 77.1 \| 66.4 \| \| SpanBERT (large) \| [94.6](https://huggingface.co/mrm8488/spanbert-large-finetuned-squadv1) \| [88.7](https://huggingface.co/mrm8488/spanbert-large-finetuned-squadv2) \| 79.6 \| [70.8](https://huggingface.co/mrm8488/spanbert-large-finetuned-tacred) \| Note: The numbers marked as * are evaluated on the development sets because those models were not submitted to the official SQuAD leaderboard. All the other numbers are test numbers. ## Model in action Fast usage with pipelines: ```python from transformers import pipeline qa_pipeline = pipeline( "question-answering", model="mrm8488/spanbert-base-finetuned-squadv1", tokenizer="SpanBERT/spanbert-base-cased" ) qa_pipeline({ 'context': "Manuel Romero has been working very hard in the repository hugginface/transformers lately", 'question': "How has been working Manuel Romero lately?" }) # Output: {'answer': 'very hard in the repository hugginface/transformers', 'end': 82, 'score': 0.327230326857725, 'start': 31} ``` > Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) > Made with <span style="color: #e25555;">&hearts;</span> in Spain
ncoop57/cm_codeparrot	e21e05b4fa4c25b96a6f18f8ff8097628257550d	2022-03-02T12:59:23.000Z	[ "pytorch", "gpt_neo", "text-generation", "transformers" ]	text-generation	false	ncoop57	null	ncoop57/cm_codeparrot	15	null	transformers	9,553	Entry not found
nielsr/dino_deits8	161957bc7e0712f3c5bd5490e62ae9c70678ae4c	2021-05-03T08:17:02.000Z	[ "pytorch", "vit", "feature-extraction", "transformers" ]	feature-extraction	false	nielsr	null	nielsr/dino_deits8	15	null	transformers	9,554	Entry not found
orzhan/t5-long-extract	9b20f3c309b8cbea80d2609bbd5a638d1c7a7385	2022-06-11T07:20:59.000Z	[ "pytorch", "t5", "feature-extraction", "transformers" ]	feature-extraction	false	orzhan	null	orzhan/t5-long-extract	15	null	transformers	9,555	T5-small model fine-tuned for extractive summarization on long documents. Repository: [GitHub](https://github.com/orzhan/t5-long-extract)
patrickvonplaten/hf-reformer-crime-and-punish	498eddd2421bd1902a147e550ffae000d7a60d55	2020-05-11T11:10:52.000Z	[ "pytorch", "reformer", "text-generation", "transformers" ]	text-generation	false	patrickvonplaten	null	patrickvonplaten/hf-reformer-crime-and-punish	15	null	transformers	9,556	Entry not found
patrickvonplaten/wav2vec2-common_voice-tr-demo	59074c9bbe282a9c02a6422082d16fcc4aa46307	2021-12-20T12:54:39.000Z	[ "pytorch", "tensorboard", "wav2vec2", "automatic-speech-recognition", "tr", "dataset:common_voice", "transformers", "speech-recognition", "common_voice", "generated_from_trainer", "license:apache-2.0", "model-index" ]	automatic-speech-recognition	false	patrickvonplaten	null	patrickvonplaten/wav2vec2-common_voice-tr-demo	15	null	transformers	9,557	--- language: - tr license: apache-2.0 tags: - speech-recognition - common_voice - generated_from_trainer datasets: - common_voice model-index: - name: wav2vec2-common_voice-tr-demo results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # wav2vec2-common_voice-tr-demo This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - TR dataset. It achieves the following results on the evaluation set: - Loss: 0.3856 - Wer: 0.3556 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 15.0 - mixed_precision_training: Native AMP ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Wer \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:------:\| \| 3.7391 \| 0.92 \| 100 \| 3.5760 \| 1.0 \| \| 2.927 \| 1.83 \| 200 \| 3.0796 \| 0.9999 \| \| 0.9009 \| 2.75 \| 300 \| 0.9278 \| 0.8226 \| \| 0.6529 \| 3.67 \| 400 \| 0.5926 \| 0.6367 \| \| 0.3623 \| 4.59 \| 500 \| 0.5372 \| 0.5692 \| \| 0.2888 \| 5.5 \| 600 \| 0.4407 \| 0.4838 \| \| 0.285 \| 6.42 \| 700 \| 0.4341 \| 0.4694 \| \| 0.0842 \| 7.34 \| 800 \| 0.4153 \| 0.4302 \| \| 0.1415 \| 8.26 \| 900 \| 0.4317 \| 0.4136 \| \| 0.1552 \| 9.17 \| 1000 \| 0.4145 \| 0.4013 \| \| 0.1184 \| 10.09 \| 1100 \| 0.4115 \| 0.3844 \| \| 0.0556 \| 11.01 \| 1200 \| 0.4182 \| 0.3862 \| \| 0.0851 \| 11.93 \| 1300 \| 0.3985 \| 0.3688 \| \| 0.0961 \| 12.84 \| 1400 \| 0.4030 \| 0.3665 \| \| 0.0596 \| 13.76 \| 1500 \| 0.3880 \| 0.3631 \| \| 0.0917 \| 14.68 \| 1600 \| 0.3878 \| 0.3582 \| ### Framework versions - Transformers 4.11.0.dev0 - Pytorch 1.9.0+cu111 - Datasets 1.12.1 - Tokenizers 0.10.3
peggyhuang/bert-base-uncased-coqa	ae54d05fa6b4c7b6c04a9cd28c1fd26bfd23d4fa	2021-11-19T09:05:00.000Z	[ "pytorch", "bert", "question-answering", "transformers", "autotrain_compatible" ]	question-answering	false	peggyhuang	null	peggyhuang/bert-base-uncased-coqa	15	null	transformers	9,558	Entry not found
pinecone/bert-retriever-squad2	edb4465f3fc105473bf54cb7d4674d0d043d9185	2022-01-03T02:38:02.000Z	[ "pytorch", "bert", "feature-extraction", "sentence-transformers", "sentence-similarity", "transformers" ]	sentence-similarity	false	pinecone	null	pinecone/bert-retriever-squad2	15	null	sentence-transformers	9,559	--- pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(*encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the Sentence Embeddings Benchmark: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: DataLoader: `sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 5429 with parameters: ``` {'batch_size': 24} ``` Loss*: `sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss` with parameters: ``` {'scale': 20.0, 'similarity_fct': 'cos_sim'} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'transformers.optimization.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 542, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
pkushiqiang/bert-title-org	113cf90e11c9920cc55c2a885813d30166991300	2022-02-28T06:16:37.000Z	[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	pkushiqiang	null	pkushiqiang/bert-title-org	15	null	transformers	9,560	Entry not found
pmthangk09/bert-base-uncased-superglue-multirc	5b503587b2e7cee15046475079d14a13aff1e616	2021-05-20T02:50:34.000Z	[ "pytorch", "tf", "jax", "bert", "text-classification", "transformers" ]	text-classification	false	pmthangk09	null	pmthangk09/bert-base-uncased-superglue-multirc	15	null	transformers	9,561	Entry not found
pszemraj/t5-base-askscience	a7582c22c55096ed39fe3261852d94d767e1da98	2022-02-19T22:50:18.000Z	[ "pytorch", "t5", "text2text-generation", "en", "dataset:eli5", "transformers", "qa", "askscience", "lfqa", "information retrieval", "autotrain_compatible" ]	text2text-generation	false	pszemraj	null	pszemraj/t5-base-askscience	15	null	transformers	9,562	--- language: - en tags: - t5 - qa - askscience - lfqa - information retrieval datasets: - eli5 metrics: - rouge widget: - text: "why aren't there more planets in our solar system?" example_title: "solar system" - text: "question: what is a probability distribution? context: I am just learning about statistics." example_title: "probability distribution" - text: "question: What are the underlying physical processes by which exercise helps us lose weight? context: I started working out two weeks ago and already feel a lot better, and started to think about it and became deeply confused." example_title: "pumpen" - text: "what is a neural network?" example_title: "deep learning" - text: "What are the primary mechanisms that computers use to understand human language?" example_title: "NLP" inference: parameters: max_length: 96 no_repeat_ngram_size: 2 encoder_no_repeat_ngram_size: 4 repetition_penalty: 3.51 length_penalty: 0.8 num_beams: 4 early_stopping: True --- # t5 - base- askscience - [t5-v1_1](https://huggingface.co/google/t5-v1_1-base) trained on the entirety of the _askscience_ sub-section of the eli5 dataset for one epoch. - compare to bart on eli5 [here](https://huggingface.co/yjernite/bart_eli5) - note that for the inference API, the model is restricted to outputting 96 tokens - by using the model in python with the transformers library, you can get longer outputs. ## training - for inputs, the model was presented with the post title and the post selftext encoded as: `question: <post title> context: <post selftext>`. You may see better results if queries are posed in this fashion. - The top two replies were aggregated and presented to the model as the output text. - Training for longer will be explored, but given that the dataset has 127k examples and the loss flatlines at 0.5 epochs so this model should be fairly viable.
racai/distilbert-multi-base-romanian-cased	099fba005eedaeda5a4aabcf6e502c2df50f58db	2021-12-24T17:32:28.000Z	[ "pytorch", "tf", "jax", "distilbert", "ro", "dataset:oscar", "dataset:wikipedia", "arxiv:2112.12650", "transformers", "license:mit" ]	null	false	racai	null	racai/distilbert-multi-base-romanian-cased	15	null	transformers	9,563	--- language: ro license: mit datasets: - oscar - wikipedia --- # Romanian DistilBERT This repository contains the a Romanian cased version of DistilBERT (named DistilMulti-BERT-base-ro in the paper) that was obtained by distilling an ensemble of two teacher models: [dumitrescustefan/bert-base-romanian-cased-v1](https://huggingface.co/dumitrescustefan/bert-base-romanian-cased-v1) and [readerbench/RoBERT-base](https://huggingface.co/readerbench/RoBERT-base). The model was introduced in [this paper](https://arxiv.org/abs/2112.12650). The adjacent code can be found [here](https://github.com/racai-ai/Romanian-DistilBERT). ## Usage ```python from transformers import AutoTokenizer, AutoModel # load the tokenizer and the model tokenizer = AutoTokenizer.from_pretrained("racai/distilbert-multi-base-romanian-cased") model = AutoModel.from_pretrained("racai/distilbert-multi-base-romanian-cased") # tokenize a test sentence input_ids = tokenizer.encode("Aceasta este o propoziție de test.", add_special_tokens=True, return_tensors="pt") # run the tokens trough the model outputs = model(input_ids) print(outputs) ``` ## Model Size The model is 35% smaller than `bert-base-romanian-cased-v1` and 30% smaller than `RoBERT-base`. \| Model \| Size (MB) \| Params (Millions) \| \|--------------------------------\|:---------:\|:----------------:\| \| RoBERT-base \| 441 \| 114 \| \| bert-base-romanian-cased-v1 \| 477 \| 124 \| \| distilbert-multi-base-romanian-cased \| 312 \| 81 \| ## Evaluation We evaluated the model in comparison with its two teachers on 5 Romanian tasks: - UPOS: Universal Part of Speech (F1-macro) - XPOS: Extended Part of Speech (F1-macro) - NER: Named Entity Recognition (F1-macro) - SAPN: Sentiment Anlaysis - Positive vs Negative (Accuracy) - SAR: Sentiment Analysis - Rating (F1-macro) - DI: Dialect identification (F1-macro) - STS: Semantic Textual Similarity (Pearson) \| Model \| UPOS \| XPOS \| NER \| SAPN \| SAR \| DI \| STS \| \|--------------------------------\|:----:\|:----:\|:---:\|:----:\|:---:\|:--:\|:---:\| \| RoBERT-base \| 98.02 \| 97.15 \| 85.14 \| 98.30 \| 79.40 \| 96.07 \| 81.18 \| \| bert-base-romanian-cased-v1 \| 98.00 \| 96.46 \| 85.88 \| 98.07 \| 79.61 \| 95.58 \| 80.30 \| \| distilbert-multi-base-romanian-cased \| 98.07 \| 96.83 \| 83.22 \| 98.11 \| 79.77 \| 96.18 \| 80.66 \| ### BibTeX entry and citation info ```bibtex @article{avram2021distilling, title={Distilling the Knowledge of Romanian BERTs Using Multiple Teachers}, author={Andrei-Marius Avram and Darius Catrina and Dumitru-Clementin Cercel and Mihai Dascălu and Traian Rebedea and Vasile Păiş and Dan Tufiş}, journal={ArXiv}, year={2021}, volume={abs/2112.12650} } ```
ramybaly/ner_conll2003	8705927627b7f05803a0a150c8c369c376ad1383	2021-08-21T03:21:14.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "dataset:conll2003", "transformers", "generated_from_trainer", "license:apache-2.0", "autotrain_compatible" ]	token-classification	false	ramybaly	null	ramybaly/ner_conll2003	15	null	transformers	9,564	--- license: apache-2.0 tags: - generated_from_trainer datasets: - conll2003 metrics: - precision - recall - f1 - accuracy model_index: - name: ner_conll2003 results: - task: name: Token Classification type: token-classification dataset: name: conll2003 type: conll2003 args: conll2003 metric: name: Accuracy type: accuracy value: 0.9772880710440217 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # ner_conll2003 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the conll2003 dataset. It achieves the following results on the evaluation set: - Loss: 0.1495 - Precision: 0.8985 - Recall: 0.9130 - F1: 0.9057 - Accuracy: 0.9773 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 10 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Precision \| Recall \| F1 \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:---------:\|:------:\|:------:\|:--------:\| \| 0.423 \| 1.0 \| 877 \| 0.0656 \| 0.9158 \| 0.9268 \| 0.9213 \| 0.9818 \| \| 0.0575 \| 2.0 \| 1754 \| 0.0574 \| 0.9285 \| 0.9445 \| 0.9364 \| 0.9847 \| \| 0.0295 \| 3.0 \| 2631 \| 0.0631 \| 0.9414 \| 0.9456 \| 0.9435 \| 0.9859 \| \| 0.0155 \| 4.0 \| 3508 \| 0.0680 \| 0.9395 \| 0.9467 \| 0.9431 \| 0.9860 \| \| 0.0097 \| 5.0 \| 4385 \| 0.0694 \| 0.9385 \| 0.9513 \| 0.9449 \| 0.9863 \| \| 0.0059 \| 6.0 \| 5262 \| 0.0743 \| 0.9363 \| 0.9471 \| 0.9416 \| 0.9860 \| \| 0.0041 \| 7.0 \| 6139 \| 0.0803 \| 0.9371 \| 0.9518 \| 0.9444 \| 0.9862 \| ### Framework versions - Transformers 4.9.1 - Pytorch 1.9.0+cu102 - Datasets 1.11.0 - Tokenizers 0.10.2
rebeccakoganlee/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ner	badcc7e3abb95def4e8dac5fc7bb5610b6c8e865	2021-11-23T20:42:01.000Z	[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	rebeccakoganlee	null	rebeccakoganlee/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ner	15	null	transformers	9,565	Entry not found
sarahmiller137/distilbert-base-uncased-ft-conll2003	97f129c62514a7e0aa79f6c8de00c17c792065a4	2022-07-14T11:52:53.000Z	[ "pytorch", "distilbert", "token-classification", "en", "dataset:conll2003", "transformers", "token classification", "license:cc", "model-index", "autotrain_compatible" ]	token-classification	false	sarahmiller137	null	sarahmiller137/distilbert-base-uncased-ft-conll2003	15	null	transformers	9,566	--- language: - en thumbnail: url to a thumbnail used in social sharing tags: - token classification license: cc datasets: - conll2003 model-index: - name: sarahmiller137/distilbert-base-uncased-ft-conll2003 results: - task: type: token-classification name: Token Classification dataset: name: conll2003 type: conll2003 config: conll2003 split: test metrics: - name: Accuracy type: accuracy value: 0.9750189904012154 verified: true - name: Precision type: precision value: 0.9802152215150602 verified: true - name: Recall type: recall value: 0.9803021169462076 verified: true - name: F1 type: f1 value: 0.9802586673049137 verified: true - name: loss type: loss value: 0.10723897069692612 verified: true --- ## Model information: distilbert-base-uncased model finetuned using the conll2003 dataset from the datasets library. ## Intended uses & limitations This model is intended to be used for named entity recoginition tasks. The model will identify entities of persons, locations, organisations, and miscellaneous. The model will predict lables based upon the CoNLL-2003 dataset. Note that the dataset and model may not be fully represetative or suitable for all needs it is recommended that the paper for the dataset and base model card should be reviewed before using the model - - [CoNLL-2003](https://aclanthology.org/W03-0419) - [distilbert](https://huggingface.co/distilbert-base-uncased) ## How to use Load the model from the library using the following checkpoints: ```python from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("sarahmiller137/distilbert-base-uncased-ft-conll2003") model = AutoModel.from_pretrained("sarahmiller137/distilbert-base-uncased-ft-conll2003") ```
sentence-transformers/distilroberta-base-msmarco-v1	b625c1f869c7869b660b456ecf8eff290d1333e3	2022-06-16T01:04:36.000Z	[ "pytorch", "tf", "roberta", "feature-extraction", "arxiv:1908.10084", "sentence-transformers", "sentence-similarity", "transformers", "license:apache-2.0" ]	sentence-similarity	false	sentence-transformers	null	sentence-transformers/distilroberta-base-msmarco-v1	15	null	sentence-transformers	9,567	--- pipeline_tag: sentence-similarity license: apache-2.0 tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- ⚠️ This model is deprecated. Please don't use it as it produces sentence embeddings of low quality. You can find recommended sentence embedding models here: [SBERT.net - Pretrained Models](https://www.sbert.net/docs/pretrained_models.html) # sentence-transformers/distilroberta-base-msmarco-v1 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('sentence-transformers/distilroberta-base-msmarco-v1') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/distilroberta-base-msmarco-v1') model = AutoModel.from_pretrained('sentence-transformers/distilroberta-base-msmarco-v1') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(*encoded_input) # Perform pooling. In this case, max pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results For an automated evaluation of this model, see the Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/distilroberta-base-msmarco-v1) ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 350, 'do_lower_case': False}) with Transformer model: RobertaModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) ) ``` ## Citing & Authors This model was trained by [sentence-transformers](https://www.sbert.net/). If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084): ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "http://arxiv.org/abs/1908.10084", } ```
sismetanin/sbert-ru-sentiment-rutweetcorp	b7f3c2cea37655f012af16fc852c454f3f998e64	2021-05-20T06:41:48.000Z	[ "pytorch", "jax", "bert", "text-classification", "transformers" ]	text-classification	false	sismetanin	null	sismetanin/sbert-ru-sentiment-rutweetcorp	15	null	transformers	9,568	Entry not found
skplanet/dialog-koelectra-small-discriminator	d7a060551ecc231b35236a8d7d1647876c193986	2021-04-13T01:15:27.000Z	[ "pytorch", "electra", "pretraining", "transformers" ]	null	false	skplanet	null	skplanet/dialog-koelectra-small-discriminator	15	null	transformers	9,569	# Dialog-KoELECTRA Github : [https://github.com/skplanet/Dialog-KoELECTRA](https://github.com/skplanet/Dialog-KoELECTRA) ## Introduction Dialog-KoELECTRA is a language model specialized for dialogue. It was trained with 22GB colloquial and written style Korean text data. Dialog-ELECTRA model is made based on the [ELECTRA](https://openreview.net/pdf?id=r1xMH1BtvB) model. ELECTRA is a method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. <br> ## Released Models We are initially releasing small version pre-trained model. The model was trained on Korean text. We hope to release other models, such as base/large models, in the future. \| Model \| Layers \| Hidden Size \| Params \| Max<br/>Seq Len \| Learning<br/>Rate \| Batch Size \| Train Steps \| \| :---: \| :---: \| :---: \| :---: \| :---: \| :---: \| :---: \| :---: \| \| Dialog-KoELECTRA-Small \| 12 \| 256 \| 14M \| 128 \| 1e-4 \| 512 \| 700K \| <br> ## Model Performance Dialog-KoELECTRA shows strong performance in conversational downstream tasks. \| \| NSMC<br/>(acc) \| Question Pair<br/>(acc) \| Korean-Hate-Speech<br/>(F1) \| Naver NER<br/>(F1) \| KorNLI<br/>(acc) \| KorSTS<br/>(spearman) \| \| :--------------------- \| :----------------: \| :--------------------: \| :----------------: \| :------------------: \| :-----------------------: \| :-------------------------: \| \| DistilKoBERT \| 88.60 \| 92.48 \| 60.72 \| 84.65 \| 72.00 \| 72.59 \| \| Dialog-KoELECTRA-Small \| 90.01 \| 94.99 \| 68.26 \| 85.51 \| 78.54 \| 78.96 \| <br> ## Train Data <table class="tg"> <thead> <tr> <th class="tg-c3ow"></th> <th class="tg-c3ow">corpus name</th> <th class="tg-c3ow">size</th> </tr> </thead> <tbody> <tr> <td class="tg-c3ow" rowspan="4">dialog</td> <td class="tg-0pky"><a href="https://aihub.or.kr/aidata/85" target="_blank" rel="noopener noreferrer">Aihub Korean dialog corpus</a></td> <td class="tg-c3ow" rowspan="4">7GB</td> </tr> <tr> <td class="tg-0pky"><a href="https://corpus.korean.go.kr/" target="_blank" rel="noopener noreferrer">NIKL Spoken corpus</a></td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/songys/Chatbot_data" target="_blank" rel="noopener noreferrer">Korean chatbot data</a></td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/Beomi/KcBERT" target="_blank" rel="noopener noreferrer">KcBERT</a></td> </tr> <tr> <td class="tg-c3ow" rowspan="2">written</td> <td class="tg-0pky"><a href="https://corpus.korean.go.kr/" target="_blank" rel="noopener noreferrer">NIKL Newspaper corpus</a></td> <td class="tg-c3ow" rowspan="2">15GB</td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/lovit/namuwikitext" target="_blank" rel="noopener noreferrer">namuwikitext</a></td> </tr> </tbody> </table> <br> ## Vocabulary We applied morpheme analysis using [huggingface_konlpy](https://github.com/lovit/huggingface_konlpy) when creating a vocabulary dictionary. As a result of the experiment, it showed better performance than a vocabulary dictionary created without applying morpheme analysis. <table> <thead> <tr> <th>vocabulary size</th> <th>unused token size</th> <th>limit alphabet</th> <th>min frequency</th> </tr> </thead> <tbody> <tr> <td>40,000</td> <td>500</td> <td>6,000</td> <td>3</td> </tr> </tbody> </table> <br>
spencerh/leftpartisan	9cb2baaa91cdf3c9982ddced7c076550d9c32739	2021-04-23T19:27:15.000Z	[ "pytorch", "tf", "distilbert", "text-classification", "transformers" ]	text-classification	false	spencerh	null	spencerh/leftpartisan	15	null	transformers	9,570	# Text classifier using DistilBERT to determine Partisanship ## This is one of many single-class partisanship models label_0 refers to "left" while label_1 refers to "other". This model was trained on 40,000 articles. ### Best Practices This model was optimized for 512 token-length text. Any text below 150 tokens will result in inaccurate results.
superb/wav2vec2-large-superb-er	bd13d8ed2b396e23676111aacff32283c9dece5d	2021-11-04T16:03:41.000Z	[ "pytorch", "wav2vec2", "audio-classification", "en", "dataset:superb", "arxiv:2105.01051", "transformers", "speech", "audio", "license:apache-2.0" ]	audio-classification	false	superb	null	superb/wav2vec2-large-superb-er	15	null	transformers	9,571	--- language: en datasets: - superb tags: - speech - audio - wav2vec2 - audio-classification license: apache-2.0 widget: - example_title: IEMOCAP clip "happy" src: https://cdn-media.huggingface.co/speech_samples/IEMOCAP_Ses01F_impro03_F013.wav - example_title: IEMOCAP clip "neutral" src: https://cdn-media.huggingface.co/speech_samples/IEMOCAP_Ses01F_impro04_F000.wav --- # Wav2Vec2-Large for Emotion Recognition ## Model description This is a ported version of [S3PRL's Wav2Vec2 for the SUPERB Emotion Recognition task](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream/emotion). The base model is [wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60), which is pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. For more information refer to [SUPERB: Speech processing Universal PERformance Benchmark](https://arxiv.org/abs/2105.01051) ## Task and dataset description Emotion Recognition (ER) predicts an emotion class for each utterance. The most widely used ER dataset [IEMOCAP](https://sail.usc.edu/iemocap/) is adopted, and we follow the conventional evaluation protocol: we drop the unbalanced emotion classes to leave the final four classes with a similar amount of data points and cross-validate on five folds of the standard splits. For the original model's training and evaluation instructions refer to the [S3PRL downstream task README](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream#er-emotion-recognition). ## Usage examples You can use the model via the Audio Classification pipeline: ```python from datasets import load_dataset from transformers import pipeline dataset = load_dataset("anton-l/superb_demo", "er", split="session1") classifier = pipeline("audio-classification", model="superb/wav2vec2-large-superb-er") labels = classifier(dataset[0]["file"], top_k=5) ``` Or use the model directly: ```python import torch import librosa from datasets import load_dataset from transformers import Wav2Vec2ForSequenceClassification, Wav2Vec2FeatureExtractor def map_to_array(example): speech, _ = librosa.load(example["file"], sr=16000, mono=True) example["speech"] = speech return example # load a demo dataset and read audio files dataset = load_dataset("anton-l/superb_demo", "er", split="session1") dataset = dataset.map(map_to_array) model = Wav2Vec2ForSequenceClassification.from_pretrained("superb/wav2vec2-large-superb-er") feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained("superb/wav2vec2-large-superb-er") # compute attention masks and normalize the waveform if needed inputs = feature_extractor(dataset[:4]["speech"], sampling_rate=16000, padding=True, return_tensors="pt") logits = model(inputs).logits predicted_ids = torch.argmax(logits, dim=-1) labels = [model.config.id2label[_id] for _id in predicted_ids.tolist()] ``` ## Eval results The evaluation metric is accuracy. \| \| s3prl \| transformers \| \|--------\|-----------\|------------------\| \|session1**\| `0.6564` \| `N/A` \| ### BibTeX entry and citation info ```bibtex @article{yang2021superb, title={SUPERB: Speech processing Universal PERformance Benchmark}, author={Yang, Shu-wen and Chi, Po-Han and Chuang, Yung-Sung and Lai, Cheng-I Jeff and Lakhotia, Kushal and Lin, Yist Y and Liu, Andy T and Shi, Jiatong and Chang, Xuankai and Lin, Guan-Ting and others}, journal={arXiv preprint arXiv:2105.01051}, year={2021} } ```
tbrasil/classificador_de_atendimento_2_classes_v1.1	147ae7455fb7891fbcef6e27de67badb01055d22	2021-08-02T17:51:56.000Z	[ "pytorch", "bert", "text-classification", "transformers" ]	text-classification	false	tbrasil	null	tbrasil/classificador_de_atendimento_2_classes_v1.1	15	null	transformers	9,572	Entry not found
textattack/albert-base-v2-snli	706e88d23c4ec5a68fbff2c1517d0da6ef7287d1	2020-07-06T16:36:47.000Z	[ "pytorch", "albert", "text-classification", "transformers" ]	text-classification	false	textattack	null	textattack/albert-base-v2-snli	15	null	transformers	9,573	## TextAttack Model Card This `albert-base-v2` model was fine-tuned for sequence classification using TextAttack and the snli dataset loaded using the `nlp` library. The model was fine-tuned for 5 epochs with a batch size of 64, a learning rate of 2e-05, and a maximum sequence length of 64. Since this was a classification task, the model was trained with a cross-entropy loss function. The best score the model achieved on this task was 0.9060150375939849, as measured by the eval set accuracy, found after 2 epochs. For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).
tmills/roberta_sfda_sharpseed	10d59130d9a12a683c1049a7573848ccc020ea1e	2021-05-20T22:41:21.000Z	[ "pytorch", "jax", "roberta", "text-classification", "transformers" ]	text-classification	false	tmills	null	tmills/roberta_sfda_sharpseed	15	null	transformers	9,574	Entry not found
ttop324/kogpt2jnovel	70c6af4eba91fb32a746588fc52c33c82437c58a	2021-11-11T07:38:14.000Z	[ "pytorch", "gpt2", "text-generation", "ko", "transformers", "license:cc-by-nc-sa-4.0" ]	text-generation	false	ttop324	null	ttop324/kogpt2jnovel	15	null	transformers	9,575	--- language: ko tags: - gpt2 license: cc-by-nc-sa-4.0 --- korean translated japan web novel finetuned from skt/kogpt2-base-v2
uclanlp/plbart-single_task-all-summarization	486692f974ac352601f3952a154d9aa9fa4bb7de	2022-03-02T07:28:07.000Z	[ "pytorch", "plbart", "text2text-generation", "transformers", "autotrain_compatible" ]	text2text-generation	false	uclanlp	null	uclanlp/plbart-single_task-all-summarization	15	null	transformers	9,576	Entry not found
ufal/byt5-small-multilexnorm2021-hr	c98026c96146a7d3d920dc64bf082f97f1900027	2021-10-20T12:27:34.000Z	[ "pytorch", "t5", "text2text-generation", "hr", "dataset:mc4", "dataset:wikipedia", "dataset:multilexnorm", "arxiv:2105.13626", "arxiv:1907.06292", "transformers", "lexical normalization", "license:apache-2.0", "autotrain_compatible" ]	text2text-generation	false	ufal	null	ufal/byt5-small-multilexnorm2021-hr	15	null	transformers	9,577	--- language: hr datasets: - mc4 - wikipedia - multilexnorm tags: - lexical normalization license: apache-2.0 --- # Fine-tuned ByT5-small for MultiLexNorm (Croatian version) ![model image](https://github.com/ufal/multilexnorm2021/raw/master/img/overall.png) This is the official release of the fine-tuned models for the winning entry to the [W-NUT 2021: Multilingual Lexical Normalization (MultiLexNorm) shared task](https://noisy-text.github.io/2021/multi-lexnorm.html), which evaluates lexical-normalization systems on 12 social media datasets in 11 languages. Our system is based on [ByT5](https://arxiv.org/abs/2105.13626), which we first pre-train on synthetic data and then fine-tune on authentic normalization data. It achieves the best performance by a wide margin in intrinsic evaluation, and also the best performance in extrinsic evaluation through dependency parsing. In addition to these fine-tuned models, we also release the source files on [GitHub](https://github.com/ufal/multilexnorm2021) and an interactive demo on [Google Colab](https://colab.research.google.com/drive/1rxpI8IlKk-D2crFqi2hdzbTBIezqgsCg?usp=sharing). ## How to use The model was not fine-tuned in a standard sentence-to-sentence setting – instead, it was tailored to the token-to-token definition of MultiLexNorm data. Please refer to [the interactive demo on Colab notebook](https://colab.research.google.com/drive/1rxpI8IlKk-D2crFqi2hdzbTBIezqgsCg?usp=sharing) to learn how to use these models. ## How to cite ```bibtex @inproceedings{wnut-ufal, title= "{ÚFAL} at {MultiLexNorm} 2021: Improving Multilingual Lexical Normalization by Fine-tuning {ByT5}", author = "Samuel, David and Straka, Milan", booktitle = "Proceedings of the 7th Workshop on Noisy User-generated Text (W-NUT 2021)", year = "2021", publisher = "Association for Computational Linguistics", address = "Punta Cana, Dominican Republic" } ``` ## ByT5 - Small ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-small). ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task. ByT5 works especially well on noisy text data,e.g., `google/byt5-small` significantly outperforms [mt5-small](https://huggingface.co/google/mt5-small) on [TweetQA](https://arxiv.org/abs/1907.06292). Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626) Authors: Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel
unicamp-dl/ptt5-large-t5-vocab	c213b7615a0ecd776dfe6f2d95fcaca06fd03647	2021-06-23T14:32:15.000Z	[ "pytorch", "tf", "t5", "text2text-generation", "pt", "dataset:brWaC", "transformers", "tensorflow", "pt-br", "license:mit", "autotrain_compatible" ]	text2text-generation	false	unicamp-dl	null	unicamp-dl/ptt5-large-t5-vocab	15	null	transformers	9,578	--- language: pt license: mit tags: - t5 - pytorch - tensorflow - pt - pt-br datasets: - brWaC widget: - text: "Texto de exemplo em português" inference: false --- # Portuguese T5 (aka "PTT5") ## Introduction PTT5 is a T5 model pretrained in the BrWac corpus, a large collection of web pages in Portuguese, improving T5's performance on Portuguese sentence similarity and entailment tasks. It's available in three sizes (small, base and large) and two vocabularies (Google's T5 original and ours, trained on Portuguese Wikipedia). For further information or requests, please go to [PTT5 repository](https://github.com/unicamp-dl/PTT5). ## Available models \| Model \| Size \| #Params \| Vocabulary \| \| :-: \| :-: \| :-: \| :-: \| \| [unicamp-dl/ptt5-small-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-small-t5-vocab) \| small \| 60M \| Google's T5 \| \| [unicamp-dl/ptt5-base-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-base-t5-vocab) \| base \| 220M \| Google's T5 \| \| [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) \| large \| 740M \| Google's T5 \| \| [unicamp-dl/ptt5-small-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-small-portuguese-vocab) \| small \| 60M \| Portuguese \| \| [unicamp-dl/ptt5-base-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-base-portuguese-vocab) (Recommended) \| base \| 220M \| Portuguese \| \| [unicamp-dl/ptt5-large-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-large-portuguese-vocab) \| large \| 740M \| Portuguese \| ## Usage ```python # Tokenizer from transformers import T5Tokenizer # PyTorch (bare model, baremodel + language modeling head) from transformers import T5Model, T5ForConditionalGeneration # Tensorflow (bare model, baremodel + language modeling head) from transformers import TFT5Model, TFT5ForConditionalGeneration model_name = 'unicamp-dl/ptt5-base-portuguese-vocab' tokenizer = T5Tokenizer.from_pretrained(model_name) # PyTorch model_pt = T5ForConditionalGeneration.from_pretrained(model_name) # TensorFlow model_tf = TFT5ForConditionalGeneration.from_pretrained(model_name) ``` # Citation If you use PTT5, please cite: @article{ptt5_2020, title={PTT5: Pretraining and validating the T5 model on Brazilian Portuguese data}, author={Carmo, Diedre and Piau, Marcos and Campiotti, Israel and Nogueira, Rodrigo and Lotufo, Roberto}, journal={arXiv preprint arXiv:2008.09144}, year={2020} }
vasudevgupta/bigbird-pegasus-large-pubmed	5cb34a36cccf14bb7bed607bade700d74e923fc2	2021-05-04T11:12:55.000Z	[ "pytorch", "bigbird_pegasus", "text2text-generation", "transformers", "autotrain_compatible" ]	text2text-generation	false	vasudevgupta	null	vasudevgupta/bigbird-pegasus-large-pubmed	15	null	transformers	9,579	Moved here: https://huggingface.co/google/bigbird-pegasus-large-pubmed
vukpetar/trocr-small-photomath	daa6f7cd6b80a9040ddb2ca4f15061652d2068cc	2021-12-27T19:41:43.000Z	[ "pytorch", "vision-encoder-decoder", "arxiv:2109.10282", "transformers" ]	null	false	vukpetar	null	vukpetar/trocr-small-photomath	15	null	transformers	9,580	## TrOCR (small-sized model, fine-tuned on Synthetic Math Expression Dataset) TrOCR model fine-tuned on the Synthetic Math Expression Dataset. It was introduced in the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Li et al. and first released in [this repository](https://github.com/microsoft/unilm/tree/master/trocr). Disclaimer: The team releasing TrOCR did not write a model card for this model so this model card has been written by the Hugging Face team. ## Model description The TrOCR model is an encoder-decoder model, consisting of an image Transformer as encoder, and a text Transformer as decoder. The image encoder was initialized from the weights of BEiT, while the text decoder was initialized from the weights of RoBERTa. Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder. Next, the Transformer text decoder autoregressively generates tokens. ## Intended uses & limitations You can use the raw model for optical character recognition (OCR) on single text-line images. See the model hub to look for fine-tuned versions on a task that interests you. ## How to use Here is how to use this model in PyTorch: ```python from transformers import VisionEncoderDecoderModel, AutoFeatureExtractor, AutoTokenizer from PIL import Image import requests # load image from the IAM database url = 'https://drive.google.com/uc?export=view&id=15dUjO44YDe1Agw_Qi8MyODRHpUFaCFw-' image = Image.open(requests.get(url, stream=True).raw).convert("RGB") feature_extractor = AutoFeatureExtractor.from_pretrained('vukpetar/trocr-small-photomath') tokenizer = AutoTokenizer.from_pretrained("vukpetar/trocr-small-photomath") model = VisionEncoderDecoderModel.from_pretrained('vukpetar/trocr-small-photomath') pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values generated_ids = model.generate(pixel_values) generated_text = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] ``` ## BibTeX entry and citation info @misc{li2021trocr, title={TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models}, author={Minghao Li and Tengchao Lv and Lei Cui and Yijuan Lu and Dinei Florencio and Cha Zhang and Zhoujun Li and Furu Wei}, year={2021}, eprint={2109.10282}, archivePrefix={arXiv}, primaryClass={cs.CL} }
yazdipour/text-to-sparql-t5-small	f485542939bc21807227d766ecbb8e47007c989d	2021-10-19T11:17:46.000Z	[ "pytorch", "tensorboard", "t5", "text2text-generation", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index", "autotrain_compatible" ]	text2text-generation	false	yazdipour	null	yazdipour/text-to-sparql-t5-small	15	null	transformers	9,581	--- license: apache-2.0 tags: - generated_from_trainer datasets: - null metrics: - f1 model-index: - name: text-to-sparql-t5-small-2021-10-19_10-17_lastDS results: - task: name: Sequence-to-sequence Language Modeling type: text2text-generation metrics: - name: F1 type: f1 value: 0.3129461705684662 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # text-to-sparql-t5-small-2021-10-19_10-17_lastDS This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2335 - Gen Len: 19.0 - P: 0.5580 - R: 0.0884 - F1: 0.3129 - Score: 5.9585 - Bleu-precisions: [90.11303396628615, 80.34125695971072, 73.81487011728768, 69.48796722990271] - Bleu-bp: 0.0763 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Gen Len \| P \| R \| F1 \| Score \| Bleu-precisions \| Bleu-bp \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:-------:\|:------:\|:------:\|:------:\|:------:\|:----------------------------------------------------------------------------:\|:-------:\| \| 0.3166 \| 1.0 \| 4807 \| 0.2335 \| 19.0 \| 0.5580 \| 0.0884 \| 0.3129 \| 5.9585 \| [90.11303396628615, 80.34125695971072, 73.81487011728768, 69.48796722990271] \| 0.0763 \| ### Framework versions - Transformers 4.10.0 - Pytorch 1.9.0+cu111 - Datasets 1.13.3 - Tokenizers 0.10.3
yuvraj/xSumm	6403af6f8f4eaf246bc94eef9d4ec1df88e2eca9	2020-12-11T22:05:01.000Z	[ "pytorch", "bart", "text2text-generation", "en", "transformers", "summarization", "extreme summarization", "autotrain_compatible" ]	summarization	false	yuvraj	null	yuvraj/xSumm	15	null	transformers	9,582	--- language: "en" tags: - summarization - extreme summarization --- ## Model description BartForConditionalGenerationModel for extreme summarization- creates a one line abstractive summary of a given article ## How to use PyTorch model available ```python from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline tokenizer = AutoTokenizer.from_pretrained("yuvraj/xSumm") model = AutoModelWithLMHead.from_pretrained("yuvraj/xSumm") xsumm = pipeline('summarization', model=model, tokenizer=tokenizer) xsumm("<text to be summarized>") ## Limitations and bias Trained on a small fraction of the xsumm training dataset
zhiheng-huang/bert-large-uncased-whole-word-masking-embedding-relative-key-query	8e5156c80b48db5fbe0868ca18d4e4e462a896b0	2021-05-20T09:48:50.000Z	[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	zhiheng-huang	null	zhiheng-huang/bert-large-uncased-whole-word-masking-embedding-relative-key-query	15	null	transformers	9,583	Entry not found
Davlan/xlm-roberta-base-masakhaner	643ee144abafa9c5fbe5f71f25d8a0118b6344a3	2022-02-25T15:23:22.000Z	[ "pytorch", "xlm-roberta", "token-classification", "am", "ha", "ig", "rw", "lg", "luo", "pcm", "sw", "wo", "yo", "multilingual", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "autotrain_compatible" ]	token-classification	false	Davlan	null	Davlan/xlm-roberta-base-masakhaner	15	null	transformers	9,584	Hugging Face's logo --- language: - am - ha - ig - rw - lg - luo - pcm - sw - wo - yo - multilingual datasets: - masakhaner --- # xlm-roberta-base-masakhaner ## Model description xlm-roberta-base-masakhaner is the first Named Entity Recognition model for 10 African languages (Amharic, Hausa, Igbo, Kinyarwanda, Luganda, Nigerian Pidgin, Swahili, Wolof, and Yorùbá) based on a fine-tuned XLM-RoBERTa large model. It achieves the state-of-the-art performance for the NER task. It has been trained to recognize four types of entities: dates & times (DATE), location (LOC), organizations (ORG), and person (PER). Specifically, this model is a xlm-roberta-large model that was fine-tuned on an aggregation of African language datasets obtained from Masakhane [MasakhaNER](https://github.com/masakhane-io/masakhane-ner) dataset. ## Intended uses & limitations #### How to use You can use this model with Transformers pipeline for NER. ```python from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline tokenizer = AutoTokenizer.from_pretrained("Davlan/xlm-roberta-base-masakhaner") model = AutoModelForTokenClassification.from_pretrained("Davlan/xlm-roberta-base-masakhaner") nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Emir of Kano turban Zhang wey don spend 18 years for Nigeria" ner_results = nlp(example) print(ner_results) ``` #### Limitations and bias This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains. ## Training data This model was fine-tuned on 10 African NER datasets (Amharic, Hausa, Igbo, Kinyarwanda, Luganda, Nigerian Pidgin, Swahili, Wolof, and Yorùbá) Masakhane [MasakhaNER](https://github.com/masakhane-io/masakhane-ner) dataset The training dataset distinguishes between the beginning and continuation of an entity so that if there are back-to-back entities of the same type, the model can output where the second entity begins. As in the dataset, each token will be classified as one of the following classes: Abbreviation\|Description -\|- O\|Outside of a named entity B-DATE \|Beginning of a DATE entity right after another DATE entity I-DATE \|DATE entity B-PER \|Beginning of a person’s name right after another person’s name I-PER \|Person’s name B-ORG \|Beginning of an organisation right after another organisation I-ORG \|Organisation B-LOC \|Beginning of a location right after another location I-LOC \|Location ## Training procedure This model was trained on a single NVIDIA V100 GPU with recommended hyperparameters from the [original MasakhaNER paper](https://arxiv.org/abs/2103.11811) which trained & evaluated the model on MasakhaNER corpus. ### BibTeX entry and citation info ``` @article{adelani21tacl, title = {Masakha{NER}: Named Entity Recognition for African Languages}, author = {David Ifeoluwa Adelani and Jade Abbott and Graham Neubig and Daniel D'souza and Julia Kreutzer and Constantine Lignos and Chester Palen-Michel and Happy Buzaaba and Shruti Rijhwani and Sebastian Ruder and Stephen Mayhew and Israel Abebe Azime and Shamsuddeen Muhammad and Chris Chinenye Emezue and Joyce Nakatumba-Nabende and Perez Ogayo and Anuoluwapo Aremu and Catherine Gitau and Derguene Mbaye and Jesujoba Alabi and Seid Muhie Yimam and Tajuddeen Gwadabe and Ignatius Ezeani and Rubungo Andre Niyongabo and Jonathan Mukiibi and Verrah Otiende and Iroro Orife and Davis David and Samba Ngom and Tosin Adewumi and Paul Rayson and Mofetoluwa Adeyemi and Gerald Muriuki and Emmanuel Anebi and Chiamaka Chukwuneke and Nkiruka Odu and Eric Peter Wairagala and Samuel Oyerinde and Clemencia Siro and Tobius Saul Bateesa and Temilola Oloyede and Yvonne Wambui and Victor Akinode and Deborah Nabagereka and Maurice Katusiime and Ayodele Awokoya and Mouhamadane MBOUP and Dibora Gebreyohannes and Henok Tilaye and Kelechi Nwaike and Degaga Wolde and Abdoulaye Faye and Blessing Sibanda and Orevaoghene Ahia and Bonaventure F. P. Dossou and Kelechi Ogueji and Thierno Ibrahima DIOP and Abdoulaye Diallo and Adewale Akinfaderin and Tendai Marengereke and Salomey Osei}, journal = {Transactions of the Association for Computational Linguistics (TACL)}, month = {}, url = {https://arxiv.org/abs/2103.11811}, year = {2021} } ```
ghadeermobasher/BC5CDR-Chem2-imbalanced-BiomedNLP-PubMedBERT-base-uncased-abstract	a95c14a635a2938902dc6d864e6b1fb147e1faa9	2022-03-01T06:00:08.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC5CDR-Chem2-imbalanced-BiomedNLP-PubMedBERT-base-uncased-abstract	15	null	transformers	9,585	Entry not found
ghadeermobasher/Model_org	ec0a976a03a68831924a915e003e3cbe8eee4ee6	2022-03-01T21:25:54.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_org	15	null	transformers	9,586	Entry not found
ghadeermobasher/Model_imb	e01715bfd053cf8e19121f5b09131f2e2394a1ee	2022-03-01T21:26:51.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_imb	15	null	transformers	9,587	Entry not found
ghadeermobasher/Model_imb_1	1a90a396146a91d18e13ceae7a50bb42962c0250	2022-03-02T04:10:59.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_imb_1	15	null	transformers	9,588	Entry not found
ghadeermobasher/Model_org_1	9925b3108d651f35452c969d7289aca4d75f6ab1	2022-03-02T04:16:51.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_org_1	15	null	transformers	9,589	Entry not found
ghadeermobasher/Model_imb_2	4fdc981bc5a81f8d794ab339886b2630f0d7b089	2022-03-02T11:34:17.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_imb_2	15	null	transformers	9,590	Entry not found
ghadeermobasher/Model_co_imb	d4dcc81191114c756cac56ba59b7babe3d471a85	2022-03-01T23:08:18.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/Model_co_imb	15	null	transformers	9,591	Entry not found
ActivationAI/distilbert-base-uncased-finetuned-emotion	dbf4470880ff3b73f22975241cd309bdf8e2195f	2022-03-02T03:40:08.000Z	[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:emotion", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]	text-classification	false	ActivationAI	null	ActivationAI/distilbert-base-uncased-finetuned-emotion	15	null	transformers	9,592	--- license: apache-2.0 tags: - generated_from_trainer datasets: - emotion metrics: - accuracy - f1 model-index: - name: distilbert-base-uncased-finetuned-emotion results: - task: name: Text Classification type: text-classification dataset: name: emotion type: emotion args: default metrics: - name: Accuracy type: accuracy value: 0.928 - name: F1 type: f1 value: 0.9280065074208208 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # distilbert-base-uncased-finetuned-emotion This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset. It achieves the following results on the evaluation set: - Loss: 0.2128 - Accuracy: 0.928 - F1: 0.9280 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 64 - eval_batch_size: 64 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| F1 \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\|:------:\| \| 0.8151 \| 1.0 \| 250 \| 0.3043 \| 0.907 \| 0.9035 \| \| 0.24 \| 2.0 \| 500 \| 0.2128 \| 0.928 \| 0.9280 \| ### Framework versions - Transformers 4.11.3 - Pytorch 1.10.0+cu111 - Datasets 1.16.1 - Tokenizers 0.10.3
ghadeermobasher/BC4-original-PubmedBert	96d1e5c0b00169c4644306ddfd91da3dbe509f69	2022-03-03T02:53:18.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-original-PubmedBert	15	null	transformers	9,593	Entry not found
ghadeermobasher/BC4-original-PubmedBert_small	f72da94ac4f4208739edfaca46482bd43873f66f	2022-03-02T11:07:21.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-original-PubmedBert_small	15	null	transformers	9,594	Entry not found
ghadeermobasher/BC4-modified-PubmedBert_small	c7ffad75406d8918820a2a813bca1e9e6a013c60	2022-03-02T11:07:25.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ghadeermobasher	null	ghadeermobasher/BC4-modified-PubmedBert_small	15	null	transformers	9,595	Entry not found
ivanlau/distil-bert-uncased-finetuned-github-issues	0d35383caff319649c8996504a0f8b5b0a33dea4	2022-03-04T10:16:47.000Z	[ "pytorch", "distilbert", "text-classification", "dataset:ticket tagger", "transformers", "model-index" ]	text-classification	false	ivanlau	null	ivanlau/distil-bert-uncased-finetuned-github-issues	15	null	transformers	9,596	--- datasets: - ticket tagger metrics: - accuracy model-index: - name: distil-bert-uncased-finetuned-github-issues results: - task: name: Text Classification type: text-classification dataset: name: ticket tagger type: ticket tagger args: full metrics: - name: Accuracy type: accuracy value: 0.7862 --- # Model Description This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) and fine-tuning it on the [github ticket tagger dataset](https://tickettagger.blob.core.windows.net/datasets/dataset-labels-top3-30k-real.txt). It classifies issue into 3 common categories: Bug, Enhancement, Questions. It achieves the following results on the evaluation set: - Accuracy: 0.7862 ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-5 - train_batch_size: 16 - optimizer: AdamW with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 0 - num_epochs: 5 ### Codes https://github.com/IvanLauLinTiong/IntelliLabel
l3cube-pune/marathi-albert-v2	76ef3b6421baf9bb747e102310594550f4627587	2022-06-26T15:13:43.000Z	[ "pytorch", "albert", "fill-mask", "mr", "dataset:L3Cube-MahaCorpus", "arxiv:2202.01159", "transformers", "license:cc-by-4.0", "autotrain_compatible" ]	fill-mask	false	l3cube-pune	null	l3cube-pune/marathi-albert-v2	15	1	transformers	9,597	--- license: cc-by-4.0 language: mr datasets: - L3Cube-MahaCorpus --- ## MahaAlBERT MahaAlBERT is a Marathi AlBERT model trained on L3Cube-MahaCorpus and other publicly available Marathi monolingual datasets. [dataset link] (https://github.com/l3cube-pune/MarathiNLP) More details on the dataset, models, and baseline results can be found in our [paper] (https://arxiv.org/abs/2202.01159) ``` @InProceedings{joshi:2022:WILDRE6, author = {Joshi, Raviraj}, title = {L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resources}, booktitle = {Proceedings of The WILDRE-6 Workshop within the 13th Language Resources and Evaluation Conference}, month = {June}, year = {2022}, address = {Marseille, France}, publisher = {European Language Resources Association}, pages = {97--101} } ```
timothyshi/bart-large-cnn-finetuned-booksum-chapter	24ef62670565e5ca800a0c4365d7db48bea3f494	2022-03-07T05:13:01.000Z	[ "pytorch", "tensorboard", "bart", "text2text-generation", "transformers", "generated_from_trainer", "license:mit", "model-index", "autotrain_compatible" ]	text2text-generation	false	timothyshi	null	timothyshi/bart-large-cnn-finetuned-booksum-chapter	15	1	transformers	9,598	--- license: mit tags: - generated_from_trainer metrics: - rouge model-index: - name: bart-large-cnn-finetuned-booksum-chapter results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # bart-large-cnn-finetuned-booksum-chapter This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.1373 - Rouge1: 18.1222 - Rouge2: 3.5783 - Rougel: 13.4084 - Rougelsum: 13.5832 - Gen Len: 63.5121 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Rouge1 \| Rouge2 \| Rougel \| Rougelsum \| Gen Len \| \|:-------------:\|:-----:\|:-----:\|:---------------:\|:-------:\|:------:\|:-------:\|:---------:\|:-------:\| \| 2.5297 \| 1.0 \| 23094 \| 3.1373 \| 18.1222 \| 3.5783 \| 13.4084 \| 13.5832 \| 63.5121 \| ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.6
datarpit/toy	9ffd6ea55f01c2da71bfd7f7a3c6c5a3f5472cdb	2022-03-10T00:06:22.000Z	[ "pytorch", "distilbert", "question-answering", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index", "autotrain_compatible" ]	question-answering	false	datarpit	null	datarpit/toy	15	null	transformers	9,599	--- license: apache-2.0 tags: - generated_from_trainer model-index: - name: toy results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # toy This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2124 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 50 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| \|:-------------:\|:-----:\|:-----:\|:---------------:\| \| 0.4798 \| 1.0 \| 231 \| 0.2252 \| \| 0.3378 \| 2.0 \| 462 \| 0.1777 \| \| 0.1024 \| 3.0 \| 693 \| 0.1586 \| \| 0.0736 \| 4.0 \| 924 \| 0.1664 \| \| 0.1237 \| 5.0 \| 1155 \| 0.1692 \| \| 0.1049 \| 6.0 \| 1386 \| 0.1818 \| \| 0.0239 \| 7.0 \| 1617 \| 0.2127 \| \| 0.0036 \| 8.0 \| 1848 \| 0.1888 \| \| 0.0051 \| 9.0 \| 2079 \| 0.2061 \| \| 0.0003 \| 10.0 \| 2310 \| 0.1905 \| \| 0.0005 \| 11.0 \| 2541 \| 0.2011 \| \| 0.0003 \| 12.0 \| 2772 \| 0.1928 \| \| 0.0029 \| 13.0 \| 3003 \| 0.2563 \| \| 0.0002 \| 14.0 \| 3234 \| 0.2076 \| \| 0.0002 \| 15.0 \| 3465 \| 0.1980 \| \| 0.0001 \| 16.0 \| 3696 \| 0.2013 \| \| 0.0001 \| 17.0 \| 3927 \| 0.2089 \| \| 0.0001 \| 18.0 \| 4158 \| 0.1984 \| \| 0.0001 \| 19.0 \| 4389 \| 0.2017 \| \| 0.0001 \| 20.0 \| 4620 \| 0.2013 \| \| 0.0001 \| 21.0 \| 4851 \| 0.2142 \| \| 0.0001 \| 22.0 \| 5082 \| 0.1943 \| \| 0.0001 \| 23.0 \| 5313 \| 0.2003 \| \| 0.0 \| 24.0 \| 5544 \| 0.2015 \| \| 0.0001 \| 25.0 \| 5775 \| 0.2031 \| \| 0.0002 \| 26.0 \| 6006 \| 0.2600 \| \| 0.0022 \| 27.0 \| 6237 \| 0.2269 \| \| 0.0 \| 28.0 \| 6468 \| 0.2125 \| \| 0.0 \| 29.0 \| 6699 \| 0.2172 \| \| 0.0 \| 30.0 \| 6930 \| 0.2185 \| \| 0.0 \| 31.0 \| 7161 \| 0.2004 \| \| 0.0 \| 32.0 \| 7392 \| 0.2077 \| \| 0.0 \| 33.0 \| 7623 \| 0.2333 \| \| 0.0003 \| 34.0 \| 7854 \| 0.2102 \| \| 0.0 \| 35.0 \| 8085 \| 0.2095 \| \| 0.0 \| 36.0 \| 8316 \| 0.2030 \| \| 0.0 \| 37.0 \| 8547 \| 0.2038 \| \| 0.0 \| 38.0 \| 8778 \| 0.2062 \| \| 0.0 \| 39.0 \| 9009 \| 0.2080 \| \| 0.0 \| 40.0 \| 9240 \| 0.2083 \| \| 0.0 \| 41.0 \| 9471 \| 0.2063 \| \| 0.0 \| 42.0 \| 9702 \| 0.2146 \| \| 0.0 \| 43.0 \| 9933 \| 0.2168 \| \| 0.0 \| 44.0 \| 10164 \| 0.2112 \| \| 0.0 \| 45.0 \| 10395 \| 0.2109 \| \| 0.0 \| 46.0 \| 10626 \| 0.2116 \| \| 0.0 \| 47.0 \| 10857 \| 0.2122 \| \| 0.0 \| 48.0 \| 11088 \| 0.2122 \| \| 0.0 \| 49.0 \| 11319 \| 0.2124 \| \| 0.0 \| 50.0 \| 11550 \| 0.2124 \| ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0 - Datasets 1.18.3 - Tokenizers 0.11.6

bhavikardeshna/xlm-roberta-base-german

530e8c3dd543800078c3dfbfcde30c883480f258

2021-12-21T11:40:35.000Z

[ "pytorch", "xlm-roberta", "question-answering", "arxiv:2112.09866", "transformers", "autotrain_compatible" ]

question-answering

false

bhavikardeshna

null

bhavikardeshna/xlm-roberta-base-german

15

null

transformers

9,500

# BibTeX entry and citation info ``` @misc{pandya2021cascading, title={Cascading Adaptors to Leverage English Data to Improve Performance of Question Answering for Low-Resource Languages}, author={Hariom A. Pandya and Bhavik Ardeshna and Dr. Brijesh S. Bhatt}, year={2021}, eprint={2112.09866}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

bs-modeling-metadata/website_metadata_exp_1_model_100k_checkpoint

875249e5fc9357a93f6eab4461688b3ac18d40dc

2021-10-07T13:32:07.000Z

[ "pytorch", "gpt2", "text-generation", "transformers" ]

text-generation

false

bs-modeling-metadata

null

bs-modeling-metadata/website_metadata_exp_1_model_100k_checkpoint

15

1

transformers

9,501

Entry not found

bsingh/roberta_goEmotion

af498dcbab4ef49f7163cac455aa0d34ae7d25d8

2021-10-11T00:26:09.000Z

[ "pytorch", "roberta", "text-classification", "en", "dataset:go_emotions", "transformers", "emotions", "license:mit" ]

text-classification

false

bsingh

null

bsingh/roberta_goEmotion

15

null

transformers

9,502

--- language: en tags: - text-classification - pytorch - roberta - emotions datasets: - go_emotions license: mit widget: - text: "I am not feeling well today." --- ## This model is trained for GoEmotions dataset which contains labeled 58k Reddit comments with 28 emotions - admiration, amusement, anger, annoyance, approval, caring, confusion, curiosity, desire, disappointment, disapproval, disgust, embarrassment, excitement, fear, gratitude, grief, joy, love, nervousness, optimism, pride, realization, relief, remorse, sadness, surprise + neutral ## Training details: - The training script is provided here: https://github.com/bsinghpratap/roberta_train_goEmotion - Please feel free to start an issue in the repo if you have trouble running the model and I would try to respond as soon as possible. - The model works well on most of the emotions except: 'desire', 'disgust', 'embarrassment', 'excitement', 'fear', 'grief', 'nervousness', 'pride', 'relief', 'remorse', 'surprise'] - I'll try to fine-tune the model further and update here if RoBERTa achieves a better performance. - Each text datapoint can have more than 1 label. Most of the training set had 1 label: Counter({1: 36308, 2: 6541, 3: 532, 4: 28, 5: 1}). So currently I just used the first label for each of the datapoint. Not ideal but it does a decent job. ## Model Performance ============================================================<br> Emotion: admiration<br> ============================================================<br> GoEmotions Paper: 0.65<br> RoBERTa: 0.62<br> Support: 504<br> ============================================================<br> Emotion: amusement<br> ============================================================<br> GoEmotions Paper: 0.80<br> RoBERTa: 0.78<br> Support: 252<br> ============================================================<br> Emotion: anger<br> ============================================================<br> GoEmotions Paper: 0.47<br> RoBERTa: 0.44<br> Support: 197<br> ============================================================<br> Emotion: annoyance<br> ============================================================<br> GoEmotions Paper: 0.34<br> RoBERTa: 0.22<br> Support: 286<br> ============================================================<br> Emotion: approval<br> ============================================================<br> GoEmotions Paper: 0.36<br> RoBERTa: 0.31<br> Support: 318<br> ============================================================<br> Emotion: caring<br> ============================================================<br> GoEmotions Paper: 0.39<br> RoBERTa: 0.24<br> Support: 114<br> ============================================================<br> Emotion: confusion<br> ============================================================<br> GoEmotions Paper: 0.37<br> RoBERTa: 0.29<br> Support: 139<br> ============================================================<br> Emotion: curiosity<br> ============================================================<br> GoEmotions Paper: 0.54<br> RoBERTa: 0.48<br> Support: 233<br> ============================================================<br> Emotion: disappointment<br> ============================================================<br> GoEmotions Paper: 0.28<br> RoBERTa: 0.18<br> Support: 127<br> ============================================================<br> Emotion: disapproval<br> ============================================================<br> GoEmotions Paper: 0.39<br> RoBERTa: 0.26<br> Support: 220<br> ============================================================<br> Emotion: gratitude<br> ============================================================<br> GoEmotions Paper: 0.86<br> RoBERTa: 0.84<br> Support: 288<br> ============================================================<br> Emotion: joy<br> ============================================================<br> GoEmotions Paper: 0.51<br> RoBERTa: 0.47<br> Support: 116<br> ============================================================<br> Emotion: love<br> ============================================================<br> GoEmotions Paper: 0.78<br> RoBERTa: 0.68<br> Support: 169<br> ============================================================<br> Emotion: neutral<br> ============================================================<br> GoEmotions Paper: 0.68<br> RoBERTa: 0.61<br> Support: 1606<br> ============================================================<br> Emotion: optimism<br> ============================================================<br> GoEmotions Paper: 0.51<br> RoBERTa: 0.52<br> Support: 120<br> ============================================================<br> Emotion: realization<br> ============================================================<br> GoEmotions Paper: 0.21<br> RoBERTa: 0.15<br> Support: 109<br> ============================================================<br> Emotion: sadness<br> ============================================================<br> GoEmotions Paper: 0.49<br> RoBERTa: 0.42<br> Support: 108

cardiffnlp/bertweet-base-stance-atheism

8a4275b426ee8d4136b36ed826bd3feb2dc41f3c

2021-05-20T14:53:17.000Z

[ "pytorch", "tf", "jax", "roberta", "text-classification", "transformers" ]

text-classification

false

cardiffnlp

null

cardiffnlp/bertweet-base-stance-atheism

15

null

transformers

9,503

chrommium/rubert-base-cased-sentence-finetuned-headlines_X

3ff3429c5539d43e2a02328421cf8204c67695e4

2021-09-16T00:34:06.000Z

[ "pytorch", "tensorboard", "bert", "text-classification", "transformers", "generated_from_trainer", "model-index" ]

text-classification

false

chrommium

null

chrommium/rubert-base-cased-sentence-finetuned-headlines_X

15

null

transformers

9,504

--- tags: - generated_from_trainer metrics: - accuracy model-index: - name: rubert-base-cased-sentence-finetuned-headlines_X results: - task: name: Text Classification type: text-classification metrics: - name: Accuracy type: accuracy value: 0.952 ---  # rubert-base-cased-sentence-finetuned-headlines_X This model is a fine-tuned version of [DeepPavlov/rubert-base-cased-sentence](https://huggingface.co/DeepPavlov/rubert-base-cased-sentence) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.2535 - Accuracy: 0.952 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | No log | 1.0 | 157 | 0.2759 | 0.912 | | No log | 2.0 | 314 | 0.2538 | 0.936 | | No log | 3.0 | 471 | 0.2556 | 0.945 | | 0.1908 | 4.0 | 628 | 0.2601 | 0.95 | | 0.1908 | 5.0 | 785 | 0.2535 | 0.952 | ### Framework versions - Transformers 4.10.2 - Pytorch 1.9.0+cu102 - Datasets 1.12.1 - Tokenizers 0.10.3

classla/bcms-bertic-frenk-hate

f41364e4a917e75feddf03e7525b9aa001650aca

2022-06-01T09:31:46.000Z

[ "pytorch", "bert", "text-classification", "hr", "arxiv:1906.02045", "transformers", "hate-speech" ]

text-classification

false

classla

null

classla/bcms-bertic-frenk-hate

15

null

transformers

9,505

--- language: "hr" tags: - text-classification - hate-speech widget: - text: "Potpredsjednik Vlade i ministar branitelja Tomo Medved komentirao je Vladine planove za zakonsku zabranu pozdrava 'za dom spremni'." --- # bcms-bertic-frenk-hate Text classification model based on [`classla/bcms-bertic`](https://huggingface.co/classla/bcms-bertic) and fine-tuned on the [FRENK dataset](https://www.clarin.si/repository/xmlui/handle/11356/1433) comprising of LGBT and migrant hatespeech. Only the Croatian subset of the data was used for fine-tuning and the dataset has been relabeled for binary classification (offensive or acceptable). ## Fine-tuning hyperparameters Fine-tuning was performed with `simpletransformers`. Beforehand a brief hyperparameter optimisation was performed and the presumed optimal hyperparameters are: ```python model_args = { "num_train_epochs": 12, "learning_rate": 1e-5, "train_batch_size": 74} ``` ## Performance The same pipeline was run with two other transformer models and `fasttext` for comparison. Accuracy and macro F1 score were recorded for each of the 6 fine-tuning sessions and post festum analyzed. | model | average accuracy | average macro F1 | |----------------------------|------------------|------------------| | bcms-bertic-frenk-hate | 0.8313 | 0.8219 | | EMBEDDIA/crosloengual-bert | 0.8054 | 0.796 | | xlm-roberta-base | 0.7175 | 0.7049 | | fasttext | 0.771 | 0.754 | From recorded accuracies and macro F1 scores p-values were also calculated: Comparison with `crosloengual-bert`: | test | accuracy p-value | macro F1 p-value | |----------------|------------------|------------------| | Wilcoxon | 0.00781 | 0.00781 | | Mann Whithney | 0.00108 | 0.00108 | | Student t-test | 2.43e-10 | 1.27e-10 | Comparison with `xlm-roberta-base`: | test | accuracy p-value | macro F1 p-value | |----------------|------------------|------------------| | Wilcoxon | 0.00781 | 0.00781 | | Mann Whithney | 0.00107 | 0.00108 | | Student t-test | 4.83e-11 | 5.61e-11 | ## Use examples ```python from simpletransformers.classification import ClassificationModel model = ClassificationModel( "bert", "5roop/bcms-bertic-frenk-hate", use_cuda=True, ) predictions, logit_output = model.predict(['Ne odbacujem da će RH primiti još migranata iz Afganistana, no neće biti novog vala', "Potpredsjednik Vlade i ministar branitelja Tomo Medved komentirao je Vladine planove za zakonsku zabranu pozdrava 'za dom spremni' "]) predictions ### Output: ### array([0, 0]) ``` ## Citation If you use the model, please cite the following paper on which the original model is based: ``` @inproceedings{ljubesic-lauc-2021-bertic, title = "{BERT}i{\'c} - The Transformer Language Model for {B}osnian, {C}roatian, {M}ontenegrin and {S}erbian", author = "Ljube{\v{s}}i{\'c}, Nikola and Lauc, Davor", booktitle = "Proceedings of the 8th Workshop on Balto-Slavic Natural Language Processing", month = apr, year = "2021", address = "Kiyv, Ukraine", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/2021.bsnlp-1.5", pages = "37--42", } ``` and the dataset used for fine-tuning: ``` @misc{ljubešić2019frenk, title={The FRENK Datasets of Socially Unacceptable Discourse in Slovene and English}, author={Nikola Ljubešić and Darja Fišer and Tomaž Erjavec}, year={2019}, eprint={1906.02045}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/1906.02045} } ```

crystina-z/monoELECTRA_LCE_nneg31

47923dbe2fb0dedea1c1572940b1289806838a92

2022-02-11T18:02:52.000Z

[ "pytorch", "tf", "electra", "text-classification", "transformers" ]

text-classification

false

crystina-z

null

crystina-z/monoELECTRA_LCE_nneg31

15

null

transformers

9,506

Entry not found

dbmdz/electra-base-german-europeana-cased-generator

195c6427c576e68a7c2a97de2e20421fc506c58c

2020-07-26T00:53:55.000Z

[ "pytorch", "tf", "electra", "fill-mask", "transformers", "autotrain_compatible" ]

fill-mask

false

dbmdz

null

dbmdz/electra-base-german-europeana-cased-generator

15

null

transformers

9,507

Entry not found

fabriceyhc/bert-base-uncased-yahoo_answers_topics

968176fc24a2eb73cac26ab4312d8b22da98486a

2021-09-21T00:54:22.000Z

[ "pytorch", "bert", "text-classification", "dataset:yahoo_answers_topics", "transformers", "generated_from_trainer", "sibyl", "license:apache-2.0", "model-index" ]

text-classification

false

fabriceyhc

null

fabriceyhc/bert-base-uncased-yahoo_answers_topics

15

1

transformers

9,508

--- license: apache-2.0 tags: - generated_from_trainer - sibyl datasets: - yahoo_answers_topics metrics: - accuracy model-index: - name: bert-base-uncased-yahoo_answers_topics results: - task: name: Text Classification type: text-classification dataset: name: yahoo_answers_topics type: yahoo_answers_topics args: yahoo_answers_topics metrics: - name: Accuracy type: accuracy value: 0.7499166666666667 ---  # bert-base-uncased-yahoo_answers_topics This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the yahoo_answers_topics dataset. It achieves the following results on the evaluation set: - Loss: 0.8092 - Accuracy: 0.7499 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 86625 - training_steps: 866250 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:-----:|:---------------:|:--------:| | 2.162 | 0.01 | 2000 | 1.7444 | 0.5681 | | 1.3126 | 0.02 | 4000 | 1.0081 | 0.7054 | | 0.9592 | 0.03 | 6000 | 0.9021 | 0.7234 | | 0.8903 | 0.05 | 8000 | 0.8827 | 0.7276 | | 0.8685 | 0.06 | 10000 | 0.8540 | 0.7341 | | 0.8422 | 0.07 | 12000 | 0.8547 | 0.7365 | | 0.8535 | 0.08 | 14000 | 0.8264 | 0.7372 | | 0.8178 | 0.09 | 16000 | 0.8331 | 0.7389 | | 0.8325 | 0.1 | 18000 | 0.8242 | 0.7411 | | 0.8181 | 0.12 | 20000 | 0.8356 | 0.7437 | | 0.8171 | 0.13 | 22000 | 0.8090 | 0.7451 | | 0.8092 | 0.14 | 24000 | 0.8469 | 0.7392 | | 0.8057 | 0.15 | 26000 | 0.8185 | 0.7478 | | 0.8085 | 0.16 | 28000 | 0.8090 | 0.7467 | | 0.8229 | 0.17 | 30000 | 0.8225 | 0.7417 | | 0.8151 | 0.18 | 32000 | 0.8262 | 0.7419 | | 0.81 | 0.2 | 34000 | 0.8149 | 0.7383 | | 0.8073 | 0.21 | 36000 | 0.8225 | 0.7441 | | 0.816 | 0.22 | 38000 | 0.8037 | 0.744 | | 0.8217 | 0.23 | 40000 | 0.8409 | 0.743 | | 0.82 | 0.24 | 42000 | 0.8286 | 0.7385 | | 0.8101 | 0.25 | 44000 | 0.8282 | 0.7413 | | 0.8254 | 0.27 | 46000 | 0.8170 | 0.7414 | ### Framework versions - Transformers 4.10.2 - Pytorch 1.7.1 - Datasets 1.6.1 - Tokenizers 0.10.3

fbaigt/procbert

20814e122765866e213447ebe2618d2f0b90cbf1

2021-11-08T15:08:01.000Z

[ "pytorch", "bert", "feature-extraction", "en", "dataset:pubmed", "dataset:chemical patent", "dataset:cooking recipe", "arxiv:2109.04711", "transformers" ]

feature-extraction

false

fbaigt

null

fbaigt/procbert

15

1

transformers

9,509

--- language: - en datasets: - pubmed - chemical patent - cooking recipe --- ## ProcBERT ProcBERT is a pre-trained language model specifically for procedural text. It was pre-trained on a large-scale procedural corpus (PubMed articles/chemical patents/cooking recipes) containing over 12B tokens and shows great performance on downstream tasks. More details can be found in the following [paper](https://arxiv.org/abs/2109.04711): ``` @inproceedings{bai-etal-2021-pre, title = "Pre-train or Annotate? Domain Adaptation with a Constrained Budget", author = "Bai, Fan and Ritter, Alan and Xu, Wei", booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing", month = nov, year = "2021", address = "Online and Punta Cana, Dominican Republic", publisher = "Association for Computational Linguistics", } ``` ## Usage ``` from transformers import * tokenizer = AutoTokenizer.from_pretrained("fbaigt/procbert") model = AutoModelForTokenClassification.from_pretrained("fbaigt/procbert") ``` More usage details can be found [here](https://github.com/bflashcp3f/ProcBERT).

federicopascual/finetune-sentiment-analysis-model-3000-samples

595ae6575f96bc971fd033b3560a98e80ede9517

2021-12-30T19:29:48.000Z

[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:imdb", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]

text-classification

false

federicopascual

null

federicopascual/finetune-sentiment-analysis-model-3000-samples

15

null

transformers

9,510

--- license: apache-2.0 tags: - generated_from_trainer datasets: - imdb metrics: - accuracy - f1 model-index: - name: finetune-sentiment-analysis-model-3000-samples results: - task: name: Text Classification type: text-classification dataset: name: imdb type: imdb args: plain_text metrics: - name: Accuracy type: accuracy value: 0.8866666666666667 - name: F1 type: f1 value: 0.8944099378881988 ---  # finetune-sentiment-analysis-model-3000-samples This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.4558 - Accuracy: 0.8867 - F1: 0.8944 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results ### Framework versions - Transformers 4.15.0 - Pytorch 1.10.0+cu111 - Datasets 1.17.0 - Tokenizers 0.10.3

figurative-nlp/t5-figurative-paraphrase

4c382b695540ace5fa8ce647e3fcd67a372a93f8

2022-02-17T12:21:53.000Z

[ "pytorch", "t5", "text2text-generation", "transformers", "autotrain_compatible" ]

text2text-generation

false

figurative-nlp

null

figurative-nlp/t5-figurative-paraphrase

15

2

transformers

9,511

This model can convert the figurative/metaphorical expression to the literal expression. Below is the usage of our model: from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("figurative-nlp/t5-figurative-paraphrase") model = AutoModelForSeq2SeqLM.from_pretrained("figurative-nlp/t5-figurative-paraphrase") input_ids = tokenizer( "paraphrase the sentence : i will talk this story to you from A to Z", return_tensors="pt" ).input_ids # Batch size 1 outputs = model.generate(input_ids,num_beams = 5) result = tokenizer.decode(outputs[0], skip_special_tokens=True) #result : i will talk this story to you from beginning to end.. For example: **Input**: He is always bang on when he makes a speech. **Output**: He is always presice when he makes a speech. **Input**: He always buy what he said. **Output**: He always agree with what he said. **Input**: Your team will be done like dinner if they play against the all-star team. **Output**: Your team will be defeated if they play against the all-star team. (the one is not particularly accurate) Note: the figurative language here includes metaphor, idiom and simile. We don't guarantee that the results generated results are satisfactory to you. We are trying to improve the effect of the model.

fnlp/elasticbert-large

0a5689cea93ed0bf88c87bcd623e0de0f98516e2

2021-10-28T11:05:49.000Z

[ "pytorch", "elasticbert", "fill-mask", "arxiv:2110.07038", "transformers", "autotrain_compatible" ]

fill-mask

false

fnlp

null

fnlp/elasticbert-large

15

2

transformers

9,512

# ElasticBERT-LARGE ## Model description This is an implementation of the `large` version of ElasticBERT. [**Towards Efficient NLP: A Standard Evaluation and A Strong Baseline**](https://arxiv.org/pdf/2110.07038.pdf) Xiangyang Liu, Tianxiang Sun, Junliang He, Lingling Wu, Xinyu Zhang, Hao Jiang, Zhao Cao, Xuanjing Huang, Xipeng Qiu ## Code link [**fastnlp/elasticbert**](https://github.com/fastnlp/ElasticBERT) ## Usage ```python >>> from transformers import BertTokenizer as ElasticBertTokenizer >>> from models.configuration_elasticbert import ElasticBertConfig >>> from models.modeling_elasticbert import ElasticBertForSequenceClassification >>> num_output_layers = 1 >>> config = ElasticBertConfig.from_pretrained('fnlp/elasticbert-large', num_output_layers=num_output_layers ) >>> tokenizer = ElasticBertTokenizer.from_pretrained('fnlp/elasticbert-large') >>> model = ElasticBertForSequenceClassification.from_pretrained('fnlp/elasticbert-large', config=config) >>> input_ids = tokenizer.encode('The actors are fantastic .', return_tensors='pt') >>> outputs = model(input_ids) ``` ## Citation ```bibtex @article{liu2021elasticbert, author = {Xiangyang Liu and Tianxiang Sun and Junliang He and Lingling Wu and Xinyu Zhang and Hao Jiang and Zhao Cao and Xuanjing Huang and Xipeng Qiu}, title = {Towards Efficient {NLP:} {A} Standard Evaluation and {A} Strong Baseline}, journal = {CoRR}, volume = {abs/2110.07038}, year = {2021}, url = {https://arxiv.org/abs/2110.07038}, eprinttype = {arXiv}, eprint = {2110.07038}, timestamp = {Fri, 22 Oct 2021 13:33:09 +0200}, biburl = {https://dblp.org/rec/journals/corr/abs-2110-07038.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ```

geekfeed/gpt2_ja

6c297f7e58fc6e7c75d654941380620cd3710660

2021-05-21T16:11:52.000Z

[ "pytorch", "jax", "gpt2", "feature-extraction", "transformers" ]

feature-extraction

false

geekfeed

null

geekfeed/gpt2_ja

15

null

transformers

9,513

hello

ghadeermobasher/BC2GM-Gene-Modified_scibert_scivocab_cased

cf11c70aceb0994ec32ebedfa0a2e878043b12f9

2022-01-23T19:55:04.000Z

[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC2GM-Gene-Modified_scibert_scivocab_cased

15

null

transformers

9,514

Entry not found

ghadeermobasher/BC2GM-Gene_ImbalancedPubMedBERT

2574cb8b0ee0fe4af0e9b272c51f855bfa3c1b01

2022-01-22T01:44:25.000Z

[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC2GM-Gene_ImbalancedPubMedBERT

15

null

transformers

9,515

Entry not found

ghadeermobasher/BC4-Original-biobert-v1.1

ab10a139dbc7b0c0c9bc6a4558a206e5d73fb3e5

2022-02-24T14:45:29.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-Original-biobert-v1.1

15

null

transformers

9,516

Entry not found

ghadeermobasher/BC4-Original-bluebert_pubmed_uncased_L-12_H-768_A-12

cbceb41840ab0b1b4d9cd8d4cc7a19ac477fae8d

2022-02-24T14:22:37.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-Original-bluebert_pubmed_uncased_L-12_H-768_A-12

15

null

transformers

9,517

Entry not found

ghadeermobasher/BC4-Original-scibert_scivocab_uncased

1bb6e36961e4a831a996c806637ac891282bf3e9

2022-02-24T14:28:32.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-Original-scibert_scivocab_uncased

15

null

transformers

9,518

Entry not found

ghadeermobasher/BC5CDR-Chemical-imbalanced-bluebert_pubmed_uncased_L-12_H-768_A-12_latest

17f9443fd2d369763bf7c64b4eb9206803f0df27

2022-02-21T23:07:06.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC5CDR-Chemical-imbalanced-bluebert_pubmed_uncased_L-12_H-768_A-12_latest

15

null

transformers

9,519

Entry not found

glasses/resnet18

a15b2ef76c4e01cc6b3f4518f56c1c98722d6793

2021-11-30T20:06:28.000Z

[ "pytorch", "dataset:imagenet", "arxiv:1512.03385", "arxiv:1812.01187", "transformers", "image-classification", "license:apache-2.0" ]

image-classification

false

glasses

null

glasses/resnet18

15

null

transformers

9,520

--- license: apache-2.0 tags: - image-classification datasets: - imagenet --- # resnet18 Implementation of ResNet proposed in [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) ``` python ResNet.resnet18() ResNet.resnet26() ResNet.resnet34() ResNet.resnet50() ResNet.resnet101() ResNet.resnet152() ResNet.resnet200() Variants (d) proposed in `Bag of Tricks for Image Classification with Convolutional Neural Networks <https://arxiv.org/pdf/1812.01187.pdf`_ ResNet.resnet26d() ResNet.resnet34d() ResNet.resnet50d() # You can construct your own one by chaning `stem` and `block` resnet101d = ResNet.resnet101(stem=ResNetStemC, block=partial(ResNetBottleneckBlock, shortcut=ResNetShorcutD)) ``` Examples: ``` python # change activation ResNet.resnet18(activation = nn.SELU) # change number of classes (default is 1000 ) ResNet.resnet18(n_classes=100) # pass a different block ResNet.resnet18(block=SENetBasicBlock) # change the steam model = ResNet.resnet18(stem=ResNetStemC) change shortcut model = ResNet.resnet18(block=partial(ResNetBasicBlock, shortcut=ResNetShorcutD)) # store each feature x = torch.rand((1, 3, 224, 224)) # get features model = ResNet.resnet18() # first call .features, this will activate the forward hooks and tells the model you'll like to get the features model.encoder.features model(torch.randn((1,3,224,224))) # get the features from the encoder features = model.encoder.features print([x.shape for x in features]) #[torch.Size([1, 64, 112, 112]), torch.Size([1, 64, 56, 56]), torch.Size([1, 128, 28, 28]), torch.Size([1, 256, 14, 14])] ```

gpssohi/distilbart-qgen-3-3

66e9d4f41a1c4bf55bdab0a9eb904476542d5d06

2022-01-12T08:29:26.000Z

[ "pytorch", "bart", "text2text-generation", "en", "dataset:squad", "transformers", "question-generation", "summarization", "license:apache-2.0", "autotrain_compatible" ]

summarization

false

gpssohi

null

gpssohi/distilbart-qgen-3-3

15

2

transformers

9,521

--- language: en tags: - question-generation - summarization license: apache-2.0 datasets: - squad --- # Introduction This model checkpoint is obtained by first fine-tuning the sshleifer/distilbart-cnn-6-6 summarization checkpoint on the SQuAD dataset. After this, the 6-6 fine-tuned model is distilled down to a 3-3 model which gives us the final checkpoint. [GitHub Link for training scripts.](https://github.com/darth-c0d3r/bart-question-generation) # Usage The input format is as follows: `[answer] <s> [passage]`. The model will predict the question that corresponds to the answer from the passage. # Plot ![Distillation Run](distill_run_21.png) # Dataset The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Hence, the input to the model will be a passage context and an answer, and the output / target will be the question for the given answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chat-bots to lead a conversation. The final dataset is created by taking the union of the following Question Answering Datasets. The dataset must have the following three columns: context, question, answer. ## [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowd-workers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable. We use the SQuAD 1.1 variant which does not have unanswerable questions. So, every question will have a corresponding answer and vice-versa. ### Preprocessing The first step is to remove questions which don't have answers. After that, we split the train set into Train and Eval sets and treat the dev set as the test set. ### Stats **Original Dataset** | Split | Num Docs | Num Contexts | Ques w/ Ans | Ques w/o Ans | Num Unique Ans | | ----- | -------- | ------------ | ----------- | ------------ | -------------- | | Train | 442 | 19035 | 86821 | 43498 | 86821 | | Dev | 35 | 1204 | 5928 | 5945 | 10279 | **After Preprocessing** | Split | Num Rows | Context | Answer | Question | | ----- | -------- | ---------- | ------ | -------- | | Train | 80995 | 653,120,20 | 43,3,1 | 40,10,1 | | Eval | 5826 | 445,123,67 | 28,3,1 | 29,10,3 | | Test | 10297 | 629,129,25 | 29,4,1 | 31,10,3 | The numbers in the columns indicate max, avg, min number of words.

gpssohi/distilbart-qgen-6-6

18d85ee5d5482d7af7b5f719048f4bc641c3a5ff

2022-01-12T08:29:13.000Z

[ "pytorch", "bart", "text2text-generation", "en", "dataset:squad", "transformers", "summarization", "question-generation", "license:apache-2.0", "autotrain_compatible" ]

summarization

false

gpssohi

null

gpssohi/distilbart-qgen-6-6

15

1

transformers

9,522

--- language: en tags: - summarization - question-generation license: apache-2.0 datasets: - squad --- # Introduction This model checkpoint is obtained by fine-tuning the `sshleifer/distilbart-cnn-6-6` summarization checkpoint on the SQuAD dataset. [GitHub Link for training scripts.](https://github.com/darth-c0d3r/bart-question-generation) # Usage The input format is as follows: `[answer] <s> [passage]`. The model will predict the question that corresponds to the answer from the passage. # Plot ![Training Run](train_run_6.png) # Dataset The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Hence, the input to the model will be a passage context and an answer, and the output / target will be the question for the given answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chat-bots to lead a conversation. The final dataset is created by taking the union of the following Question Answering Datasets. The dataset must have the following three columns: context, question, answer. ## [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowd-workers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable. We use the SQuAD 1.1 variant which does not have unanswerable questions. So, every question will have a corresponding answer and vice-versa. ### Preprocessing The first step is to remove questions that don't have answers. After that, we split the train set into Train and Eval sets and treat the dev set as the test set. ### Stats **Original Dataset** | Split | Num Docs | Num Contexts | Ques w/ Ans | Ques w/o Ans | Num Unique Ans | | ----- | -------- | ------------ | ----------- | ------------ | -------------- | | Train | 442 | 19035 | 86821 | 43498 | 86821 | | Dev | 35 | 1204 | 5928 | 5945 | 10279 | **After Preprocessing** | Split | Num Rows | Context | Answer | Question | | ----- | -------- | ---------- | ------ | -------- | | Train | 80995 | 653,120,20 | 43,3,1 | 40,10,1 | | Eval | 5826 | 445,123,67 | 28,3,1 | 29,10,3 | | Test | 10297 | 629,129,25 | 29,4,1 | 31,10,3 | The numbers in the columns indicate max, avg, min number of words.

gwkim22/domain_base2_disc

253c926654ffcae55fe363bca751474d03b90ec7

2021-07-19T01:56:14.000Z

[ "pytorch", "electra", "pretraining", "transformers" ]

null

false

gwkim22

null

gwkim22/domain_base2_disc

15

1

transformers

9,523

"domain_base2_disc_0719"

hamzaMM/questionClassifier

d1d326c6965fb4f91070df4356562dade3b37364

2021-12-02T20:08:26.000Z

[ "pytorch", "bert", "text-classification", "transformers" ]

text-classification

false

hamzaMM

null

hamzaMM/questionClassifier

15

2

transformers

9,524

Entry not found

hrdipto/wav2vec2-xls-r-300m-bangla-command-generated-data-finetune

6361d8dfe95c21a0fe20389a195e9de9aab1de02

2022-02-14T08:58:20.000Z

[ "pytorch", "tensorboard", "wav2vec2", "automatic-speech-recognition", "transformers", "generated_from_trainer", "model-index" ]

automatic-speech-recognition

false

hrdipto

null

hrdipto/wav2vec2-xls-r-300m-bangla-command-generated-data-finetune

15

null

transformers

9,525

--- tags: - generated_from_trainer model-index: - name: wav2vec2-xls-r-300m-bangla-command-generated-data-finetune results: [] ---  # wav2vec2-xls-r-300m-bangla-command-generated-data-finetune This model is a fine-tuned version of [hrdipto/wav2vec2-xls-r-300m-bangla-command-data](https://huggingface.co/hrdipto/wav2vec2-xls-r-300m-bangla-command-data) on the None dataset. It achieves the following results on the evaluation set: - eval_loss: 0.0099 - eval_wer: 0.0208 - eval_runtime: 2.5526 - eval_samples_per_second: 75.217 - eval_steps_per_second: 9.402 - epoch: 71.43 - step: 2000 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 100 - mixed_precision_training: Native AMP ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.0

huggingartists/kanye-west

ef0b90df2f597af17783d7ae3477a01de520f35c

2022-05-05T00:27:21.000Z

[ "pytorch", "jax", "gpt2", "text-generation", "en", "dataset:huggingartists/kanye-west", "transformers", "huggingartists", "lyrics", "lm-head", "causal-lm" ]

text-generation

false

huggingartists

null

huggingartists/kanye-west

15

null

transformers

9,526

--- language: en datasets: - huggingartists/kanye-west tags: - huggingartists - lyrics - lm-head - causal-lm widget: - text: "I am" --- <div class="inline-flex flex-col" style="line-height: 1.5;"> <div class="flex"> <div style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://images.genius.com/54520386ec39aca6408c7e2c156ae10a.399x399x1.png')"> </div> </div> <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div> <div style="text-align: center; font-size: 16px; font-weight: 800">Kanye West</div> <a href="https://genius.com/artists/kanye-west"> <div style="text-align: center; font-size: 14px;">@kanye-west</div> </a> </div> I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists). Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)! ## How does it work? To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist). ## Training data The model was trained on lyrics from Kanye West. Dataset is available [here](https://huggingface.co/datasets/huggingartists/kanye-west). And can be used with: ```python from datasets import load_dataset dataset = load_dataset("huggingartists/kanye-west") ``` [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/hl7afoso/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Kanye West's lyrics. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/28dw8m5v) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/28dw8m5v/artifacts) is logged and versioned. ## How to use You can use this model directly with a pipeline for text generation: ```python from transformers import pipeline generator = pipeline('text-generation', model='huggingartists/kanye-west') generator("I am", num_return_sequences=5) ``` Or with Transformers library: ```python from transformers import AutoTokenizer, AutoModelWithLMHead tokenizer = AutoTokenizer.from_pretrained("huggingartists/kanye-west") model = AutoModelWithLMHead.from_pretrained("huggingartists/kanye-west") ``` ## Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About *Built by Aleksey Korshuk* [![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk) [![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk) [![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky) For more details, visit the project repository. [![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)

huggingtweets/amazon

7bc08510372ae3b213814f77f110ae1f3138dd3a

2021-05-21T18:33:06.000Z

[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]

text-generation

false

huggingtweets

null

huggingtweets/amazon

15

null

transformers

9,527

--- language: en thumbnail: https://www.huggingtweets.com/amazon/1609713999453/predictions.png tags: - huggingtweets widget: - text: "My dream is" --- <link rel="stylesheet" href="https://unpkg.com/@tailwindcss/[email protected]/dist/typography.min.css"> <style> @media (prefers-color-scheme: dark) { .prose { color: #E2E8F0 !important; } .prose h2, .prose h3, .prose a, .prose thead { color: #F7FAFC !important; } } </style> <section class='prose'> <div> <div style="width: 132px; height:132px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/949070360103698432/kXSiPeTk_400x400.jpg')"> </div> <div style="margin-top: 8px; font-size: 19px; font-weight: 800">Amazon 🤖 AI Bot </div> <div style="font-size: 15px; color: #657786">@amazon bot</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://app.wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-model-to-generate-tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on [@amazon's tweets](https://twitter.com/amazon). <table style='border-width:0'> <thead style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #CBD5E0'> <th style='border-width:0'>Data</th> <th style='border-width:0'>Quantity</th> </tr> </thead> <tbody style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Tweets downloaded</td> <td style='border-width:0'>3242</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Retweets</td> <td style='border-width:0'>40</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Short tweets</td> <td style='border-width:0'>60</td> </tr> <tr style='border-width:0'> <td style='border-width:0'>Tweets kept</td> <td style='border-width:0'>3142</td> </tr> </tbody> </table> [Explore the data](https://wandb.ai/wandb/huggingtweets/runs/1fd78mc2/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @amazon's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/76pxw0n0) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/76pxw0n0/artifacts) is logged and versioned. ## Intended uses & limitations ### How to use You can use this model directly with a pipeline for text generation: <pre><code><span style="color:#03A9F4">from</span> transformers <span style="color:#03A9F4">import</span> pipeline generator = pipeline(<span style="color:#FF9800">'text-generation'</span>, model=<span style="color:#FF9800">'huggingtweets/amazon'</span>) generator(<span style="color:#FF9800">"My dream is"</span>, num_return_sequences=<span style="color:#8BC34A">5</span>)</code></pre> ### Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About *Built by Boris Dayma* </section> [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) <section class='prose'> For more details, visit the project repository. </section> [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets)

huggingtweets/deepleffen

9f6a7a1f1bb73ef33f7cead6ee0b72ba37411d4f

2022-06-03T17:34:54.000Z

[ "pytorch", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]

text-generation

false

huggingtweets

null

huggingtweets/deepleffen

15

null

transformers

9,528

--- language: en thumbnail: http://www.huggingtweets.com/deepleffen/1654277690184/predictions.png tags: - huggingtweets widget: - text: "My dream is" --- <div class="inline-flex flex-col" style="line-height: 1.5;"> <div class="flex"> <div style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1241879678455078914/e2EdZIrr_400x400.jpg')"> </div> <div style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')"> </div> <div style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')"> </div> </div> <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI BOT 🤖</div> <div style="text-align: center; font-size: 16px; font-weight: 800">Deep Leffen Bot</div> <div style="text-align: center; font-size: 14px;">@deepleffen</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on tweets from Deep Leffen Bot. | Data | Deep Leffen Bot | | --- | --- | | Tweets downloaded | 589 | | Retweets | 14 | | Short tweets | 27 | | Tweets kept | 548 | [Explore the data](https://wandb.ai/wandb/huggingtweets/runs/1p32tock/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @deepleffen's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/imjjixah) for full transparency and reproducibility. At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/imjjixah/artifacts) is logged and versioned. ## How to use You can use this model directly with a pipeline for text generation: ```python from transformers import pipeline generator = pipeline('text-generation', model='huggingtweets/deepleffen') generator("My dream is", num_return_sequences=5) ``` ## Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About *Built by Boris Dayma* [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) For more details, visit the project repository. [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets)

huggingtweets/nvidia

cceee848844258789df63f96b72b676fade2a4aa

2021-05-22T17:00:36.000Z

[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers", "huggingtweets" ]

text-generation

false

huggingtweets

null

huggingtweets/nvidia

15

null

transformers

9,529

--- language: en thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true tags: - huggingtweets widget: - text: "My dream is" --- <link rel="stylesheet" href="https://unpkg.com/@tailwindcss/[email protected]/dist/typography.min.css"> <style> @media (prefers-color-scheme: dark) { .prose { color: #E2E8F0 !important; } .prose h2, .prose h3, .prose a, .prose thead { color: #F7FAFC !important; } } </style> <section class='prose'> <div> <div style="width: 132px; height:132px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1145524454170062848/U4lxVYEw_400x400.png')"> </div> <div style="margin-top: 8px; font-size: 19px; font-weight: 800">NVIDIA 🤖 AI Bot </div> <div style="font-size: 15px; color: #657786">@nvidia bot</div> </div> I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets). Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)! ## How does it work? The model uses the following pipeline. ![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true) To understand how the model was developed, check the [W&B report](https://app.wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-model-to-generate-tweets--VmlldzoxMTY5MjI). ## Training data The model was trained on [@nvidia's tweets](https://twitter.com/nvidia). <table style='border-width:0'> <thead style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #CBD5E0'> <th style='border-width:0'>Data</th> <th style='border-width:0'>Quantity</th> </tr> </thead> <tbody style='border-width:0'> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Tweets downloaded</td> <td style='border-width:0'>3222</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Retweets</td> <td style='border-width:0'>1876</td> </tr> <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'> <td style='border-width:0'>Short tweets</td> <td style='border-width:0'>18</td> </tr> <tr style='border-width:0'> <td style='border-width:0'>Tweets kept</td> <td style='border-width:0'>1328</td> </tr> </tbody> </table> [Explore the data](https://app.wandb.ai/wandb/huggingtweets/runs/35e4bboc/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline. ## Training procedure The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @nvidia's tweets. Hyperparameters and metrics are recorded in the [W&B training run](https://app.wandb.ai/wandb/huggingtweets/runs/2t7z7a45) for full transparency and reproducibility. At the end of training, [the final model](https://app.wandb.ai/wandb/huggingtweets/runs/2t7z7a45/artifacts) is logged and versioned. ## Intended uses & limitations ### How to use You can use this model directly with a pipeline for text generation: <pre><code><span style="color:#03A9F4">from</span> transformers <span style="color:#03A9F4">import</span> pipeline generator = pipeline(<span style="color:#FF9800">'text-generation'</span>, model=<span style="color:#FF9800">'huggingtweets/nvidia'</span>) generator(<span style="color:#FF9800">"My dream is"</span>, num_return_sequences=<span style="color:#8BC34A">5</span>)</code></pre> ### Limitations and bias The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias). In addition, the data present in the user's tweets further affects the text generated by the model. ## About *Built by Boris Dayma* </section> [![Follow](https://img.shields.io/twitter/follow/borisdayma?style=social)](https://twitter.com/intent/follow?screen_name=borisdayma) <section class='prose'> For more details, visit the project repository. </section> [![GitHub stars](https://img.shields.io/github/stars/borisdayma/huggingtweets?style=social)](https://github.com/borisdayma/huggingtweets)

imzachjohnson/autonlp-spinner-check-16492731

3e96dbddaf4b1b2c760fbf196391ac77ecfc7890

2021-10-11T00:02:11.000Z

[ "pytorch", "bert", "text-classification", "en", "dataset:imzachjohnson/autonlp-data-spinner-check", "transformers", "autonlp" ]

text-classification

false

imzachjohnson

null

imzachjohnson/autonlp-spinner-check-16492731

15

null

transformers

9,530

--- tags: autonlp language: en widget: - text: "I love AutoNLP 🤗" datasets: - imzachjohnson/autonlp-data-spinner-check --- # Model Trained Using AutoNLP - Problem type: Binary Classification - Model ID: 16492731 ## Validation Metrics - Loss: 0.21610039472579956 - Accuracy: 0.9155366722657816 - Precision: 0.9530714194995978 - Recall: 0.944871149164778 - AUC: 0.9553238723676906 - F1: 0.9489535692456846 ## Usage You can use cURL to access this model: ``` $ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoNLP"}' https://api-inference.huggingface.co/models/imzachjohnson/autonlp-spinner-check-16492731 ``` Or Python API: ``` from transformers import AutoModelForSequenceClassification, AutoTokenizer model = AutoModelForSequenceClassification.from_pretrained("imzachjohnson/autonlp-spinner-check-16492731", use_auth_token=True) tokenizer = AutoTokenizer.from_pretrained("imzachjohnson/autonlp-spinner-check-16492731", use_auth_token=True) inputs = tokenizer("I love AutoNLP", return_tensors="pt") outputs = model(**inputs) ```

jaimin/plagiarism_checker

63fdab21cd1ee8fa220533eeb00c77238156728f

2021-08-20T05:44:24.000Z

[ "pytorch", "longformer", "text-classification", "transformers" ]

text-classification

false

jaimin

null

jaimin/plagiarism_checker

15

null

transformers

9,531

"hello"

kanishka/GlossBERT

0cc3b83af5496e27ebcc95ef0cf37ea0a9281a7a

2021-09-22T08:54:41.000Z

[ "pytorch", "bert", "en", "dataset:SemCor3.0", "arxiv:1908.07245", "transformers", "glossbert", "license:mit" ]

null

false

kanishka

null

kanishka/GlossBERT

15

null

transformers

9,532

--- language: en tags: - glossbert license: mit datasets: - SemCor3.0 --- ## GlossBERT A BERT-based model fine-tuned on SemCor 3.0 to perform word-sense-disambiguation by leveraging gloss information. This model is the research output of the paper titled: '[GlossBERT: BERT for Word Sense Disambiguation with Gloss Knowledge](https://arxiv.org/pdf/1908.07245.pdf)' Disclaimer: This model was built and trained by a group of researchers different than the repository's author. The original model code can be found on github: https://github.com/HSLCY/GlossBERT ## Usage The following code loads GlossBERT: ```py from transformers import AutoTokenizer, BertForSequenceClassification tokenizer = AutoTokenizer.from_pretrained('kanishka/GlossBERT') model = BertForSequenceClassification.from_pretrained('kanishka/GlossBERT') ``` ## Citation If you use this model in any of your projects, please cite the original authors using the following bibtex: ``` @inproceedings{huang-etal-2019-glossbert, title = "{G}loss{BERT}: {BERT} for Word Sense Disambiguation with Gloss Knowledge", author = "Huang, Luyao and Sun, Chi and Qiu, Xipeng and Huang, Xuanjing", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)", month = nov, year = "2019", address = "Hong Kong, China", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/D19-1355", doi = "10.18653/v1/D19-1355", pages = "3507--3512" } ```

keshan/sinhala-roberta-oscar

655873a1b237c7e09c424d0a55bb9fb05456248e

2021-07-14T06:28:47.000Z

[ "pytorch", "jax", "tensorboard", "roberta", "fill-mask", "si", "dataset:oscar", "arxiv:1907.11692", "transformers", "oscar", "Sinhala", "autotrain_compatible" ]

fill-mask

false

keshan

null

keshan/sinhala-roberta-oscar

15

null

transformers

9,533

--- language: si tags: - oscar - Sinhala - roberta - fill-mask widget: - text: "මම සිංහල භාෂාව <mask>" datasets: - oscar --- ### Overview This is a slightly smaller model trained on [OSCAR](https://oscar-corpus.com/) Sinhala dedup dataset. As Sinhala is one of those low resource languages, there are only a handful of models been trained. So, this would be a great place to start training for more downstream tasks. ## Model Specification The model chosen for training is [Roberta](https://arxiv.org/abs/1907.11692) with the following specifications: 1. vocab_size=50265 2. max_position_embeddings=514 3. num_attention_heads=12 4. num_hidden_layers=12 5. type_vocab_size=1 ## How to Use You can use this model directly with a pipeline for masked language modeling: ```py from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline model = AutoModelWithLMHead.from_pretrained("keshan/sinhala-roberta-oscar") tokenizer = AutoTokenizer.from_pretrained("keshan/sinhala-roberta-oscar") fill_mask = pipeline('fill-mask', model=model, tokenizer=tokenizer) fill_mask("මම ගෙදර <mask>.") ```

kurianbenoy/distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb

465c2a001ccd95ff2faee805ffe909ef79fdf366

2022-02-21T11:55:41.000Z

[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:imdb", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]

text-classification

false

kurianbenoy

null

kurianbenoy/distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb

15

null

transformers

9,534

--- license: apache-2.0 tags: - generated_from_trainer datasets: - imdb metrics: - accuracy model-index: - name: distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb results: - task: name: Text Classification type: text-classification dataset: name: imdb type: imdb args: plain_text metrics: - name: Accuracy type: accuracy value: 0.93032 ---  # distilbert-base-uncased-finetuned-sst-2-english-finetuned-imdb This model is a fine-tuned version of [distilbert-base-uncased-finetuned-sst-2-english](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.2165 - Accuracy: 0.9303 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.2749 | 1.0 | 3125 | 0.2165 | 0.9303 | ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.0

leonardvorbeck/wav2vec2-large-robust-SB300

e630e8662f2d52c6be25ccd3e95ba417f7c8b21b

2021-08-26T12:22:18.000Z

[ "pytorch", "wav2vec2", "automatic-speech-recognition", "en", "dataset:libri_light", "dataset:common_voice", "dataset:switchboard", "dataset:fisher", "arxiv:2104.01027", "transformers", "speech", "CTC", "Attention", "license:apache-2.0" ]

automatic-speech-recognition

false

leonardvorbeck

null

leonardvorbeck/wav2vec2-large-robust-SB300

15

1

transformers

9,535

--- language: en datasets: - libri_light - common_voice - switchboard - fisher tags: - speech - automatic-speech-recognition - CTC - Attention - wav2vec2 license: apache-2.0 --- # Wav2Vec2-Large-Robust - Finetuned on Switchboard (300 hours) ## Note : Model has not been initialized. If you want to use it without further finetuning, do a forward pass first to recalculate the normalized weights of the positional convolutional layer : ```ipython with torch.no_grad(): model(torch.randn((1,300_000))) ``` [Facebook's Wav2Vec2](https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/) The base model pretrained on 16kHz sampled speech audio. Speech datasets from multiple domains were used to pretrain the model: - [Libri-Light](https://github.com/facebookresearch/libri-light): open-source audio books from the LibriVox project; clean, read-out audio data - [CommonVoice](https://huggingface.co/datasets/common_voice): crowd-source collected audio data; read-out text snippets - [Switchboard](https://catalog.ldc.upenn.edu/LDC97S62): telephone speech corpus; noisy telephone data - [Fisher](https://catalog.ldc.upenn.edu/LDC2004T19): conversational telephone speech; noisy telephone data When using the model make sure that your speech input is also sampled at 16Khz. Check out [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information. [Paper Robust Wav2Vec2](https://arxiv.org/abs/2104.01027) Authors: Wei-Ning Hsu, Anuroop Sriram, Alexei Baevski, Tatiana Likhomanenko, Qiantong Xu, Vineel Pratap, Jacob Kahn, Ann Lee, Ronan Collobert, Gabriel Synnaeve, Michael Auli **Abstract** Self-supervised learning of speech representations has been a very active research area but most work is focused on a single domain such as read audio books for which there exist large quantities of labeled and unlabeled data. In this paper, we explore more general setups where the domain of the unlabeled data for pre-training data differs from the domain of the labeled data for fine-tuning, which in turn may differ from the test data domain. Our experiments show that using target domain data during pre-training leads to large performance improvements across a variety of setups. On a large-scale competitive setup, we show that pre-training on unlabeled in-domain data reduces the gap between models trained on in-domain and out-of-domain labeled data by 66%-73%. This has obvious practical implications since it is much easier to obtain unlabeled target domain data than labeled data. Moreover, we find that pre-training on multiple domains improves generalization performance on domains not seen during training. Code and models will be made available at this https URL. The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20. # Usage See [this notebook](https://colab.research.google.com/drive/1FjTsqbYKphl9kL-eILgUc-bl4zVThL8F?usp=sharing) for more information on how to fine-tune the model.

luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5

45fa2fa42a97b0478a98380d53a7a50ad0177cec

2021-10-26T11:37:15.000Z

[ "pytorch", "roberta", "transformers" ]

null

false

luffycodes

null

luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5

15

null

transformers

9,536

Entry not found

luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5_norm_bb_mlm_loss

ef1df072f813dcd203c30e97b7b273ed1f4e33ad

2021-10-26T04:03:29.000Z

[ "pytorch", "roberta", "transformers" ]

null

false

luffycodes

null

luffycodes/bb_narataka_roberta_large_nli_bsz_16_bb_bsz_16_nli_lr_1e5_bb_lr_1e5_norm_bb_mlm_loss

15

null

transformers

9,537

Entry not found

m3hrdadfi/zabanshenas-roberta-base-mix

a82c58d3632aaedec457182ea6e65d523fc960b0

2021-06-24T19:40:27.000Z

[ "pytorch", "tf", "roberta", "text-classification", "multilingual", "dataset:wili_2018", "transformers", "license:apache-2.0" ]

text-classification

false

m3hrdadfi

null

m3hrdadfi/zabanshenas-roberta-base-mix

15

1

transformers

9,538

--- language: multilingual license: apache-2.0 datasets: - wili_2018 --- # Zabanshenas - Language Detector Zabanshenas is a Transformer-based solution for identifying the most likely language of a written document/text. Zabanshenas is a Persian word that has two meanings: - A person who studies linguistics. - A way to identify the type of written language. ## How to use Follow [Zabanshenas repo](https://github.com/m3hrdadfi/zabanshenas) for more information! ## Evaluation The following tables summarize the scores obtained by model overall and per each class. ### By Paragraph | language | precision | recall | f1-score | |:--------------------------------------:|:---------:|:--------:|:--------:| | Achinese (ace) | 1.000000 | 0.982143 | 0.990991 | | Afrikaans (afr) | 1.000000 | 1.000000 | 1.000000 | | Alemannic German (als) | 1.000000 | 0.946429 | 0.972477 | | Amharic (amh) | 1.000000 | 0.982143 | 0.990991 | | Old English (ang) | 0.981818 | 0.964286 | 0.972973 | | Arabic (ara) | 0.846154 | 0.982143 | 0.909091 | | Aragonese (arg) | 1.000000 | 1.000000 | 1.000000 | | Egyptian Arabic (arz) | 0.979592 | 0.857143 | 0.914286 | | Assamese (asm) | 0.981818 | 0.964286 | 0.972973 | | Asturian (ast) | 0.964912 | 0.982143 | 0.973451 | | Avar (ava) | 0.941176 | 0.905660 | 0.923077 | | Aymara (aym) | 0.964912 | 0.982143 | 0.973451 | | South Azerbaijani (azb) | 0.965517 | 1.000000 | 0.982456 | | Azerbaijani (aze) | 1.000000 | 1.000000 | 1.000000 | | Bashkir (bak) | 1.000000 | 0.978261 | 0.989011 | | Bavarian (bar) | 0.843750 | 0.964286 | 0.900000 | | Central Bikol (bcl) | 1.000000 | 0.982143 | 0.990991 | | Belarusian (Taraschkewiza) (be-tarask) | 1.000000 | 0.875000 | 0.933333 | | Belarusian (bel) | 0.870968 | 0.964286 | 0.915254 | | Bengali (ben) | 0.982143 | 0.982143 | 0.982143 | | Bhojpuri (bho) | 1.000000 | 0.928571 | 0.962963 | | Banjar (bjn) | 0.981132 | 0.945455 | 0.962963 | | Tibetan (bod) | 1.000000 | 0.982143 | 0.990991 | | Bosnian (bos) | 0.552632 | 0.375000 | 0.446809 | | Bishnupriya (bpy) | 1.000000 | 0.982143 | 0.990991 | | Breton (bre) | 1.000000 | 0.964286 | 0.981818 | | Bulgarian (bul) | 1.000000 | 0.964286 | 0.981818 | | Buryat (bxr) | 0.946429 | 0.946429 | 0.946429 | | Catalan (cat) | 0.982143 | 0.982143 | 0.982143 | | Chavacano (cbk) | 0.914894 | 0.767857 | 0.834951 | | Min Dong (cdo) | 1.000000 | 0.982143 | 0.990991 | | Cebuano (ceb) | 1.000000 | 1.000000 | 1.000000 | | Czech (ces) | 1.000000 | 1.000000 | 1.000000 | | Chechen (che) | 1.000000 | 1.000000 | 1.000000 | | Cherokee (chr) | 1.000000 | 0.963636 | 0.981481 | | Chuvash (chv) | 0.938776 | 0.958333 | 0.948454 | | Central Kurdish (ckb) | 1.000000 | 1.000000 | 1.000000 | | Cornish (cor) | 1.000000 | 1.000000 | 1.000000 | | Corsican (cos) | 1.000000 | 0.982143 | 0.990991 | | Crimean Tatar (crh) | 1.000000 | 0.946429 | 0.972477 | | Kashubian (csb) | 1.000000 | 0.963636 | 0.981481 | | Welsh (cym) | 1.000000 | 1.000000 | 1.000000 | | Danish (dan) | 1.000000 | 1.000000 | 1.000000 | | German (deu) | 0.828125 | 0.946429 | 0.883333 | | Dimli (diq) | 0.964912 | 0.982143 | 0.973451 | | Dhivehi (div) | 1.000000 | 1.000000 | 1.000000 | | Lower Sorbian (dsb) | 1.000000 | 0.982143 | 0.990991 | | Doteli (dty) | 0.940000 | 0.854545 | 0.895238 | | Emilian (egl) | 1.000000 | 0.928571 | 0.962963 | | Modern Greek (ell) | 1.000000 | 1.000000 | 1.000000 | | English (eng) | 0.588889 | 0.946429 | 0.726027 | | Esperanto (epo) | 1.000000 | 0.982143 | 0.990991 | | Estonian (est) | 0.963636 | 0.946429 | 0.954955 | | Basque (eus) | 1.000000 | 0.982143 | 0.990991 | | Extremaduran (ext) | 0.982143 | 0.982143 | 0.982143 | | Faroese (fao) | 1.000000 | 1.000000 | 1.000000 | | Persian (fas) | 0.948276 | 0.982143 | 0.964912 | | Finnish (fin) | 1.000000 | 1.000000 | 1.000000 | | French (fra) | 0.710145 | 0.875000 | 0.784000 | | Arpitan (frp) | 1.000000 | 0.946429 | 0.972477 | | Western Frisian (fry) | 0.982143 | 0.982143 | 0.982143 | | Friulian (fur) | 1.000000 | 0.982143 | 0.990991 | | Gagauz (gag) | 0.981132 | 0.945455 | 0.962963 | | Scottish Gaelic (gla) | 0.982143 | 0.982143 | 0.982143 | | Irish (gle) | 0.949153 | 1.000000 | 0.973913 | | Galician (glg) | 1.000000 | 1.000000 | 1.000000 | | Gilaki (glk) | 0.981132 | 0.945455 | 0.962963 | | Manx (glv) | 1.000000 | 1.000000 | 1.000000 | | Guarani (grn) | 1.000000 | 0.964286 | 0.981818 | | Gujarati (guj) | 1.000000 | 0.982143 | 0.990991 | | Hakka Chinese (hak) | 0.981818 | 0.964286 | 0.972973 | | Haitian Creole (hat) | 1.000000 | 1.000000 | 1.000000 | | Hausa (hau) | 1.000000 | 0.945455 | 0.971963 | | Serbo-Croatian (hbs) | 0.448276 | 0.464286 | 0.456140 | | Hebrew (heb) | 1.000000 | 0.982143 | 0.990991 | | Fiji Hindi (hif) | 0.890909 | 0.890909 | 0.890909 | | Hindi (hin) | 0.981481 | 0.946429 | 0.963636 | | Croatian (hrv) | 0.500000 | 0.636364 | 0.560000 | | Upper Sorbian (hsb) | 0.955556 | 1.000000 | 0.977273 | | Hungarian (hun) | 1.000000 | 1.000000 | 1.000000 | | Armenian (hye) | 1.000000 | 0.981818 | 0.990826 | | Igbo (ibo) | 0.918033 | 1.000000 | 0.957265 | | Ido (ido) | 1.000000 | 1.000000 | 1.000000 | | Interlingue (ile) | 1.000000 | 0.962264 | 0.980769 | | Iloko (ilo) | 0.947368 | 0.964286 | 0.955752 | | Interlingua (ina) | 1.000000 | 1.000000 | 1.000000 | | Indonesian (ind) | 0.761905 | 0.872727 | 0.813559 | | Icelandic (isl) | 1.000000 | 1.000000 | 1.000000 | | Italian (ita) | 0.861538 | 1.000000 | 0.925620 | | Jamaican Patois (jam) | 1.000000 | 0.946429 | 0.972477 | | Javanese (jav) | 0.964912 | 0.982143 | 0.973451 | | Lojban (jbo) | 1.000000 | 1.000000 | 1.000000 | | Japanese (jpn) | 1.000000 | 1.000000 | 1.000000 | | Karakalpak (kaa) | 0.965517 | 1.000000 | 0.982456 | | Kabyle (kab) | 1.000000 | 0.964286 | 0.981818 | | Kannada (kan) | 0.982143 | 0.982143 | 0.982143 | | Georgian (kat) | 1.000000 | 0.964286 | 0.981818 | | Kazakh (kaz) | 0.980769 | 0.980769 | 0.980769 | | Kabardian (kbd) | 1.000000 | 0.982143 | 0.990991 | | Central Khmer (khm) | 0.960784 | 0.875000 | 0.915888 | | Kinyarwanda (kin) | 0.981132 | 0.928571 | 0.954128 | | Kirghiz (kir) | 1.000000 | 1.000000 | 1.000000 | | Komi-Permyak (koi) | 0.962264 | 0.910714 | 0.935780 | | Konkani (kok) | 0.964286 | 0.981818 | 0.972973 | | Komi (kom) | 1.000000 | 0.962264 | 0.980769 | | Korean (kor) | 1.000000 | 1.000000 | 1.000000 | | Karachay-Balkar (krc) | 1.000000 | 0.982143 | 0.990991 | | Ripuarisch (ksh) | 1.000000 | 0.964286 | 0.981818 | | Kurdish (kur) | 1.000000 | 0.964286 | 0.981818 | | Ladino (lad) | 1.000000 | 1.000000 | 1.000000 | | Lao (lao) | 0.961538 | 0.909091 | 0.934579 | | Latin (lat) | 0.877193 | 0.943396 | 0.909091 | | Latvian (lav) | 0.963636 | 0.946429 | 0.954955 | | Lezghian (lez) | 1.000000 | 0.964286 | 0.981818 | | Ligurian (lij) | 1.000000 | 0.964286 | 0.981818 | | Limburgan (lim) | 0.938776 | 1.000000 | 0.968421 | | Lingala (lin) | 0.980769 | 0.927273 | 0.953271 | | Lithuanian (lit) | 0.982456 | 1.000000 | 0.991150 | | Lombard (lmo) | 1.000000 | 1.000000 | 1.000000 | | Northern Luri (lrc) | 1.000000 | 0.928571 | 0.962963 | | Latgalian (ltg) | 1.000000 | 0.982143 | 0.990991 | | Luxembourgish (ltz) | 0.949153 | 1.000000 | 0.973913 | | Luganda (lug) | 1.000000 | 1.000000 | 1.000000 | | Literary Chinese (lzh) | 1.000000 | 1.000000 | 1.000000 | | Maithili (mai) | 0.931034 | 0.964286 | 0.947368 | | Malayalam (mal) | 1.000000 | 0.982143 | 0.990991 | | Banyumasan (map-bms) | 0.977778 | 0.785714 | 0.871287 | | Marathi (mar) | 0.949153 | 1.000000 | 0.973913 | | Moksha (mdf) | 0.980000 | 0.890909 | 0.933333 | | Eastern Mari (mhr) | 0.981818 | 0.964286 | 0.972973 | | Minangkabau (min) | 1.000000 | 1.000000 | 1.000000 | | Macedonian (mkd) | 1.000000 | 0.981818 | 0.990826 | | Malagasy (mlg) | 0.981132 | 1.000000 | 0.990476 | | Maltese (mlt) | 0.982456 | 1.000000 | 0.991150 | | Min Nan Chinese (nan) | 1.000000 | 1.000000 | 1.000000 | | Mongolian (mon) | 1.000000 | 0.981818 | 0.990826 | | Maori (mri) | 1.000000 | 1.000000 | 1.000000 | | Western Mari (mrj) | 0.982456 | 1.000000 | 0.991150 | | Malay (msa) | 0.862069 | 0.892857 | 0.877193 | | Mirandese (mwl) | 1.000000 | 0.982143 | 0.990991 | | Burmese (mya) | 1.000000 | 1.000000 | 1.000000 | | Erzya (myv) | 0.818182 | 0.964286 | 0.885246 | | Mazanderani (mzn) | 0.981481 | 1.000000 | 0.990654 | | Neapolitan (nap) | 1.000000 | 0.981818 | 0.990826 | | Navajo (nav) | 1.000000 | 1.000000 | 1.000000 | | Classical Nahuatl (nci) | 0.981481 | 0.946429 | 0.963636 | | Low German (nds) | 0.982143 | 0.982143 | 0.982143 | | West Low German (nds-nl) | 1.000000 | 1.000000 | 1.000000 | | Nepali (macrolanguage) (nep) | 0.881356 | 0.928571 | 0.904348 | | Newari (new) | 1.000000 | 0.909091 | 0.952381 | | Dutch (nld) | 0.982143 | 0.982143 | 0.982143 | | Norwegian Nynorsk (nno) | 1.000000 | 1.000000 | 1.000000 | | Bokmål (nob) | 1.000000 | 1.000000 | 1.000000 | | Narom (nrm) | 0.981818 | 0.964286 | 0.972973 | | Northern Sotho (nso) | 1.000000 | 1.000000 | 1.000000 | | Occitan (oci) | 0.903846 | 0.839286 | 0.870370 | | Livvi-Karelian (olo) | 0.982456 | 1.000000 | 0.991150 | | Oriya (ori) | 0.964912 | 0.982143 | 0.973451 | | Oromo (orm) | 0.982143 | 0.982143 | 0.982143 | | Ossetian (oss) | 0.982143 | 1.000000 | 0.990991 | | Pangasinan (pag) | 0.980000 | 0.875000 | 0.924528 | | Pampanga (pam) | 0.928571 | 0.896552 | 0.912281 | | Panjabi (pan) | 1.000000 | 1.000000 | 1.000000 | | Papiamento (pap) | 1.000000 | 0.964286 | 0.981818 | | Picard (pcd) | 0.849057 | 0.849057 | 0.849057 | | Pennsylvania German (pdc) | 0.854839 | 0.946429 | 0.898305 | | Palatine German (pfl) | 0.946429 | 0.946429 | 0.946429 | | Western Panjabi (pnb) | 0.981132 | 0.962963 | 0.971963 | | Polish (pol) | 0.933333 | 1.000000 | 0.965517 | | Portuguese (por) | 0.774648 | 0.982143 | 0.866142 | | Pushto (pus) | 1.000000 | 0.910714 | 0.953271 | | Quechua (que) | 0.962963 | 0.928571 | 0.945455 | | Tarantino dialect (roa-tara) | 1.000000 | 0.964286 | 0.981818 | | Romansh (roh) | 1.000000 | 0.928571 | 0.962963 | | Romanian (ron) | 0.965517 | 1.000000 | 0.982456 | | Rusyn (rue) | 0.946429 | 0.946429 | 0.946429 | | Aromanian (rup) | 0.962963 | 0.928571 | 0.945455 | | Russian (rus) | 0.859375 | 0.982143 | 0.916667 | | Yakut (sah) | 1.000000 | 0.982143 | 0.990991 | | Sanskrit (san) | 0.982143 | 0.982143 | 0.982143 | | Sicilian (scn) | 1.000000 | 1.000000 | 1.000000 | | Scots (sco) | 0.982143 | 0.982143 | 0.982143 | | Samogitian (sgs) | 1.000000 | 0.982143 | 0.990991 | | Sinhala (sin) | 0.964912 | 0.982143 | 0.973451 | | Slovak (slk) | 1.000000 | 0.982143 | 0.990991 | | Slovene (slv) | 1.000000 | 0.981818 | 0.990826 | | Northern Sami (sme) | 0.962264 | 0.962264 | 0.962264 | | Shona (sna) | 0.933333 | 1.000000 | 0.965517 | | Sindhi (snd) | 1.000000 | 1.000000 | 1.000000 | | Somali (som) | 0.948276 | 1.000000 | 0.973451 | | Spanish (spa) | 0.739130 | 0.910714 | 0.816000 | | Albanian (sqi) | 0.982143 | 0.982143 | 0.982143 | | Sardinian (srd) | 1.000000 | 0.982143 | 0.990991 | | Sranan (srn) | 1.000000 | 1.000000 | 1.000000 | | Serbian (srp) | 1.000000 | 0.946429 | 0.972477 | | Saterfriesisch (stq) | 1.000000 | 0.964286 | 0.981818 | | Sundanese (sun) | 1.000000 | 0.977273 | 0.988506 | | Swahili (macrolanguage) (swa) | 1.000000 | 1.000000 | 1.000000 | | Swedish (swe) | 1.000000 | 1.000000 | 1.000000 | | Silesian (szl) | 1.000000 | 0.981481 | 0.990654 | | Tamil (tam) | 0.982143 | 1.000000 | 0.990991 | | Tatar (tat) | 1.000000 | 1.000000 | 1.000000 | | Tulu (tcy) | 0.982456 | 1.000000 | 0.991150 | | Telugu (tel) | 1.000000 | 0.920000 | 0.958333 | | Tetum (tet) | 1.000000 | 0.964286 | 0.981818 | | Tajik (tgk) | 1.000000 | 1.000000 | 1.000000 | | Tagalog (tgl) | 1.000000 | 1.000000 | 1.000000 | | Thai (tha) | 0.932203 | 0.982143 | 0.956522 | | Tongan (ton) | 1.000000 | 0.964286 | 0.981818 | | Tswana (tsn) | 1.000000 | 1.000000 | 1.000000 | | Turkmen (tuk) | 1.000000 | 0.982143 | 0.990991 | | Turkish (tur) | 0.901639 | 0.982143 | 0.940171 | | Tuvan (tyv) | 1.000000 | 0.964286 | 0.981818 | | Udmurt (udm) | 1.000000 | 0.982143 | 0.990991 | | Uighur (uig) | 1.000000 | 0.982143 | 0.990991 | | Ukrainian (ukr) | 0.963636 | 0.946429 | 0.954955 | | Urdu (urd) | 1.000000 | 0.982143 | 0.990991 | | Uzbek (uzb) | 1.000000 | 1.000000 | 1.000000 | | Venetian (vec) | 1.000000 | 0.982143 | 0.990991 | | Veps (vep) | 0.982456 | 1.000000 | 0.991150 | | Vietnamese (vie) | 0.964912 | 0.982143 | 0.973451 | | Vlaams (vls) | 1.000000 | 0.982143 | 0.990991 | | Volapük (vol) | 1.000000 | 1.000000 | 1.000000 | | Võro (vro) | 0.964286 | 0.964286 | 0.964286 | | Waray (war) | 1.000000 | 0.982143 | 0.990991 | | Walloon (wln) | 1.000000 | 1.000000 | 1.000000 | | Wolof (wol) | 0.981481 | 0.963636 | 0.972477 | | Wu Chinese (wuu) | 0.981481 | 0.946429 | 0.963636 | | Xhosa (xho) | 1.000000 | 0.964286 | 0.981818 | | Mingrelian (xmf) | 1.000000 | 0.964286 | 0.981818 | | Yiddish (yid) | 1.000000 | 1.000000 | 1.000000 | | Yoruba (yor) | 0.964912 | 0.982143 | 0.973451 | | Zeeuws (zea) | 1.000000 | 0.982143 | 0.990991 | | Cantonese (zh-yue) | 0.981481 | 0.946429 | 0.963636 | | Standard Chinese (zho) | 0.932203 | 0.982143 | 0.956522 | | accuracy | 0.963055 | 0.963055 | 0.963055 | | macro avg | 0.966424 | 0.963216 | 0.963891 | | weighted avg | 0.966040 | 0.963055 | 0.963606 | ### By Sentence | language | precision | recall | f1-score | |:--------------------------------------:|:---------:|:--------:|:--------:| | Achinese (ace) | 0.754545 | 0.873684 | 0.809756 | | Afrikaans (afr) | 0.708955 | 0.940594 | 0.808511 | | Alemannic German (als) | 0.870130 | 0.752809 | 0.807229 | | Amharic (amh) | 1.000000 | 0.820000 | 0.901099 | | Old English (ang) | 0.966667 | 0.906250 | 0.935484 | | Arabic (ara) | 0.907692 | 0.967213 | 0.936508 | | Aragonese (arg) | 0.921569 | 0.959184 | 0.940000 | | Egyptian Arabic (arz) | 0.964286 | 0.843750 | 0.900000 | | Assamese (asm) | 0.964286 | 0.870968 | 0.915254 | | Asturian (ast) | 0.880000 | 0.795181 | 0.835443 | | Avar (ava) | 0.864198 | 0.843373 | 0.853659 | | Aymara (aym) | 1.000000 | 0.901961 | 0.948454 | | South Azerbaijani (azb) | 0.979381 | 0.989583 | 0.984456 | | Azerbaijani (aze) | 0.989899 | 0.960784 | 0.975124 | | Bashkir (bak) | 0.837209 | 0.857143 | 0.847059 | | Bavarian (bar) | 0.741935 | 0.766667 | 0.754098 | | Central Bikol (bcl) | 0.962963 | 0.928571 | 0.945455 | | Belarusian (Taraschkewiza) (be-tarask) | 0.857143 | 0.733333 | 0.790419 | | Belarusian (bel) | 0.775510 | 0.752475 | 0.763819 | | Bengali (ben) | 0.861111 | 0.911765 | 0.885714 | | Bhojpuri (bho) | 0.965517 | 0.933333 | 0.949153 | | Banjar (bjn) | 0.891566 | 0.880952 | 0.886228 | | Tibetan (bod) | 1.000000 | 1.000000 | 1.000000 | | Bosnian (bos) | 0.375000 | 0.323077 | 0.347107 | | Bishnupriya (bpy) | 0.986301 | 1.000000 | 0.993103 | | Breton (bre) | 0.951613 | 0.893939 | 0.921875 | | Bulgarian (bul) | 0.945055 | 0.877551 | 0.910053 | | Buryat (bxr) | 0.955556 | 0.843137 | 0.895833 | | Catalan (cat) | 0.692308 | 0.750000 | 0.720000 | | Chavacano (cbk) | 0.842857 | 0.641304 | 0.728395 | | Min Dong (cdo) | 0.972973 | 1.000000 | 0.986301 | | Cebuano (ceb) | 0.981308 | 0.954545 | 0.967742 | | Czech (ces) | 0.944444 | 0.915385 | 0.929687 | | Chechen (che) | 0.875000 | 0.700000 | 0.777778 | | Cherokee (chr) | 1.000000 | 0.970588 | 0.985075 | | Chuvash (chv) | 0.875000 | 0.836957 | 0.855556 | | Central Kurdish (ckb) | 1.000000 | 0.983051 | 0.991453 | | Cornish (cor) | 0.979592 | 0.969697 | 0.974619 | | Corsican (cos) | 0.986842 | 0.925926 | 0.955414 | | Crimean Tatar (crh) | 0.958333 | 0.907895 | 0.932432 | | Kashubian (csb) | 0.920354 | 0.904348 | 0.912281 | | Welsh (cym) | 0.971014 | 0.943662 | 0.957143 | | Danish (dan) | 0.865169 | 0.777778 | 0.819149 | | German (deu) | 0.721311 | 0.822430 | 0.768559 | | Dimli (diq) | 0.915966 | 0.923729 | 0.919831 | | Dhivehi (div) | 1.000000 | 0.991228 | 0.995595 | | Lower Sorbian (dsb) | 0.898876 | 0.879121 | 0.888889 | | Doteli (dty) | 0.821429 | 0.638889 | 0.718750 | | Emilian (egl) | 0.988095 | 0.922222 | 0.954023 | | Modern Greek (ell) | 0.988636 | 0.966667 | 0.977528 | | English (eng) | 0.522727 | 0.784091 | 0.627273 | | Esperanto (epo) | 0.963855 | 0.930233 | 0.946746 | | Estonian (est) | 0.922222 | 0.873684 | 0.897297 | | Basque (eus) | 1.000000 | 0.941176 | 0.969697 | | Extremaduran (ext) | 0.925373 | 0.885714 | 0.905109 | | Faroese (fao) | 0.855072 | 0.887218 | 0.870849 | | Persian (fas) | 0.879630 | 0.979381 | 0.926829 | | Finnish (fin) | 0.952830 | 0.943925 | 0.948357 | | French (fra) | 0.676768 | 0.943662 | 0.788235 | | Arpitan (frp) | 0.867925 | 0.807018 | 0.836364 | | Western Frisian (fry) | 0.956989 | 0.890000 | 0.922280 | | Friulian (fur) | 1.000000 | 0.857143 | 0.923077 | | Gagauz (gag) | 0.939024 | 0.802083 | 0.865169 | | Scottish Gaelic (gla) | 1.000000 | 0.879121 | 0.935673 | | Irish (gle) | 0.989247 | 0.958333 | 0.973545 | | Galician (glg) | 0.910256 | 0.922078 | 0.916129 | | Gilaki (glk) | 0.964706 | 0.872340 | 0.916201 | | Manx (glv) | 1.000000 | 0.965517 | 0.982456 | | Guarani (grn) | 0.983333 | 1.000000 | 0.991597 | | Gujarati (guj) | 1.000000 | 0.991525 | 0.995745 | | Hakka Chinese (hak) | 0.955224 | 0.955224 | 0.955224 | | Haitian Creole (hat) | 0.833333 | 0.666667 | 0.740741 | | Hausa (hau) | 0.936709 | 0.913580 | 0.925000 | | Serbo-Croatian (hbs) | 0.452830 | 0.410256 | 0.430493 | | Hebrew (heb) | 0.988235 | 0.976744 | 0.982456 | | Fiji Hindi (hif) | 0.936709 | 0.840909 | 0.886228 | | Hindi (hin) | 0.965517 | 0.756757 | 0.848485 | | Croatian (hrv) | 0.443820 | 0.537415 | 0.486154 | | Upper Sorbian (hsb) | 0.951613 | 0.830986 | 0.887218 | | Hungarian (hun) | 0.854701 | 0.909091 | 0.881057 | | Armenian (hye) | 1.000000 | 0.816327 | 0.898876 | | Igbo (ibo) | 0.974359 | 0.926829 | 0.950000 | | Ido (ido) | 0.975000 | 0.987342 | 0.981132 | | Interlingue (ile) | 0.880597 | 0.921875 | 0.900763 | | Iloko (ilo) | 0.882353 | 0.821918 | 0.851064 | | Interlingua (ina) | 0.952381 | 0.895522 | 0.923077 | | Indonesian (ind) | 0.606383 | 0.695122 | 0.647727 | | Icelandic (isl) | 0.978261 | 0.882353 | 0.927835 | | Italian (ita) | 0.910448 | 0.910448 | 0.910448 | | Jamaican Patois (jam) | 0.988764 | 0.967033 | 0.977778 | | Javanese (jav) | 0.903614 | 0.862069 | 0.882353 | | Lojban (jbo) | 0.943878 | 0.929648 | 0.936709 | | Japanese (jpn) | 1.000000 | 0.764706 | 0.866667 | | Karakalpak (kaa) | 0.940171 | 0.901639 | 0.920502 | | Kabyle (kab) | 0.985294 | 0.837500 | 0.905405 | | Kannada (kan) | 0.975806 | 0.975806 | 0.975806 | | Georgian (kat) | 0.953704 | 0.903509 | 0.927928 | | Kazakh (kaz) | 0.934579 | 0.877193 | 0.904977 | | Kabardian (kbd) | 0.987952 | 0.953488 | 0.970414 | | Central Khmer (khm) | 0.928571 | 0.829787 | 0.876404 | | Kinyarwanda (kin) | 0.953125 | 0.938462 | 0.945736 | | Kirghiz (kir) | 0.927632 | 0.881250 | 0.903846 | | Komi-Permyak (koi) | 0.750000 | 0.776786 | 0.763158 | | Konkani (kok) | 0.893491 | 0.872832 | 0.883041 | | Komi (kom) | 0.734177 | 0.690476 | 0.711656 | | Korean (kor) | 0.989899 | 0.989899 | 0.989899 | | Karachay-Balkar (krc) | 0.928571 | 0.917647 | 0.923077 | | Ripuarisch (ksh) | 0.915789 | 0.896907 | 0.906250 | | Kurdish (kur) | 0.977528 | 0.935484 | 0.956044 | | Ladino (lad) | 0.985075 | 0.904110 | 0.942857 | | Lao (lao) | 0.896552 | 0.812500 | 0.852459 | | Latin (lat) | 0.741935 | 0.831325 | 0.784091 | | Latvian (lav) | 0.710526 | 0.878049 | 0.785455 | | Lezghian (lez) | 0.975309 | 0.877778 | 0.923977 | | Ligurian (lij) | 0.951807 | 0.897727 | 0.923977 | | Limburgan (lim) | 0.909091 | 0.921053 | 0.915033 | | Lingala (lin) | 0.942857 | 0.814815 | 0.874172 | | Lithuanian (lit) | 0.892857 | 0.925926 | 0.909091 | | Lombard (lmo) | 0.766234 | 0.951613 | 0.848921 | | Northern Luri (lrc) | 0.972222 | 0.875000 | 0.921053 | | Latgalian (ltg) | 0.895349 | 0.865169 | 0.880000 | | Luxembourgish (ltz) | 0.882353 | 0.750000 | 0.810811 | | Luganda (lug) | 0.946429 | 0.883333 | 0.913793 | | Literary Chinese (lzh) | 1.000000 | 1.000000 | 1.000000 | | Maithili (mai) | 0.893617 | 0.823529 | 0.857143 | | Malayalam (mal) | 1.000000 | 0.975000 | 0.987342 | | Banyumasan (map-bms) | 0.924242 | 0.772152 | 0.841379 | | Marathi (mar) | 0.874126 | 0.919118 | 0.896057 | | Moksha (mdf) | 0.771242 | 0.830986 | 0.800000 | | Eastern Mari (mhr) | 0.820000 | 0.860140 | 0.839590 | | Minangkabau (min) | 0.973684 | 0.973684 | 0.973684 | | Macedonian (mkd) | 0.895652 | 0.953704 | 0.923767 | | Malagasy (mlg) | 1.000000 | 0.966102 | 0.982759 | | Maltese (mlt) | 0.987952 | 0.964706 | 0.976190 | | Min Nan Chinese (nan) | 0.975000 | 1.000000 | 0.987342 | | Mongolian (mon) | 0.954545 | 0.933333 | 0.943820 | | Maori (mri) | 0.985294 | 1.000000 | 0.992593 | | Western Mari (mrj) | 0.966292 | 0.914894 | 0.939891 | | Malay (msa) | 0.770270 | 0.695122 | 0.730769 | | Mirandese (mwl) | 0.970588 | 0.891892 | 0.929577 | | Burmese (mya) | 1.000000 | 0.964286 | 0.981818 | | Erzya (myv) | 0.535714 | 0.681818 | 0.600000 | | Mazanderani (mzn) | 0.968750 | 0.898551 | 0.932331 | | Neapolitan (nap) | 0.892308 | 0.865672 | 0.878788 | | Navajo (nav) | 0.984375 | 0.984375 | 0.984375 | | Classical Nahuatl (nci) | 0.901408 | 0.761905 | 0.825806 | | Low German (nds) | 0.896226 | 0.913462 | 0.904762 | | West Low German (nds-nl) | 0.873563 | 0.835165 | 0.853933 | | Nepali (macrolanguage) (nep) | 0.704545 | 0.861111 | 0.775000 | | Newari (new) | 0.920000 | 0.741935 | 0.821429 | | Dutch (nld) | 0.925926 | 0.872093 | 0.898204 | | Norwegian Nynorsk (nno) | 0.847059 | 0.808989 | 0.827586 | | Bokmål (nob) | 0.861386 | 0.852941 | 0.857143 | | Narom (nrm) | 0.966667 | 0.983051 | 0.974790 | | Northern Sotho (nso) | 0.897436 | 0.921053 | 0.909091 | | Occitan (oci) | 0.958333 | 0.696970 | 0.807018 | | Livvi-Karelian (olo) | 0.967742 | 0.937500 | 0.952381 | | Oriya (ori) | 0.933333 | 1.000000 | 0.965517 | | Oromo (orm) | 0.977528 | 0.915789 | 0.945652 | | Ossetian (oss) | 0.958333 | 0.841463 | 0.896104 | | Pangasinan (pag) | 0.847328 | 0.909836 | 0.877470 | | Pampanga (pam) | 0.969697 | 0.780488 | 0.864865 | | Panjabi (pan) | 1.000000 | 1.000000 | 1.000000 | | Papiamento (pap) | 0.876190 | 0.920000 | 0.897561 | | Picard (pcd) | 0.707317 | 0.568627 | 0.630435 | | Pennsylvania German (pdc) | 0.827273 | 0.827273 | 0.827273 | | Palatine German (pfl) | 0.882353 | 0.914634 | 0.898204 | | Western Panjabi (pnb) | 0.964286 | 0.931034 | 0.947368 | | Polish (pol) | 0.859813 | 0.910891 | 0.884615 | | Portuguese (por) | 0.535714 | 0.833333 | 0.652174 | | Pushto (pus) | 0.989362 | 0.902913 | 0.944162 | | Quechua (que) | 0.979167 | 0.903846 | 0.940000 | | Tarantino dialect (roa-tara) | 0.964912 | 0.901639 | 0.932203 | | Romansh (roh) | 0.914894 | 0.895833 | 0.905263 | | Romanian (ron) | 0.880597 | 0.880597 | 0.880597 | | Rusyn (rue) | 0.932584 | 0.805825 | 0.864583 | | Aromanian (rup) | 0.783333 | 0.758065 | 0.770492 | | Russian (rus) | 0.517986 | 0.765957 | 0.618026 | | Yakut (sah) | 0.954023 | 0.922222 | 0.937853 | | Sanskrit (san) | 0.866667 | 0.951220 | 0.906977 | | Sicilian (scn) | 0.984375 | 0.940299 | 0.961832 | | Scots (sco) | 0.851351 | 0.900000 | 0.875000 | | Samogitian (sgs) | 0.977011 | 0.876289 | 0.923913 | | Sinhala (sin) | 0.406154 | 0.985075 | 0.575163 | | Slovak (slk) | 0.956989 | 0.872549 | 0.912821 | | Slovene (slv) | 0.907216 | 0.854369 | 0.880000 | | Northern Sami (sme) | 0.949367 | 0.892857 | 0.920245 | | Shona (sna) | 0.936508 | 0.855072 | 0.893939 | | Sindhi (snd) | 0.984962 | 0.992424 | 0.988679 | | Somali (som) | 0.949153 | 0.848485 | 0.896000 | | Spanish (spa) | 0.584158 | 0.746835 | 0.655556 | | Albanian (sqi) | 0.988095 | 0.912088 | 0.948571 | | Sardinian (srd) | 0.957746 | 0.931507 | 0.944444 | | Sranan (srn) | 0.985714 | 0.945205 | 0.965035 | | Serbian (srp) | 0.950980 | 0.889908 | 0.919431 | | Saterfriesisch (stq) | 0.962500 | 0.875000 | 0.916667 | | Sundanese (sun) | 0.778846 | 0.910112 | 0.839378 | | Swahili (macrolanguage) (swa) | 0.915493 | 0.878378 | 0.896552 | | Swedish (swe) | 0.989247 | 0.958333 | 0.973545 | | Silesian (szl) | 0.944444 | 0.904255 | 0.923913 | | Tamil (tam) | 0.990000 | 0.970588 | 0.980198 | | Tatar (tat) | 0.942029 | 0.902778 | 0.921986 | | Tulu (tcy) | 0.980519 | 0.967949 | 0.974194 | | Telugu (tel) | 0.965986 | 0.965986 | 0.965986 | | Tetum (tet) | 0.898734 | 0.855422 | 0.876543 | | Tajik (tgk) | 0.974684 | 0.939024 | 0.956522 | | Tagalog (tgl) | 0.965909 | 0.934066 | 0.949721 | | Thai (tha) | 0.923077 | 0.882353 | 0.902256 | | Tongan (ton) | 0.970149 | 0.890411 | 0.928571 | | Tswana (tsn) | 0.888889 | 0.926316 | 0.907216 | | Turkmen (tuk) | 0.968000 | 0.889706 | 0.927203 | | Turkish (tur) | 0.871287 | 0.926316 | 0.897959 | | Tuvan (tyv) | 0.948454 | 0.859813 | 0.901961 | | Udmurt (udm) | 0.989362 | 0.894231 | 0.939394 | | Uighur (uig) | 1.000000 | 0.953333 | 0.976109 | | Ukrainian (ukr) | 0.893617 | 0.875000 | 0.884211 | | Urdu (urd) | 1.000000 | 1.000000 | 1.000000 | | Uzbek (uzb) | 0.636042 | 0.886700 | 0.740741 | | Venetian (vec) | 1.000000 | 0.941176 | 0.969697 | | Veps (vep) | 0.858586 | 0.965909 | 0.909091 | | Vietnamese (vie) | 1.000000 | 0.940476 | 0.969325 | | Vlaams (vls) | 0.885714 | 0.898551 | 0.892086 | | Volapük (vol) | 0.975309 | 0.975309 | 0.975309 | | Võro (vro) | 0.855670 | 0.864583 | 0.860104 | | Waray (war) | 0.972222 | 0.909091 | 0.939597 | | Walloon (wln) | 0.742138 | 0.893939 | 0.810997 | | Wolof (wol) | 0.882979 | 0.954023 | 0.917127 | | Wu Chinese (wuu) | 0.961538 | 0.833333 | 0.892857 | | Xhosa (xho) | 0.934066 | 0.867347 | 0.899471 | | Mingrelian (xmf) | 0.958333 | 0.929293 | 0.943590 | | Yiddish (yid) | 0.984375 | 0.875000 | 0.926471 | | Yoruba (yor) | 0.868421 | 0.857143 | 0.862745 | | Zeeuws (zea) | 0.879518 | 0.793478 | 0.834286 | | Cantonese (zh-yue) | 0.896552 | 0.812500 | 0.852459 | | Standard Chinese (zho) | 0.906250 | 0.935484 | 0.920635 | | accuracy | 0.881051 | 0.881051 | 0.881051 | | macro avg | 0.903245 | 0.880618 | 0.888996 | | weighted avg | 0.894174 | 0.881051 | 0.884520 | ### By Token (3 to 5) | language | precision | recall | f1-score | |:--------------------------------------:|:---------:|:--------:|:--------:| | Achinese (ace) | 0.873846 | 0.827988 | 0.850299 | | Afrikaans (afr) | 0.638060 | 0.732334 | 0.681954 | | Alemannic German (als) | 0.673780 | 0.547030 | 0.603825 | | Amharic (amh) | 0.997743 | 0.954644 | 0.975717 | | Old English (ang) | 0.840816 | 0.693603 | 0.760148 | | Arabic (ara) | 0.768737 | 0.840749 | 0.803132 | | Aragonese (arg) | 0.493671 | 0.505181 | 0.499360 | | Egyptian Arabic (arz) | 0.823529 | 0.741935 | 0.780606 | | Assamese (asm) | 0.948454 | 0.893204 | 0.920000 | | Asturian (ast) | 0.490000 | 0.508299 | 0.498982 | | Avar (ava) | 0.813636 | 0.655678 | 0.726166 | | Aymara (aym) | 0.795833 | 0.779592 | 0.787629 | | South Azerbaijani (azb) | 0.832836 | 0.863777 | 0.848024 | | Azerbaijani (aze) | 0.867470 | 0.800000 | 0.832370 | | Bashkir (bak) | 0.851852 | 0.750000 | 0.797688 | | Bavarian (bar) | 0.560897 | 0.522388 | 0.540958 | | Central Bikol (bcl) | 0.708229 | 0.668235 | 0.687651 | | Belarusian (Taraschkewiza) (be-tarask) | 0.615635 | 0.526462 | 0.567568 | | Belarusian (bel) | 0.539952 | 0.597855 | 0.567430 | | Bengali (ben) | 0.830275 | 0.885086 | 0.856805 | | Bhojpuri (bho) | 0.723118 | 0.691517 | 0.706965 | | Banjar (bjn) | 0.619586 | 0.726269 | 0.668699 | | Tibetan (bod) | 0.999537 | 0.991728 | 0.995617 | | Bosnian (bos) | 0.330849 | 0.403636 | 0.363636 | | Bishnupriya (bpy) | 0.941634 | 0.949020 | 0.945312 | | Breton (bre) | 0.772222 | 0.745308 | 0.758527 | | Bulgarian (bul) | 0.771505 | 0.706897 | 0.737789 | | Buryat (bxr) | 0.741935 | 0.753149 | 0.747500 | | Catalan (cat) | 0.528716 | 0.610136 | 0.566516 | | Chavacano (cbk) | 0.409449 | 0.312625 | 0.354545 | | Min Dong (cdo) | 0.951264 | 0.936057 | 0.943599 | | Cebuano (ceb) | 0.888298 | 0.876640 | 0.882431 | | Czech (ces) | 0.806045 | 0.758294 | 0.781441 | | Chechen (che) | 0.857143 | 0.600000 | 0.705882 | | Cherokee (chr) | 0.997840 | 0.952577 | 0.974684 | | Chuvash (chv) | 0.874346 | 0.776744 | 0.822660 | | Central Kurdish (ckb) | 0.984848 | 0.953545 | 0.968944 | | Cornish (cor) | 0.747596 | 0.807792 | 0.776529 | | Corsican (cos) | 0.673913 | 0.708571 | 0.690808 | | Crimean Tatar (crh) | 0.498801 | 0.700337 | 0.582633 | | Kashubian (csb) | 0.797059 | 0.794721 | 0.795888 | | Welsh (cym) | 0.829609 | 0.841360 | 0.835443 | | Danish (dan) | 0.649789 | 0.622222 | 0.635707 | | German (deu) | 0.559406 | 0.763514 | 0.645714 | | Dimli (diq) | 0.835580 | 0.763547 | 0.797941 | | Dhivehi (div) | 1.000000 | 0.980645 | 0.990228 | | Lower Sorbian (dsb) | 0.740484 | 0.694805 | 0.716918 | | Doteli (dty) | 0.616314 | 0.527132 | 0.568245 | | Emilian (egl) | 0.822993 | 0.769625 | 0.795414 | | Modern Greek (ell) | 0.972043 | 0.963753 | 0.967880 | | English (eng) | 0.260492 | 0.724346 | 0.383183 | | Esperanto (epo) | 0.766764 | 0.716621 | 0.740845 | | Estonian (est) | 0.698885 | 0.673835 | 0.686131 | | Basque (eus) | 0.882716 | 0.841176 | 0.861446 | | Extremaduran (ext) | 0.570605 | 0.511628 | 0.539510 | | Faroese (fao) | 0.773987 | 0.784017 | 0.778970 | | Persian (fas) | 0.709836 | 0.809346 | 0.756332 | | Finnish (fin) | 0.866261 | 0.796089 | 0.829694 | | French (fra) | 0.496263 | 0.700422 | 0.580927 | | Arpitan (frp) | 0.663366 | 0.584302 | 0.621329 | | Western Frisian (fry) | 0.750000 | 0.756148 | 0.753061 | | Friulian (fur) | 0.713555 | 0.675545 | 0.694030 | | Gagauz (gag) | 0.728125 | 0.677326 | 0.701807 | | Scottish Gaelic (gla) | 0.831601 | 0.817996 | 0.824742 | | Irish (gle) | 0.868852 | 0.801296 | 0.833708 | | Galician (glg) | 0.469816 | 0.454315 | 0.461935 | | Gilaki (glk) | 0.703883 | 0.687204 | 0.695444 | | Manx (glv) | 0.873047 | 0.886905 | 0.879921 | | Guarani (grn) | 0.848580 | 0.793510 | 0.820122 | | Gujarati (guj) | 0.995643 | 0.926978 | 0.960084 | | Hakka Chinese (hak) | 0.898403 | 0.904971 | 0.901675 | | Haitian Creole (hat) | 0.719298 | 0.518987 | 0.602941 | | Hausa (hau) | 0.815353 | 0.829114 | 0.822176 | | Serbo-Croatian (hbs) | 0.343465 | 0.244589 | 0.285714 | | Hebrew (heb) | 0.891304 | 0.933941 | 0.912125 | | Fiji Hindi (hif) | 0.662577 | 0.664615 | 0.663594 | | Hindi (hin) | 0.782301 | 0.778169 | 0.780229 | | Croatian (hrv) | 0.360308 | 0.374000 | 0.367026 | | Upper Sorbian (hsb) | 0.745763 | 0.611111 | 0.671756 | | Hungarian (hun) | 0.876812 | 0.846154 | 0.861210 | | Armenian (hye) | 0.988201 | 0.917808 | 0.951705 | | Igbo (ibo) | 0.825397 | 0.696429 | 0.755448 | | Ido (ido) | 0.760479 | 0.814103 | 0.786378 | | Interlingue (ile) | 0.701299 | 0.580645 | 0.635294 | | Iloko (ilo) | 0.688356 | 0.844538 | 0.758491 | | Interlingua (ina) | 0.577889 | 0.588235 | 0.583016 | | Indonesian (ind) | 0.415879 | 0.514019 | 0.459770 | | Icelandic (isl) | 0.855263 | 0.790754 | 0.821745 | | Italian (ita) | 0.474576 | 0.561247 | 0.514286 | | Jamaican Patois (jam) | 0.826087 | 0.791667 | 0.808511 | | Javanese (jav) | 0.670130 | 0.658163 | 0.664093 | | Lojban (jbo) | 0.896861 | 0.917431 | 0.907029 | | Japanese (jpn) | 0.931373 | 0.848214 | 0.887850 | | Karakalpak (kaa) | 0.790393 | 0.827744 | 0.808637 | | Kabyle (kab) | 0.828571 | 0.759162 | 0.792350 | | Kannada (kan) | 0.879357 | 0.847545 | 0.863158 | | Georgian (kat) | 0.916399 | 0.907643 | 0.912000 | | Kazakh (kaz) | 0.900901 | 0.819672 | 0.858369 | | Kabardian (kbd) | 0.923345 | 0.892256 | 0.907534 | | Central Khmer (khm) | 0.976667 | 0.816156 | 0.889226 | | Kinyarwanda (kin) | 0.824324 | 0.726190 | 0.772152 | | Kirghiz (kir) | 0.674766 | 0.779698 | 0.723447 | | Komi-Permyak (koi) | 0.652830 | 0.633700 | 0.643123 | | Konkani (kok) | 0.778865 | 0.728938 | 0.753075 | | Komi (kom) | 0.737374 | 0.572549 | 0.644592 | | Korean (kor) | 0.984615 | 0.967603 | 0.976035 | | Karachay-Balkar (krc) | 0.869416 | 0.857627 | 0.863481 | | Ripuarisch (ksh) | 0.709859 | 0.649485 | 0.678331 | | Kurdish (kur) | 0.883777 | 0.862884 | 0.873206 | | Ladino (lad) | 0.660920 | 0.576441 | 0.615797 | | Lao (lao) | 0.986175 | 0.918455 | 0.951111 | | Latin (lat) | 0.581250 | 0.636986 | 0.607843 | | Latvian (lav) | 0.824513 | 0.797844 | 0.810959 | | Lezghian (lez) | 0.898955 | 0.793846 | 0.843137 | | Ligurian (lij) | 0.662903 | 0.677100 | 0.669927 | | Limburgan (lim) | 0.615385 | 0.581818 | 0.598131 | | Lingala (lin) | 0.836207 | 0.763780 | 0.798354 | | Lithuanian (lit) | 0.756329 | 0.804714 | 0.779772 | | Lombard (lmo) | 0.556818 | 0.536986 | 0.546722 | | Northern Luri (lrc) | 0.838574 | 0.753296 | 0.793651 | | Latgalian (ltg) | 0.759531 | 0.755102 | 0.757310 | | Luxembourgish (ltz) | 0.645062 | 0.614706 | 0.629518 | | Luganda (lug) | 0.787535 | 0.805797 | 0.796562 | | Literary Chinese (lzh) | 0.921951 | 0.949749 | 0.935644 | | Maithili (mai) | 0.777778 | 0.761658 | 0.769634 | | Malayalam (mal) | 0.993377 | 0.949367 | 0.970874 | | Banyumasan (map-bms) | 0.531429 | 0.453659 | 0.489474 | | Marathi (mar) | 0.748744 | 0.818681 | 0.782152 | | Moksha (mdf) | 0.728745 | 0.800000 | 0.762712 | | Eastern Mari (mhr) | 0.790323 | 0.760870 | 0.775316 | | Minangkabau (min) | 0.953271 | 0.886957 | 0.918919 | | Macedonian (mkd) | 0.816399 | 0.849722 | 0.832727 | | Malagasy (mlg) | 0.925187 | 0.918317 | 0.921739 | | Maltese (mlt) | 0.869421 | 0.890017 | 0.879599 | | Min Nan Chinese (nan) | 0.743707 | 0.820707 | 0.780312 | | Mongolian (mon) | 0.852194 | 0.838636 | 0.845361 | | Maori (mri) | 0.934726 | 0.937173 | 0.935948 | | Western Mari (mrj) | 0.818792 | 0.827119 | 0.822934 | | Malay (msa) | 0.508065 | 0.376119 | 0.432247 | | Mirandese (mwl) | 0.650407 | 0.685225 | 0.667362 | | Burmese (mya) | 0.995968 | 0.972441 | 0.984064 | | Erzya (myv) | 0.475783 | 0.503012 | 0.489019 | | Mazanderani (mzn) | 0.775362 | 0.701639 | 0.736661 | | Neapolitan (nap) | 0.628993 | 0.595349 | 0.611708 | | Navajo (nav) | 0.955882 | 0.937500 | 0.946602 | | Classical Nahuatl (nci) | 0.679758 | 0.589005 | 0.631136 | | Low German (nds) | 0.669789 | 0.690821 | 0.680143 | | West Low German (nds-nl) | 0.513889 | 0.504545 | 0.509174 | | Nepali (macrolanguage) (nep) | 0.640476 | 0.649758 | 0.645084 | | Newari (new) | 0.928571 | 0.745902 | 0.827273 | | Dutch (nld) | 0.553763 | 0.553763 | 0.553763 | | Norwegian Nynorsk (nno) | 0.569277 | 0.519231 | 0.543103 | | Bokmål (nob) | 0.519856 | 0.562500 | 0.540338 | | Narom (nrm) | 0.691275 | 0.605882 | 0.645768 | | Northern Sotho (nso) | 0.950276 | 0.815166 | 0.877551 | | Occitan (oci) | 0.483444 | 0.366834 | 0.417143 | | Livvi-Karelian (olo) | 0.816850 | 0.790780 | 0.803604 | | Oriya (ori) | 0.981481 | 0.963636 | 0.972477 | | Oromo (orm) | 0.885714 | 0.829218 | 0.856536 | | Ossetian (oss) | 0.822006 | 0.855219 | 0.838284 | | Pangasinan (pag) | 0.842105 | 0.715655 | 0.773748 | | Pampanga (pam) | 0.770000 | 0.435028 | 0.555957 | | Panjabi (pan) | 0.996154 | 0.984791 | 0.990440 | | Papiamento (pap) | 0.674672 | 0.661670 | 0.668108 | | Picard (pcd) | 0.407895 | 0.356322 | 0.380368 | | Pennsylvania German (pdc) | 0.487047 | 0.509485 | 0.498013 | | Palatine German (pfl) | 0.614173 | 0.570732 | 0.591656 | | Western Panjabi (pnb) | 0.926267 | 0.887417 | 0.906426 | | Polish (pol) | 0.797059 | 0.734417 | 0.764457 | | Portuguese (por) | 0.500914 | 0.586724 | 0.540434 | | Pushto (pus) | 0.941489 | 0.898477 | 0.919481 | | Quechua (que) | 0.854167 | 0.797665 | 0.824950 | | Tarantino dialect (roa-tara) | 0.669794 | 0.724138 | 0.695906 | | Romansh (roh) | 0.745527 | 0.760649 | 0.753012 | | Romanian (ron) | 0.805486 | 0.769048 | 0.786845 | | Rusyn (rue) | 0.718543 | 0.645833 | 0.680251 | | Aromanian (rup) | 0.288482 | 0.730245 | 0.413580 | | Russian (rus) | 0.530120 | 0.690583 | 0.599805 | | Yakut (sah) | 0.853521 | 0.865714 | 0.859574 | | Sanskrit (san) | 0.931343 | 0.896552 | 0.913616 | | Sicilian (scn) | 0.734139 | 0.618321 | 0.671271 | | Scots (sco) | 0.571429 | 0.540816 | 0.555701 | | Samogitian (sgs) | 0.829167 | 0.748120 | 0.786561 | | Sinhala (sin) | 0.909474 | 0.935065 | 0.922092 | | Slovak (slk) | 0.738235 | 0.665782 | 0.700139 | | Slovene (slv) | 0.671123 | 0.662269 | 0.666667 | | Northern Sami (sme) | 0.800676 | 0.825784 | 0.813036 | | Shona (sna) | 0.761702 | 0.724696 | 0.742739 | | Sindhi (snd) | 0.950172 | 0.946918 | 0.948542 | | Somali (som) | 0.849462 | 0.802030 | 0.825065 | | Spanish (spa) | 0.325234 | 0.413302 | 0.364017 | | Albanian (sqi) | 0.875899 | 0.832479 | 0.853637 | | Sardinian (srd) | 0.750000 | 0.711061 | 0.730012 | | Sranan (srn) | 0.888889 | 0.771084 | 0.825806 | | Serbian (srp) | 0.824561 | 0.814356 | 0.819427 | | Saterfriesisch (stq) | 0.790087 | 0.734417 | 0.761236 | | Sundanese (sun) | 0.764192 | 0.631769 | 0.691700 | | Swahili (macrolanguage) (swa) | 0.763496 | 0.796247 | 0.779528 | | Swedish (swe) | 0.838284 | 0.723647 | 0.776758 | | Silesian (szl) | 0.819788 | 0.750809 | 0.783784 | | Tamil (tam) | 0.985765 | 0.955172 | 0.970228 | | Tatar (tat) | 0.469780 | 0.795349 | 0.590674 | | Tulu (tcy) | 0.893300 | 0.873786 | 0.883436 | | Telugu (tel) | 1.000000 | 0.913690 | 0.954899 | | Tetum (tet) | 0.765116 | 0.744344 | 0.754587 | | Tajik (tgk) | 0.828418 | 0.813158 | 0.820717 | | Tagalog (tgl) | 0.751468 | 0.757396 | 0.754420 | | Thai (tha) | 0.933884 | 0.807143 | 0.865900 | | Tongan (ton) | 0.920245 | 0.923077 | 0.921659 | | Tswana (tsn) | 0.873397 | 0.889070 | 0.881164 | | Turkmen (tuk) | 0.898438 | 0.837887 | 0.867107 | | Turkish (tur) | 0.666667 | 0.716981 | 0.690909 | | Tuvan (tyv) | 0.857143 | 0.805063 | 0.830287 | | Udmurt (udm) | 0.865517 | 0.756024 | 0.807074 | | Uighur (uig) | 0.991597 | 0.967213 | 0.979253 | | Ukrainian (ukr) | 0.771341 | 0.702778 | 0.735465 | | Urdu (urd) | 0.877647 | 0.855505 | 0.866434 | | Uzbek (uzb) | 0.655652 | 0.797040 | 0.719466 | | Venetian (vec) | 0.611111 | 0.527233 | 0.566082 | | Veps (vep) | 0.672862 | 0.688213 | 0.680451 | | Vietnamese (vie) | 0.932406 | 0.914230 | 0.923228 | | Vlaams (vls) | 0.594427 | 0.501305 | 0.543909 | | Volapük (vol) | 0.765625 | 0.942308 | 0.844828 | | Võro (vro) | 0.797203 | 0.740260 | 0.767677 | | Waray (war) | 0.930876 | 0.930876 | 0.930876 | | Walloon (wln) | 0.636804 | 0.693931 | 0.664141 | | Wolof (wol) | 0.864220 | 0.845601 | 0.854809 | | Wu Chinese (wuu) | 0.848921 | 0.830986 | 0.839858 | | Xhosa (xho) | 0.837398 | 0.759214 | 0.796392 | | Mingrelian (xmf) | 0.943396 | 0.874126 | 0.907441 | | Yiddish (yid) | 0.955729 | 0.897311 | 0.925599 | | Yoruba (yor) | 0.812010 | 0.719907 | 0.763190 | | Zeeuws (zea) | 0.617737 | 0.550409 | 0.582133 | | Cantonese (zh-yue) | 0.859649 | 0.649007 | 0.739623 | | Standard Chinese (zho) | 0.845528 | 0.781955 | 0.812500 | | accuracy | 0.749527 | 0.749527 | 0.749527 | | macro avg | 0.762866 | 0.742101 | 0.749261 | | weighted avg | 0.762006 | 0.749527 | 0.752910 | ## Questions? Post a Github issue from [HERE](https://github.com/m3hrdadfi/zabanshenas/issues).

manishiitg/output

5ddc5ec9f5bf098c3e9de99c27773112ce34c510

2021-05-20T17:44:39.000Z

[ "pytorch", "jax", "roberta", "fill-mask", "transformers", "autotrain_compatible" ]

fill-mask

false

manishiitg

null

manishiitg/output

15

null

transformers

9,539

Entry not found

mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili

ab8ad3c070d1fea9b697c2870129bca2d4a6f760

2021-11-25T09:04:07.000Z

[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]

token-classification

false

mbeukman

null

mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili

15

null

transformers

9,540

--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-kinyarwanda](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation|Description -|- O|Outside of a named entity B-DATE |Beginning of a DATE entity right after another DATE entity I-DATE |DATE entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organisation right after another organisation I-ORG |Organisation B-LOC |Beginning of a location right after another location I-LOC |Location | Model Name | Staring point | Evaluation / Fine-tune Language | F1 | Precision | Recall | F1 (DATE) | F1 (LOC) | F1 (ORG) | F1 (PER) | | -------------------------------------------------- | -------------------- | -------------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | | [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) (This model) | [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) | [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) | swa | 88.36 | 86.95 | 89.82 | 86.00 | 91.00 | 77.00 | 94.00 | | [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) | [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) | swa | 87.75 | 86.55 | 88.97 | 85.00 | 92.00 | 77.00 | 91.00 | | [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) | [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) | swa | 88.93 | 87.64 | 90.25 | 83.00 | 92.00 | 79.00 | 95.00 | | [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) | [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) | swa | 87.93 | 86.91 | 88.97 | 83.00 | 91.00 | 76.00 | 94.00 | | [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) | [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) | [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) | swa | 90.36 | 88.59 | 92.20 | 86.00 | 93.00 | 79.00 | 96.00 | | [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) | [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) | swa | 87.80 | 86.50 | 89.14 | 86.00 | 90.00 | 78.00 | 93.00 | | [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) | [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) | swa | 87.73 | 86.67 | 88.80 | 85.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) | [base](https://huggingface.co/xlm-roberta-base) | swa | 88.71 | 86.84 | 90.67 | 83.00 | 91.00 | 79.00 | 95.00 | ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```

mbeukman/xlm-roberta-base-finetuned-ner-swahili

d09c7418ffdc293886fb957cc014eadacd60718b

2021-11-25T09:04:40.000Z

[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]

token-classification

false

mbeukman

null

mbeukman/xlm-roberta-base-finetuned-ner-swahili

15

1

transformers

9,541

--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation|Description -|- O|Outside of a named entity B-DATE |Beginning of a DATE entity right after another DATE entity I-DATE |DATE entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organisation right after another organisation I-ORG |Organisation B-LOC |Beginning of a location right after another location I-LOC |Location | Model Name | Staring point | Evaluation / Fine-tune Language | F1 | Precision | Recall | F1 (DATE) | F1 (LOC) | F1 (ORG) | F1 (PER) | | -------------------------------------------------- | -------------------- | -------------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | | [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) (This model) | [base](https://huggingface.co/xlm-roberta-base) | swa | 88.71 | 86.84 | 90.67 | 83.00 | 91.00 | 79.00 | 95.00 | | [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) | [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) | swa | 88.36 | 86.95 | 89.82 | 86.00 | 91.00 | 77.00 | 94.00 | | [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) | [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) | swa | 87.75 | 86.55 | 88.97 | 85.00 | 92.00 | 77.00 | 91.00 | | [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) | [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) | [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) | swa | 88.93 | 87.64 | 90.25 | 83.00 | 92.00 | 79.00 | 95.00 | | [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) | [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) | swa | 87.93 | 86.91 | 88.97 | 83.00 | 91.00 | 76.00 | 94.00 | | [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) | [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) | [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) | swa | 90.36 | 88.59 | 92.20 | 86.00 | 93.00 | 79.00 | 96.00 | | [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) | [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) | swa | 87.80 | 86.50 | 89.14 | 86.00 | 90.00 | 78.00 | 93.00 | | [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) | [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) | swa | 87.73 | 86.67 | 88.80 | 85.00 | 91.00 | 75.00 | 93.00 | ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```

mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo

728511ffd5d6856f5d75db779b2719d53ab75f07

2021-11-25T09:04:58.000Z

[ "pytorch", "xlm-roberta", "token-classification", "luo", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]

token-classification

false

mbeukman

null

mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo

15

null

transformers

9,542

--- language: - luo tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Jii 2 moko jowito ngimagi ka machielo 1 to ohinyore marach mokalo e masira makoch mar apaya mane otimore e apaya mawuok Oyugis kochimo Chabera e sub county ma Rachuonyo East e County ma Homa Bay ewii odhiambo makawuononi" --- # xlm-roberta-base-finetuned-swahili-finetuned-ner-luo This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-swahili](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Luo part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation|Description -|- O|Outside of a named entity B-DATE |Beginning of a DATE entity right after another DATE entity I-DATE |DATE entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organisation right after another organisation I-ORG |Organisation B-LOC |Beginning of a location right after another location I-LOC |Location | Model Name | Staring point | Evaluation / Fine-tune Language | F1 | Precision | Recall | F1 (DATE) | F1 (LOC) | F1 (ORG) | F1 (PER) | | -------------------------------------------------- | -------------------- | -------------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | | [xlm-roberta-base-finetuned-swahili-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo) (This model) | [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) | luo | 78.13 | 77.75 | 78.52 | 65.00 | 82.00 | 61.00 | 89.00 | | [xlm-roberta-base-finetuned-luo-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-luo) | [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) | luo | 78.71 | 78.91 | 78.52 | 72.00 | 84.00 | 59.00 | 87.00 | | [xlm-roberta-base-finetuned-ner-luo](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-luo) | [base](https://huggingface.co/xlm-roberta-base) | luo | 75.99 | 76.18 | 75.80 | 71.00 | 76.00 | 62.00 | 85.00 | ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-luo' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Jii 2 moko jowito ngimagi ka machielo 1 to ohinyore marach mokalo e masira makoch mar apaya mane otimore e apaya mawuok Oyugis kochimo Chabera e sub county ma Rachuonyo East e County ma Homa Bay ewii odhiambo makawuononi" ner_results = nlp(example) print(ner_results) ```

mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili

80a0b4230b7b706e50ce5038e9f18b21f44c1198

2021-11-25T09:05:15.000Z

[ "pytorch", "xlm-roberta", "token-classification", "sw", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "NER", "autotrain_compatible" ]

token-classification

false

mbeukman

null

mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili

15

null

transformers

9,543

--- language: - sw tags: - NER datasets: - masakhaner metrics: - f1 - precision - recall widget: - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." --- # xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili This is a token classification (specifically NER) model that fine-tuned [xlm-roberta-base-finetuned-yoruba](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) on the [MasakhaNER](https://arxiv.org/abs/2103.11811) dataset, specifically the Swahili part. More information, and other similar models can be found in the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). ## About This model is transformer based and was fine-tuned on the MasakhaNER dataset. It is a named entity recognition dataset, containing mostly news articles in 10 different African languages. The model was fine-tuned for 50 epochs, with a maximum sequence length of 200, 32 batch size, 5e-5 learning rate. This process was repeated 5 times (with different random seeds), and this uploaded model performed the best out of those 5 seeds (aggregate F1 on test set). This model was fine-tuned by me, Michael Beukman while doing a project at the University of the Witwatersrand, Johannesburg. This is version 1, as of 20 November 2021. This model is licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0). ### Contact & More information For more information about the models, including training scripts, detailed results and further resources, you can visit the the [main Github repository](https://github.com/Michael-Beukman/NERTransfer). You can contact me by filing an issue on this repository. ### Training Resources In the interest of openness, and reporting resources used, we list here how long the training process took, as well as what the minimum resources would be to reproduce this. Fine-tuning each model on the NER dataset took between 10 and 30 minutes, and was performed on a NVIDIA RTX3090 GPU. To use a batch size of 32, at least 14GB of GPU memory was required, although it was just possible to fit these models in around 6.5GB's of VRAM when using a batch size of 1. ## Data The train, evaluation and test datasets were taken directly from the MasakhaNER [Github](https://github.com/masakhane-io/masakhane-ner) repository, with minimal to no preprocessing, as the original dataset is already of high quality. The motivation for the use of this data is that it is the "first large, publicly available, high quality dataset for named entity recognition (NER) in ten African languages" ([source](https://arxiv.org/pdf/2103.11811.pdf)). The high-quality data, as well as the groundwork laid by the paper introducing it are some more reasons why this dataset was used. For evaluation, the dedicated test split was used, which is from the same distribution as the training data, so this model may not generalise to other distributions, and further testing would need to be done to investigate this. The exact distribution of the data is covered in detail [here](https://arxiv.org/abs/2103.11811). ## Intended Use This model are intended to be used for NLP research into e.g. interpretability or transfer learning. Using this model in production is not supported, as generalisability and downright performance is limited. In particular, this is not designed to be used in any important downstream task that could affect people, as harm could be caused by the limitations of the model, described next. ## Limitations This model was only trained on one (relatively small) dataset, covering one task (NER) in one domain (news articles) and in a set span of time. The results may not generalise, and the model may perform badly, or in an unfair / biased way if used on other tasks. Although the purpose of this project was to investigate transfer learning, the performance on languages that the model was not trained for does suffer. Because this model used xlm-roberta-base as its starting point (potentially with domain adaptive fine-tuning on specific languages), this model's limitations can also apply here. These can include being biased towards the hegemonic viewpoint of most of its training data, being ungrounded and having subpar results on other languages (possibly due to unbalanced training data). As [Adelani et al. (2021)](https://arxiv.org/abs/2103.11811) showed, the models in general struggled with entities that were either longer than 3 words and entities that were not contained in the training data. This could bias the models towards not finding, e.g. names of people that have many words, possibly leading to a misrepresentation in the results. Similarly, names that are uncommon, and may not have been found in the training data (due to e.g. different languages) would also be predicted less often. Additionally, this model has not been verified in practice, and other, more subtle problems may become prevalent if used without any verification that it does what it is supposed to. ### Privacy & Ethical Considerations The data comes from only publicly available news sources, the only available data should cover public figures and those that agreed to be reported on. See the original MasakhaNER paper for more details. No explicit ethical considerations or adjustments were made during fine-tuning of this model. ## Metrics The language adaptive models achieve (mostly) superior performance over starting with xlm-roberta-base. Our main metric was the aggregate F1 score for all NER categories. These metrics are on the test set for MasakhaNER, so the data distribution is similar to the training set, so these results do not directly indicate how well these models generalise. We do find large variation in transfer results when starting from different seeds (5 different seeds were tested), indicating that the fine-tuning process for transfer might be unstable. The metrics used were chosen to be consistent with previous work, and to facilitate research. Other metrics may be more appropriate for other purposes. ## Caveats and Recommendations In general, this model performed worse on the 'date' category compared to others, so if dates are a critical factor, then that might need to be taken into account and addressed, by for example collecting and annotating more data. ## Model Structure Here are some performance details on this specific model, compared to others we trained. All of these metrics were calculated on the test set, and the seed was chosen that gave the best overall F1 score. The first three result columns are averaged over all categories, and the latter 4 provide performance broken down by category. This model can predict the following label for a token ([source](https://huggingface.co/Davlan/xlm-roberta-large-masakhaner)): Abbreviation|Description -|- O|Outside of a named entity B-DATE |Beginning of a DATE entity right after another DATE entity I-DATE |DATE entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organisation right after another organisation I-ORG |Organisation B-LOC |Beginning of a location right after another location I-LOC |Location | Model Name | Staring point | Evaluation / Fine-tune Language | F1 | Precision | Recall | F1 (DATE) | F1 (LOC) | F1 (ORG) | F1 (PER) | | -------------------------------------------------- | -------------------- | -------------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | -------------- | | [xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili) (This model) | [yor](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-yoruba) | swa | 87.73 | 86.67 | 88.80 | 85.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-hausa-finetuned-ner-swahili) | [hau](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-hausa) | swa | 88.36 | 86.95 | 89.82 | 86.00 | 91.00 | 77.00 | 94.00 | | [xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-igbo-finetuned-ner-swahili) | [ibo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-igbo) | swa | 87.75 | 86.55 | 88.97 | 85.00 | 92.00 | 77.00 | 91.00 | | [xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-kinyarwanda-finetuned-ner-swahili) | [kin](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-kinyarwanda) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luganda-finetuned-ner-swahili) | [lug](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luganda) | swa | 88.93 | 87.64 | 90.25 | 83.00 | 92.00 | 79.00 | 95.00 | | [xlm-roberta-base-finetuned-luo-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-luo-finetuned-ner-swahili) | [luo](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-luo) | swa | 87.93 | 86.91 | 88.97 | 83.00 | 91.00 | 76.00 | 94.00 | | [xlm-roberta-base-finetuned-naija-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-naija-finetuned-ner-swahili) | [pcm](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-naija) | swa | 87.26 | 85.15 | 89.48 | 83.00 | 91.00 | 75.00 | 93.00 | | [xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-swahili-finetuned-ner-swahili) | [swa](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-swahili) | swa | 90.36 | 88.59 | 92.20 | 86.00 | 93.00 | 79.00 | 96.00 | | [xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-wolof-finetuned-ner-swahili) | [wol](https://huggingface.co/Davlan/xlm-roberta-base-finetuned-wolof) | swa | 87.80 | 86.50 | 89.14 | 86.00 | 90.00 | 78.00 | 93.00 | | [xlm-roberta-base-finetuned-ner-swahili](https://huggingface.co/mbeukman/xlm-roberta-base-finetuned-ner-swahili) | [base](https://huggingface.co/xlm-roberta-base) | swa | 88.71 | 86.84 | 90.67 | 83.00 | 91.00 | 79.00 | 95.00 | ## Usage To use this model (or others), you can do the following, just changing the model name ([source](https://huggingface.co/dslim/bert-base-NER)): ``` from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline model_name = 'mbeukman/xlm-roberta-base-finetuned-yoruba-finetuned-ner-swahili' tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForTokenClassification.from_pretrained(model_name) nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa , watu takriban 14 zaidi wamepata maambukizi ya Covid - 19 ." ner_results = nlp(example) print(ner_results) ```

meghanabhange/hinglish-sentence-bert

f7e2387d18a11062ab0ba7eb2919ee70d41d3795

2021-05-19T23:17:18.000Z

[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]

fill-mask

false

meghanabhange

null

meghanabhange/hinglish-sentence-bert

15

null

transformers

9,544

Entry not found

midas/gupshup_e2e_pegasus

4ddf0da7354c31ae27cdce2436ba6b87c6d21537

2021-11-14T02:07:37.000Z

[ "pytorch", "pegasus", "text2text-generation", "arxiv:1910.04073", "transformers", "autotrain_compatible" ]

text2text-generation

false

midas

null

midas/gupshup_e2e_pegasus

15

null

transformers

9,545

# Gupshup GupShup: Summarizing Open-Domain Code-Switched Conversations EMNLP 2021 Paper: [https://aclanthology.org/2021.emnlp-main.499.pdf](https://aclanthology.org/2021.emnlp-main.499.pdf) Github: [https://github.com/midas-research/gupshup](https://github.com/midas-research/gupshup) ### Dataset Please request for the Gupshup data using [this Google form](https://docs.google.com/forms/d/1zvUk7WcldVF3RCoHdWzQPzPprtSJClrnHoIOYbzaJEI/edit?ts=61381ec0). Dataset is available for `Hinglish Dilaogues to English Summarization`(h2e) and `English Dialogues to English Summarization`(e2e). For each task, Dialogues/conversastion have `.source`(train.source) as file extension whereas Summary has `.target`(train.target) file extension. ".source" file need to be provided to `input_path` and ".target" file to `reference_path` argument in the scripts. ## Models All model weights are available on the Huggingface model hub. Users can either directly download these weights in their local and provide this path to `model_name` argument in the scripts or use the provided alias (to `model_name` argument) in scripts directly; this will lead to download weights automatically by scripts. Model names were aliased in "gupshup_TASK_MODEL" sense, where "TASK" can be h2e,e2e and MODEL can be mbart, pegasus, etc., as listed below. **1. Hinglish Dialogues to English Summary (h2e)** | Model | Huggingface Alias | |---------|-------------------------------------------------------------------------------| | mBART | [midas/gupshup_h2e_mbart](https://huggingface.co/midas/gupshup_h2e_mbart) | | PEGASUS | [midas/gupshup_h2e_pegasus](https://huggingface.co/midas/gupshup_h2e_pegasus) | | T5 MTL | [midas/gupshup_h2e_t5_mtl](https://huggingface.co/midas/gupshup_h2e_t5_mtl) | | T5 | [midas/gupshup_h2e_t5](https://huggingface.co/midas/gupshup_h2e_t5) | | BART | [midas/gupshup_h2e_bart](https://huggingface.co/midas/gupshup_h2e_bart) | | GPT-2 | [midas/gupshup_h2e_gpt](https://huggingface.co/midas/gupshup_h2e_gpt) | **2. English Dialogues to English Summary (e2e)** | Model | Huggingface Alias | |---------|-------------------------------------------------------------------------------| | mBART | [midas/gupshup_e2e_mbart](https://huggingface.co/midas/gupshup_e2e_mbart) | | PEGASUS | [midas/gupshup_e2e_pegasus](https://huggingface.co/midas/gupshup_e2e_pegasus) | | T5 MTL | [midas/gupshup_e2e_t5_mtl](https://huggingface.co/midas/gupshup_e2e_t5_mtl) | | T5 | [midas/gupshup_e2e_t5](https://huggingface.co/midas/gupshup_e2e_t5) | | BART | [midas/gupshup_e2e_bart](https://huggingface.co/midas/gupshup_e2e_bart) | | GPT-2 | [midas/gupshup_e2e_gpt](https://huggingface.co/midas/gupshup_e2e_gpt) | ## Inference ### Using command line 1. Clone this repo and create a python virtual environment (https://docs.python.org/3/library/venv.html). Install the required packages using ``` git clone https://github.com/midas-research/gupshup.git pip install -r requirements.txt ``` 2. run_eval script has the following arguments. * **model_name** : Path or alias to one of our models available on Huggingface as listed above. * **input_path** : Source file or path to file containing conversations, which will be summarized. * **save_path** : File path where to save summaries generated by the model. * **reference_path** : Target file or path to file containing summaries, used to calculate matrices. * **score_path** : File path where to save scores. * **bs** : Batch size * **device**: Cuda devices to use. Please make sure you have downloaded the Gupshup dataset using the above google form and provide the correct path to these files in the argument's `input_path` and `refrence_path.` Or you can simply put `test.source` and `test.target` in `data/h2e/`(hinglish to english) or `data/e2e/`(english to english) folder. For example, to generate English summaries from Hinglish dialogues using the mbart model, run the following command ``` python run_eval.py \ --model_name midas/gupshup_h2e_mbart \ --input_path data/h2e/test.source \ --save_path generated_summary.txt \ --reference_path data/h2e/test.target \ --score_path scores.txt \ --bs 8 ``` Another example, to generate English summaries from English dialogues using the Pegasus model ``` python run_eval.py \ --model_name midas/gupshup_e2e_pegasus \ --input_path data/e2e/test.source \ --save_path generated_summary.txt \ --reference_path data/e2e/test.target \ --score_path scores.txt \ --bs 8 ``` Please create an issue if you are facing any difficulties in replicating the results. ### References Please cite [[1]](https://arxiv.org/abs/1910.04073) if you found the resources in this repository useful. [1] Mehnaz, Laiba, Debanjan Mahata, Rakesh Gosangi, Uma Sushmitha Gunturi, Riya Jain, Gauri Gupta, Amardeep Kumar, Isabelle G. Lee, Anish Acharya, and Rajiv Shah. [*GupShup: Summarizing Open-Domain Code-Switched Conversations*](https://aclanthology.org/2021.emnlp-main.499.pdf) ``` @inproceedings{mehnaz2021gupshup, title={GupShup: Summarizing Open-Domain Code-Switched Conversations}, author={Mehnaz, Laiba and Mahata, Debanjan and Gosangi, Rakesh and Gunturi, Uma Sushmitha and Jain, Riya and Gupta, Gauri and Kumar, Amardeep and Lee, Isabelle G and Acharya, Anish and Shah, Rajiv}, booktitle={Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing}, pages={6177--6192}, year={2021} } ```

mmcquade11/reviews-sentiment-analysis

312c90a7ffe06f57619b45485e136e2c22c973b1

2021-12-01T18:52:49.000Z

[ "pytorch", "distilbert", "text-classification", "transformers" ]

text-classification

false

mmcquade11

null

mmcquade11/reviews-sentiment-analysis

15

null

transformers

9,546

Entry not found

mofawzy/bert-ajgt

2b31eb1e999581c5ccc7cf3c7d47ca3d2f9f50a0

2022-02-17T19:56:26.000Z

[ "pytorch", "bert", "text-classification", "ar", "dataset:AJGT", "transformers", "AJGT" ]

text-classification

false

mofawzy

null

mofawzy/bert-ajgt

15

null

transformers

9,547

--- language: - ar datasets: - AJGT tags: - AJGT widget: - text: "يهدي الله من يشاء" - text: "الاسلوب قذر وقمامه" --- # BERT-AJGT Arabic version bert model fine tuned on AJGT dataset ## Data The model were fine-tuned on ~1800 sentence from twitter for Jordanian dialect. ## Results | class | precision | recall | f1-score | Support | |----------|-----------|--------|----------|---------| | 0 | 0.9462 | 0.9778 | 0.9617 | 90 | | 1 | 0.9399 | 0.9689 | 0.9542 | 90 | | Accuracy | | | 0.9611 | 180 | ## How to use You can use these models by installing `torch` or `tensorflow` and Huggingface library `transformers`. And you can use it directly by initializing it like this: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer model_name="mofawzy/bert-ajgt" model = AutoModelForSequenceClassification.from_pretrained(model_name,num_labels=2) tokenizer = AutoTokenizer.from_pretrained(model_name) ```

monologg/koelectra-small-generator

a7d050043fdb98b63ebcb747df7004b5b94dc3b8

2020-12-26T16:23:42.000Z

[ "pytorch", "electra", "fill-mask", "ko", "transformers", "autotrain_compatible" ]

fill-mask

false

monologg

null

monologg/koelectra-small-generator

15

null

transformers

9,548

--- language: ko --- # KoELECTRA (Small Generator) Pretrained ELECTRA Language Model for Korean (`koelectra-small-generator`) For more detail, please see [original repository](https://github.com/monologg/KoELECTRA/blob/master/README_EN.md). ## Usage ### Load model and tokenizer ```python >>> from transformers import ElectraModel, ElectraTokenizer >>> model = ElectraModel.from_pretrained("monologg/koelectra-small-generator") >>> tokenizer = ElectraTokenizer.from_pretrained("monologg/koelectra-small-generator") ``` ### Tokenizer example ```python >>> from transformers import ElectraTokenizer >>> tokenizer = ElectraTokenizer.from_pretrained("monologg/koelectra-small-generator") >>> tokenizer.tokenize("[CLS] 한국어 ELECTRA를 공유합니다. [SEP]") ['[CLS]', '한국어', 'E', '##L', '##EC', '##T', '##RA', '##를', '공유', '##합니다', '.', '[SEP]'] >>> tokenizer.convert_tokens_to_ids(['[CLS]', '한국어', 'E', '##L', '##EC', '##T', '##RA', '##를', '공유', '##합니다', '.', '[SEP]']) [2, 18429, 41, 6240, 15229, 6204, 20894, 5689, 12622, 10690, 18, 3] ``` ## Example using ElectraForMaskedLM ```python from transformers import pipeline fill_mask = pipeline( "fill-mask", model="monologg/koelectra-small-generator", tokenizer="monologg/koelectra-small-generator" ) print(fill_mask("나는 {} 밥을 먹었다.".format(fill_mask.tokenizer.mask_token))) ```

monsoon-nlp/es-seq2seq-gender-decoder

e271163d2c94e74e2ebba5315c4c0d1e7e598ac2

2021-05-20T00:09:13.000Z

[ "pytorch", "bert", "text-generation", "es", "transformers" ]

text-generation

false

monsoon-nlp

null

monsoon-nlp/es-seq2seq-gender-decoder

15

null

transformers

9,549

--- language: es --- # es-seq2seq-gender (decoder) This is a seq2seq model (decoder half) to "flip" gender in Spanish sentences. The model can augment your existing Spanish data, or generate counterfactuals to test a model's decisions (would changing the gender of the subject or speaker change output?). Intended Examples: - el profesor viejo => la profesora vieja (article, noun, adjective all flip) - una actriz => un actor (irregular noun) - el lingüista => la lingüista (irregular noun) - la biblioteca => la biblioteca (no person, no flip) People's names are unchanged in this version, but you can use packages such as https://pypi.org/project/gender-guesser/ ## Sample code https://colab.research.google.com/drive/1Ta_YkXx93FyxqEu_zJ-W23PjPumMNHe5 ``` import torch from transformers import AutoTokenizer, EncoderDecoderModel model = EncoderDecoderModel.from_encoder_decoder_pretrained("monsoon-nlp/es-seq2seq-gender-encoder", "monsoon-nlp/es-seq2seq-gender-decoder") tokenizer = AutoTokenizer.from_pretrained('monsoon-nlp/es-seq2seq-gender-decoder') # all are same as BETO uncased original input_ids = torch.tensor(tokenizer.encode("la profesora vieja")).unsqueeze(0) generated = model.generate(input_ids, decoder_start_token_id=model.config.decoder.pad_token_id) tokenizer.decode(generated.tolist()[0]) > '[PAD] el profesor viejo profesor viejo profesor...' ``` ## Training I originally developed <a href="https://github.com/MonsoonNLP/el-la">a gender flip Python script</a> with <a href="https://huggingface.co/dccuchile/bert-base-spanish-wwm-uncased">BETO</a>, the Spanish-language BERT from Universidad de Chile, and spaCy to parse dependencies in sentences. More about this project: https://medium.com/ai-in-plain-english/gender-bias-in-spanish-bert-1f4d76780617 The seq2seq model is trained on gender-flipped text from that script run on the <a href="https://huggingface.co/datasets/muchocine">muchocine dataset</a>, and the first 6,853 lines from the <a href="https://oscar-corpus.com/">OSCAR corpus</a> (Spanish ded-duped). The encoder and decoder started with weights and vocabulary from BETO (uncased). ## Non-binary gender This model is useful to generate male and female text samples, but falls short of capturing gender diversity in the world and in the Spanish language. Some communities prefer the plural -@s to represent -os and -as, or -e and -es for gender-neutral or mixed-gender plural, or use fewer gendered professional nouns (la juez and not jueza). This is not yet embraced by the Royal Spanish Academy and is not represented in the corpora and tokenizers used to build this project. This seq2seq project and script could, in the future, help generate more text samples and prepare NLP models to understand us all better. #### Sources - https://www.nytimes.com/2020/04/15/world/americas/argentina-gender-language.html - https://www.washingtonpost.com/dc-md-va/2019/12/05/teens-argentina-are-leading-charge-gender-neutral-language/?arc404=true - https://www.theguardian.com/world/2020/jan/19/gender-neutral-language-battle-spain - https://es.wikipedia.org/wiki/Lenguaje_no_sexista - https://remezcla.com/culture/argentine-company-re-imagines-little-prince-gender-neutral-language/

monsoon-nlp/gpt-nyc

b49baf5fe2c2c03d309cbf681fd630ff15c564b9

2021-05-23T10:03:21.000Z

[ "pytorch", "jax", "gpt2", "text-generation", "transformers" ]

text-generation

false

monsoon-nlp

null

monsoon-nlp/gpt-nyc

15

1

transformers

9,550

# GPT-NYC ## About GPT2-Medium fine-tuned on questions and responses from https://reddit.com/r/asknyc I filtered comments to ones with scores >= 3, and responding directly to the original post ( = ignoring responses to other commenters). I added tokens to match NYC neighborhoods, subway stations, foods, and other common terms in the original batches of questions and comments. You would be surprised what is missing from GPT tokens! Try prompting with ```question? %% ``` or ```question? - more info %%``` ## Status I would like to continue by: - fine-tuning GPT2-Large with a larger dataset of questions - examining bias and toxicity - examining memorization vs. original responses - releasing a reusable benchmark ## Blog https://mapmeld.medium.com/gpt-nyc-part-1-9cb698b2e3d ## Notebooks ### Data processing / new tokens https://colab.research.google.com/drive/13BOw0uekoAYB4jjQtaXTn6J_VHatiRLu ### Fine-tuning GPT2 (small) https://colab.research.google.com/drive/1FnXcAh4H-k8dAzixkV5ieygV96ePh3lR ### Fine-tuning GPT2-Medium Same code as small, but on Google Cloud to use an A100 GPU ### Predictive text and probabilities Scroll to end of https://colab.research.google.com/drive/1FnXcAh4H-k8dAzixkV5ieygV96ePh3lR to see how to install git-lfs and trick ecco into loading this.

mrm8488/GPT-2-finetuned-CRD3

ce6e1457142a7aa61c564e5e32364f40f8cd3201

2021-05-23T10:10:58.000Z

[ "pytorch", "jax", "gpt2", "text-generation", "transformers" ]

text-generation

false

mrm8488

null

mrm8488/GPT-2-finetuned-CRD3

15

null

transformers

9,551

Entry not found

mrm8488/spanbert-base-finetuned-squadv1

54aff17ce703bd24116984d9abbd019c06253159

2021-05-20T00:49:33.000Z

[ "pytorch", "jax", "bert", "en", "arxiv:1907.10529", "transformers" ]

null

false

mrm8488

null

mrm8488/spanbert-base-finetuned-squadv1

15

null

transformers

9,552

--- language: en thumbnail: --- # SpanBERT base fine-tuned on SQuAD v1 [SpanBERT](https://github.com/facebookresearch/SpanBERT) created by [Facebook Research](https://github.com/facebookresearch) and fine-tuned on [SQuAD 1.1](https://rajpurkar.github.io/SQuAD-explorer/explore/1.1/dev/) for **Q&A** downstream task ([by them](https://github.com/facebookresearch/SpanBERT#finetuned-models-squad-1120-relation-extraction-coreference-resolution)). ## Details of SpanBERT [SpanBERT: Improving Pre-training by Representing and Predicting Spans](https://arxiv.org/abs/1907.10529) ## Details of the downstream task (Q&A) - Dataset 📚 🧐 ❓ [SQuAD1.1](https://rajpurkar.github.io/SQuAD-explorer/) ## Model fine-tuning 🏋️‍ You can get the fine-tuning script [here](https://github.com/facebookresearch/SpanBERT) ```bash python code/run_squad.py \ --do_train \ --do_eval \ --model spanbert-base-cased \ --train_file train-v1.1.json \ --dev_file dev-v1.1.json \ --train_batch_size 32 \ --eval_batch_size 32 \ --learning_rate 2e-5 \ --num_train_epochs 4 \ --max_seq_length 512 \ --doc_stride 128 \ --eval_metric f1 \ --output_dir squad_output \ --fp16 ``` ## Results Comparison 📝 | | SQuAD 1.1 | SQuAD 2.0 | Coref | TACRED | | ---------------------- | ------------- | --------- | ------- | ------ | | | F1 | F1 | avg. F1 | F1 | | BERT (base) | 88.5 | 76.5 | 73.1 | 67.7 | | SpanBERT (base) | **92.4** (this one) | [83.6](https://huggingface.co/mrm8488/spanbert-base-finetuned-squadv2) | 77.4 | [68.2](https://huggingface.co/mrm8488/spanbert-base-finetuned-tacred) | | BERT (large) | 91.3 | 83.3 | 77.1 | 66.4 | | SpanBERT (large) | [94.6](https://huggingface.co/mrm8488/spanbert-large-finetuned-squadv1) | [88.7](https://huggingface.co/mrm8488/spanbert-large-finetuned-squadv2) | 79.6 | [70.8](https://huggingface.co/mrm8488/spanbert-large-finetuned-tacred) | Note: The numbers marked as * are evaluated on the development sets because those models were not submitted to the official SQuAD leaderboard. All the other numbers are test numbers. ## Model in action Fast usage with **pipelines**: ```python from transformers import pipeline qa_pipeline = pipeline( "question-answering", model="mrm8488/spanbert-base-finetuned-squadv1", tokenizer="SpanBERT/spanbert-base-cased" ) qa_pipeline({ 'context': "Manuel Romero has been working very hard in the repository hugginface/transformers lately", 'question': "How has been working Manuel Romero lately?" }) # Output: {'answer': 'very hard in the repository hugginface/transformers', 'end': 82, 'score': 0.327230326857725, 'start': 31} ``` > Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) > Made with <span style="color: #e25555;">&hearts;</span> in Spain

ncoop57/cm_codeparrot

e21e05b4fa4c25b96a6f18f8ff8097628257550d

2022-03-02T12:59:23.000Z

[ "pytorch", "gpt_neo", "text-generation", "transformers" ]

text-generation

false

ncoop57

null

ncoop57/cm_codeparrot

15

null

transformers

9,553

Entry not found

nielsr/dino_deits8

161957bc7e0712f3c5bd5490e62ae9c70678ae4c

2021-05-03T08:17:02.000Z

[ "pytorch", "vit", "feature-extraction", "transformers" ]

feature-extraction

false

nielsr

null

nielsr/dino_deits8

15

null

transformers

9,554

Entry not found

orzhan/t5-long-extract

9b20f3c309b8cbea80d2609bbd5a638d1c7a7385

2022-06-11T07:20:59.000Z

[ "pytorch", "t5", "feature-extraction", "transformers" ]

feature-extraction

false

orzhan

null

orzhan/t5-long-extract

15

null

transformers

9,555

T5-small model fine-tuned for extractive summarization on long documents. Repository: [GitHub](https://github.com/orzhan/t5-long-extract)

patrickvonplaten/hf-reformer-crime-and-punish

498eddd2421bd1902a147e550ffae000d7a60d55

2020-05-11T11:10:52.000Z

[ "pytorch", "reformer", "text-generation", "transformers" ]

text-generation

false

patrickvonplaten

null

patrickvonplaten/hf-reformer-crime-and-punish

15

null

transformers

9,556

Entry not found

patrickvonplaten/wav2vec2-common_voice-tr-demo

59074c9bbe282a9c02a6422082d16fcc4aa46307

2021-12-20T12:54:39.000Z

[ "pytorch", "tensorboard", "wav2vec2", "automatic-speech-recognition", "tr", "dataset:common_voice", "transformers", "speech-recognition", "common_voice", "generated_from_trainer", "license:apache-2.0", "model-index" ]

automatic-speech-recognition

false

patrickvonplaten

null

patrickvonplaten/wav2vec2-common_voice-tr-demo

15

null

transformers

9,557

--- language: - tr license: apache-2.0 tags: - speech-recognition - common_voice - generated_from_trainer datasets: - common_voice model-index: - name: wav2vec2-common_voice-tr-demo results: [] ---  # wav2vec2-common_voice-tr-demo This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - TR dataset. It achieves the following results on the evaluation set: - Loss: 0.3856 - Wer: 0.3556 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 15.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 3.7391 | 0.92 | 100 | 3.5760 | 1.0 | | 2.927 | 1.83 | 200 | 3.0796 | 0.9999 | | 0.9009 | 2.75 | 300 | 0.9278 | 0.8226 | | 0.6529 | 3.67 | 400 | 0.5926 | 0.6367 | | 0.3623 | 4.59 | 500 | 0.5372 | 0.5692 | | 0.2888 | 5.5 | 600 | 0.4407 | 0.4838 | | 0.285 | 6.42 | 700 | 0.4341 | 0.4694 | | 0.0842 | 7.34 | 800 | 0.4153 | 0.4302 | | 0.1415 | 8.26 | 900 | 0.4317 | 0.4136 | | 0.1552 | 9.17 | 1000 | 0.4145 | 0.4013 | | 0.1184 | 10.09 | 1100 | 0.4115 | 0.3844 | | 0.0556 | 11.01 | 1200 | 0.4182 | 0.3862 | | 0.0851 | 11.93 | 1300 | 0.3985 | 0.3688 | | 0.0961 | 12.84 | 1400 | 0.4030 | 0.3665 | | 0.0596 | 13.76 | 1500 | 0.3880 | 0.3631 | | 0.0917 | 14.68 | 1600 | 0.3878 | 0.3582 | ### Framework versions - Transformers 4.11.0.dev0 - Pytorch 1.9.0+cu111 - Datasets 1.12.1 - Tokenizers 0.10.3

peggyhuang/bert-base-uncased-coqa

ae54d05fa6b4c7b6c04a9cd28c1fd26bfd23d4fa

2021-11-19T09:05:00.000Z

[ "pytorch", "bert", "question-answering", "transformers", "autotrain_compatible" ]

question-answering

false

peggyhuang

null

peggyhuang/bert-base-uncased-coqa

15

null

transformers

9,558

Entry not found

pinecone/bert-retriever-squad2

edb4465f3fc105473bf54cb7d4674d0d043d9185

2022-01-03T02:38:02.000Z

[ "pytorch", "bert", "feature-extraction", "sentence-transformers", "sentence-similarity", "transformers" ]

sentence-similarity

false

pinecone

null

pinecone/bert-retriever-squad2

15

null

sentence-transformers

9,559

--- pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.  ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results  For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 5429 with parameters: ``` {'batch_size': 24} ``` **Loss**: `sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss` with parameters: ``` {'scale': 20.0, 'similarity_fct': 'cos_sim'} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'transformers.optimization.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 542, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) ) ``` ## Citing & Authors

pkushiqiang/bert-title-org

113cf90e11c9920cc55c2a885813d30166991300

2022-02-28T06:16:37.000Z

[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

pkushiqiang

null

pkushiqiang/bert-title-org

15

null

transformers

9,560

Entry not found

pmthangk09/bert-base-uncased-superglue-multirc

5b503587b2e7cee15046475079d14a13aff1e616

2021-05-20T02:50:34.000Z

[ "pytorch", "tf", "jax", "bert", "text-classification", "transformers" ]

text-classification

false

pmthangk09

null

pmthangk09/bert-base-uncased-superglue-multirc

15

null

transformers

9,561

Entry not found

pszemraj/t5-base-askscience

a7582c22c55096ed39fe3261852d94d767e1da98

2022-02-19T22:50:18.000Z

[ "pytorch", "t5", "text2text-generation", "en", "dataset:eli5", "transformers", "qa", "askscience", "lfqa", "information retrieval", "autotrain_compatible" ]

text2text-generation

false

pszemraj

null

pszemraj/t5-base-askscience

15

null

transformers

9,562

--- language: - en tags: - t5 - qa - askscience - lfqa - information retrieval datasets: - eli5 metrics: - rouge widget: - text: "why aren't there more planets in our solar system?" example_title: "solar system" - text: "question: what is a probability distribution? context: I am just learning about statistics." example_title: "probability distribution" - text: "question: What are the underlying physical processes by which exercise helps us lose weight? context: I started working out two weeks ago and already feel a lot better, and started to think about it and became deeply confused." example_title: "pumpen" - text: "what is a neural network?" example_title: "deep learning" - text: "What are the primary mechanisms that computers use to understand human language?" example_title: "NLP" inference: parameters: max_length: 96 no_repeat_ngram_size: 2 encoder_no_repeat_ngram_size: 4 repetition_penalty: 3.51 length_penalty: 0.8 num_beams: 4 early_stopping: True --- # t5 - base- askscience - [t5-v1_1](https://huggingface.co/google/t5-v1_1-base) trained on the entirety of the _askscience_ sub-section of the eli5 dataset for one epoch. - compare to bart on eli5 [here](https://huggingface.co/yjernite/bart_eli5) - note that for the inference API, the model is restricted to outputting 96 tokens - by using the model in python with the transformers library, you can get longer outputs. ## training - for inputs, the model was presented with the post title and the post selftext encoded as: `question: <post title> context: <post selftext>`. You may see better results if queries are posed in this fashion. - The top two replies were aggregated and presented to the model as the output text. - Training for longer will be explored, but given that the dataset has 127k examples and the loss flatlines at 0.5 epochs so this model should be fairly viable.

racai/distilbert-multi-base-romanian-cased

099fba005eedaeda5a4aabcf6e502c2df50f58db

2021-12-24T17:32:28.000Z

[ "pytorch", "tf", "jax", "distilbert", "ro", "dataset:oscar", "dataset:wikipedia", "arxiv:2112.12650", "transformers", "license:mit" ]

null

false

racai

null

racai/distilbert-multi-base-romanian-cased

15

null

transformers

9,563

--- language: ro license: mit datasets: - oscar - wikipedia --- # Romanian DistilBERT This repository contains the a Romanian cased version of DistilBERT (named DistilMulti-BERT-base-ro in the paper) that was obtained by distilling an ensemble of two teacher models: [dumitrescustefan/bert-base-romanian-cased-v1](https://huggingface.co/dumitrescustefan/bert-base-romanian-cased-v1) and [readerbench/RoBERT-base](https://huggingface.co/readerbench/RoBERT-base). The model was introduced in [this paper](https://arxiv.org/abs/2112.12650). The adjacent code can be found [here](https://github.com/racai-ai/Romanian-DistilBERT). ## Usage ```python from transformers import AutoTokenizer, AutoModel # load the tokenizer and the model tokenizer = AutoTokenizer.from_pretrained("racai/distilbert-multi-base-romanian-cased") model = AutoModel.from_pretrained("racai/distilbert-multi-base-romanian-cased") # tokenize a test sentence input_ids = tokenizer.encode("Aceasta este o propoziție de test.", add_special_tokens=True, return_tensors="pt") # run the tokens trough the model outputs = model(input_ids) print(outputs) ``` ## Model Size The model is 35% smaller than `bert-base-romanian-cased-v1` and 30% smaller than `RoBERT-base`. | Model | Size (MB) | Params (Millions) | |--------------------------------|:---------:|:----------------:| | RoBERT-base | 441 | 114 | | bert-base-romanian-cased-v1 | 477 | 124 | | distilbert-multi-base-romanian-cased | 312 | 81 | ## Evaluation We evaluated the model in comparison with its two teachers on 5 Romanian tasks: - **UPOS**: Universal Part of Speech (F1-macro) - **XPOS**: Extended Part of Speech (F1-macro) - **NER**: Named Entity Recognition (F1-macro) - **SAPN**: Sentiment Anlaysis - Positive vs Negative (Accuracy) - **SAR**: Sentiment Analysis - Rating (F1-macro) - **DI**: Dialect identification (F1-macro) - **STS**: Semantic Textual Similarity (Pearson) | Model | UPOS | XPOS | NER | SAPN | SAR | DI | STS | |--------------------------------|:----:|:----:|:---:|:----:|:---:|:--:|:---:| | RoBERT-base | 98.02 | 97.15 | 85.14 | 98.30 | 79.40 | 96.07 | 81.18 | | bert-base-romanian-cased-v1 | 98.00 | 96.46 | 85.88 | 98.07 | 79.61 | 95.58 | 80.30 | | distilbert-multi-base-romanian-cased | 98.07 | 96.83 | 83.22 | 98.11 | 79.77 | 96.18 | 80.66 | ### BibTeX entry and citation info ```bibtex @article{avram2021distilling, title={Distilling the Knowledge of Romanian BERTs Using Multiple Teachers}, author={Andrei-Marius Avram and Darius Catrina and Dumitru-Clementin Cercel and Mihai Dascălu and Traian Rebedea and Vasile Păiş and Dan Tufiş}, journal={ArXiv}, year={2021}, volume={abs/2112.12650} } ```

ramybaly/ner_conll2003

8705927627b7f05803a0a150c8c369c376ad1383

2021-08-21T03:21:14.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "dataset:conll2003", "transformers", "generated_from_trainer", "license:apache-2.0", "autotrain_compatible" ]

token-classification

false

ramybaly

null

ramybaly/ner_conll2003

15

null

transformers

9,564

--- license: apache-2.0 tags: - generated_from_trainer datasets: - conll2003 metrics: - precision - recall - f1 - accuracy model_index: - name: ner_conll2003 results: - task: name: Token Classification type: token-classification dataset: name: conll2003 type: conll2003 args: conll2003 metric: name: Accuracy type: accuracy value: 0.9772880710440217 ---  # ner_conll2003 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the conll2003 dataset. It achieves the following results on the evaluation set: - Loss: 0.1495 - Precision: 0.8985 - Recall: 0.9130 - F1: 0.9057 - Accuracy: 0.9773 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.423 | 1.0 | 877 | 0.0656 | 0.9158 | 0.9268 | 0.9213 | 0.9818 | | 0.0575 | 2.0 | 1754 | 0.0574 | 0.9285 | 0.9445 | 0.9364 | 0.9847 | | 0.0295 | 3.0 | 2631 | 0.0631 | 0.9414 | 0.9456 | 0.9435 | 0.9859 | | 0.0155 | 4.0 | 3508 | 0.0680 | 0.9395 | 0.9467 | 0.9431 | 0.9860 | | 0.0097 | 5.0 | 4385 | 0.0694 | 0.9385 | 0.9513 | 0.9449 | 0.9863 | | 0.0059 | 6.0 | 5262 | 0.0743 | 0.9363 | 0.9471 | 0.9416 | 0.9860 | | 0.0041 | 7.0 | 6139 | 0.0803 | 0.9371 | 0.9518 | 0.9444 | 0.9862 | ### Framework versions - Transformers 4.9.1 - Pytorch 1.9.0+cu102 - Datasets 1.11.0 - Tokenizers 0.10.2

rebeccakoganlee/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ner

badcc7e3abb95def4e8dac5fc7bb5610b6c8e865

2021-11-23T20:42:01.000Z

[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

rebeccakoganlee

null

rebeccakoganlee/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ner

15

null

transformers

9,565

Entry not found

sarahmiller137/distilbert-base-uncased-ft-conll2003

97f129c62514a7e0aa79f6c8de00c17c792065a4

2022-07-14T11:52:53.000Z

[ "pytorch", "distilbert", "token-classification", "en", "dataset:conll2003", "transformers", "token classification", "license:cc", "model-index", "autotrain_compatible" ]

token-classification

false

sarahmiller137

null

sarahmiller137/distilbert-base-uncased-ft-conll2003

15

null

transformers

9,566

--- language: - en thumbnail: url to a thumbnail used in social sharing tags: - token classification license: cc datasets: - conll2003 model-index: - name: sarahmiller137/distilbert-base-uncased-ft-conll2003 results: - task: type: token-classification name: Token Classification dataset: name: conll2003 type: conll2003 config: conll2003 split: test metrics: - name: Accuracy type: accuracy value: 0.9750189904012154 verified: true - name: Precision type: precision value: 0.9802152215150602 verified: true - name: Recall type: recall value: 0.9803021169462076 verified: true - name: F1 type: f1 value: 0.9802586673049137 verified: true - name: loss type: loss value: 0.10723897069692612 verified: true --- ## Model information: distilbert-base-uncased model finetuned using the conll2003 dataset from the datasets library. ## Intended uses & limitations This model is intended to be used for named entity recoginition tasks. The model will identify entities of persons, locations, organisations, and miscellaneous. The model will predict lables based upon the CoNLL-2003 dataset. Note that the dataset and model may not be fully represetative or suitable for all needs it is recommended that the paper for the dataset and base model card should be reviewed before using the model - - [CoNLL-2003](https://aclanthology.org/W03-0419) - [distilbert](https://huggingface.co/distilbert-base-uncased) ## How to use Load the model from the library using the following checkpoints: ```python from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("sarahmiller137/distilbert-base-uncased-ft-conll2003") model = AutoModel.from_pretrained("sarahmiller137/distilbert-base-uncased-ft-conll2003") ```

sentence-transformers/distilroberta-base-msmarco-v1

b625c1f869c7869b660b456ecf8eff290d1333e3

2022-06-16T01:04:36.000Z

[ "pytorch", "tf", "roberta", "feature-extraction", "arxiv:1908.10084", "sentence-transformers", "sentence-similarity", "transformers", "license:apache-2.0" ]

sentence-similarity

false

sentence-transformers

null

sentence-transformers/distilroberta-base-msmarco-v1

15

null

sentence-transformers

9,567

--- pipeline_tag: sentence-similarity license: apache-2.0 tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- **⚠️ This model is deprecated. Please don't use it as it produces sentence embeddings of low quality. You can find recommended sentence embedding models here: [SBERT.net - Pretrained Models](https://www.sbert.net/docs/pretrained_models.html)** # sentence-transformers/distilroberta-base-msmarco-v1 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('sentence-transformers/distilroberta-base-msmarco-v1') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/distilroberta-base-msmarco-v1') model = AutoModel.from_pretrained('sentence-transformers/distilroberta-base-msmarco-v1') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, max pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/distilroberta-base-msmarco-v1) ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 350, 'do_lower_case': False}) with Transformer model: RobertaModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) ) ``` ## Citing & Authors This model was trained by [sentence-transformers](https://www.sbert.net/). If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084): ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "http://arxiv.org/abs/1908.10084", } ```

sismetanin/sbert-ru-sentiment-rutweetcorp

b7f3c2cea37655f012af16fc852c454f3f998e64

2021-05-20T06:41:48.000Z

[ "pytorch", "jax", "bert", "text-classification", "transformers" ]

text-classification

false

sismetanin

null

sismetanin/sbert-ru-sentiment-rutweetcorp

15

null

transformers

9,568

Entry not found

skplanet/dialog-koelectra-small-discriminator

d7a060551ecc231b35236a8d7d1647876c193986

2021-04-13T01:15:27.000Z

[ "pytorch", "electra", "pretraining", "transformers" ]

null

false

skplanet

null

skplanet/dialog-koelectra-small-discriminator

15

null

transformers

9,569

# Dialog-KoELECTRA Github : [https://github.com/skplanet/Dialog-KoELECTRA](https://github.com/skplanet/Dialog-KoELECTRA) ## Introduction **Dialog-KoELECTRA** is a language model specialized for dialogue. It was trained with 22GB colloquial and written style Korean text data. Dialog-ELECTRA model is made based on the [ELECTRA](https://openreview.net/pdf?id=r1xMH1BtvB) model. ELECTRA is a method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. <br> ## Released Models We are initially releasing small version pre-trained model. The model was trained on Korean text. We hope to release other models, such as base/large models, in the future. | Model | Layers | Hidden Size | Params | Max<br/>Seq Len | Learning<br/>Rate | Batch Size | Train Steps | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | Dialog-KoELECTRA-Small | 12 | 256 | 14M | 128 | 1e-4 | 512 | 700K | <br> ## Model Performance Dialog-KoELECTRA shows strong performance in conversational downstream tasks. | | **NSMC**<br/>(acc) | **Question Pair**<br/>(acc) | **Korean-Hate-Speech**<br/>(F1) | **Naver NER**<br/>(F1) | **KorNLI**<br/>(acc) | **KorSTS**<br/>(spearman) | | :--------------------- | :----------------: | :--------------------: | :----------------: | :------------------: | :-----------------------: | :-------------------------: | | DistilKoBERT | 88.60 | 92.48 | 60.72 | 84.65 | 72.00 | 72.59 | | **Dialog-KoELECTRA-Small** | **90.01** | **94.99** | **68.26** | **85.51** | **78.54** | **78.96** | <br> ## Train Data <table class="tg"> <thead> <tr> <th class="tg-c3ow"></th> <th class="tg-c3ow">corpus name</th> <th class="tg-c3ow">size</th> </tr> </thead> <tbody> <tr> <td class="tg-c3ow" rowspan="4">dialog</td> <td class="tg-0pky"><a href="https://aihub.or.kr/aidata/85" target="_blank" rel="noopener noreferrer">Aihub Korean dialog corpus</a></td> <td class="tg-c3ow" rowspan="4">7GB</td> </tr> <tr> <td class="tg-0pky"><a href="https://corpus.korean.go.kr/" target="_blank" rel="noopener noreferrer">NIKL Spoken corpus</a></td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/songys/Chatbot_data" target="_blank" rel="noopener noreferrer">Korean chatbot data</a></td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/Beomi/KcBERT" target="_blank" rel="noopener noreferrer">KcBERT</a></td> </tr> <tr> <td class="tg-c3ow" rowspan="2">written</td> <td class="tg-0pky"><a href="https://corpus.korean.go.kr/" target="_blank" rel="noopener noreferrer">NIKL Newspaper corpus</a></td> <td class="tg-c3ow" rowspan="2">15GB</td> </tr> <tr> <td class="tg-0pky"><a href="https://github.com/lovit/namuwikitext" target="_blank" rel="noopener noreferrer">namuwikitext</a></td> </tr> </tbody> </table> <br> ## Vocabulary We applied morpheme analysis using [huggingface_konlpy](https://github.com/lovit/huggingface_konlpy) when creating a vocabulary dictionary. As a result of the experiment, it showed better performance than a vocabulary dictionary created without applying morpheme analysis. <table> <thead> <tr> <th>vocabulary size</th> <th>unused token size</th> <th>limit alphabet</th> <th>min frequency</th> </tr> </thead> <tbody> <tr> <td>40,000</td> <td>500</td> <td>6,000</td> <td>3</td> </tr> </tbody> </table> <br>

spencerh/leftpartisan

9cb2baaa91cdf3c9982ddced7c076550d9c32739

2021-04-23T19:27:15.000Z

[ "pytorch", "tf", "distilbert", "text-classification", "transformers" ]

text-classification

false

spencerh

null

spencerh/leftpartisan

15

null

transformers

9,570

# Text classifier using DistilBERT to determine Partisanship ## This is one of many single-class partisanship models label_0 refers to "left" while label_1 refers to "other". This model was trained on 40,000 articles. ### Best Practices This model was optimized for 512 token-length text. Any text below 150 tokens will result in inaccurate results.

superb/wav2vec2-large-superb-er

bd13d8ed2b396e23676111aacff32283c9dece5d

2021-11-04T16:03:41.000Z

[ "pytorch", "wav2vec2", "audio-classification", "en", "dataset:superb", "arxiv:2105.01051", "transformers", "speech", "audio", "license:apache-2.0" ]

audio-classification

false

superb

null

superb/wav2vec2-large-superb-er

15

null

transformers

9,571

--- language: en datasets: - superb tags: - speech - audio - wav2vec2 - audio-classification license: apache-2.0 widget: - example_title: IEMOCAP clip "happy" src: https://cdn-media.huggingface.co/speech_samples/IEMOCAP_Ses01F_impro03_F013.wav - example_title: IEMOCAP clip "neutral" src: https://cdn-media.huggingface.co/speech_samples/IEMOCAP_Ses01F_impro04_F000.wav --- # Wav2Vec2-Large for Emotion Recognition ## Model description This is a ported version of [S3PRL's Wav2Vec2 for the SUPERB Emotion Recognition task](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream/emotion). The base model is [wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60), which is pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. For more information refer to [SUPERB: Speech processing Universal PERformance Benchmark](https://arxiv.org/abs/2105.01051) ## Task and dataset description Emotion Recognition (ER) predicts an emotion class for each utterance. The most widely used ER dataset [IEMOCAP](https://sail.usc.edu/iemocap/) is adopted, and we follow the conventional evaluation protocol: we drop the unbalanced emotion classes to leave the final four classes with a similar amount of data points and cross-validate on five folds of the standard splits. For the original model's training and evaluation instructions refer to the [S3PRL downstream task README](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream#er-emotion-recognition). ## Usage examples You can use the model via the Audio Classification pipeline: ```python from datasets import load_dataset from transformers import pipeline dataset = load_dataset("anton-l/superb_demo", "er", split="session1") classifier = pipeline("audio-classification", model="superb/wav2vec2-large-superb-er") labels = classifier(dataset[0]["file"], top_k=5) ``` Or use the model directly: ```python import torch import librosa from datasets import load_dataset from transformers import Wav2Vec2ForSequenceClassification, Wav2Vec2FeatureExtractor def map_to_array(example): speech, _ = librosa.load(example["file"], sr=16000, mono=True) example["speech"] = speech return example # load a demo dataset and read audio files dataset = load_dataset("anton-l/superb_demo", "er", split="session1") dataset = dataset.map(map_to_array) model = Wav2Vec2ForSequenceClassification.from_pretrained("superb/wav2vec2-large-superb-er") feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained("superb/wav2vec2-large-superb-er") # compute attention masks and normalize the waveform if needed inputs = feature_extractor(dataset[:4]["speech"], sampling_rate=16000, padding=True, return_tensors="pt") logits = model(**inputs).logits predicted_ids = torch.argmax(logits, dim=-1) labels = [model.config.id2label[_id] for _id in predicted_ids.tolist()] ``` ## Eval results The evaluation metric is accuracy. | | **s3prl** | **transformers** | |--------|-----------|------------------| |**session1**| `0.6564` | `N/A` | ### BibTeX entry and citation info ```bibtex @article{yang2021superb, title={SUPERB: Speech processing Universal PERformance Benchmark}, author={Yang, Shu-wen and Chi, Po-Han and Chuang, Yung-Sung and Lai, Cheng-I Jeff and Lakhotia, Kushal and Lin, Yist Y and Liu, Andy T and Shi, Jiatong and Chang, Xuankai and Lin, Guan-Ting and others}, journal={arXiv preprint arXiv:2105.01051}, year={2021} } ```

tbrasil/classificador_de_atendimento_2_classes_v1.1

147ae7455fb7891fbcef6e27de67badb01055d22

2021-08-02T17:51:56.000Z

[ "pytorch", "bert", "text-classification", "transformers" ]

text-classification

false

tbrasil

null

tbrasil/classificador_de_atendimento_2_classes_v1.1

15

null

transformers

9,572

Entry not found

textattack/albert-base-v2-snli

706e88d23c4ec5a68fbff2c1517d0da6ef7287d1

2020-07-06T16:36:47.000Z

[ "pytorch", "albert", "text-classification", "transformers" ]

text-classification

false

textattack

null

textattack/albert-base-v2-snli

15

null

transformers

9,573

## TextAttack Model Card This `albert-base-v2` model was fine-tuned for sequence classification using TextAttack and the snli dataset loaded using the `nlp` library. The model was fine-tuned for 5 epochs with a batch size of 64, a learning rate of 2e-05, and a maximum sequence length of 64. Since this was a classification task, the model was trained with a cross-entropy loss function. The best score the model achieved on this task was 0.9060150375939849, as measured by the eval set accuracy, found after 2 epochs. For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

tmills/roberta_sfda_sharpseed

10d59130d9a12a683c1049a7573848ccc020ea1e

2021-05-20T22:41:21.000Z

[ "pytorch", "jax", "roberta", "text-classification", "transformers" ]

text-classification

false

tmills

null

tmills/roberta_sfda_sharpseed

15

null

transformers

9,574

Entry not found

ttop324/kogpt2jnovel

70c6af4eba91fb32a746588fc52c33c82437c58a

2021-11-11T07:38:14.000Z

[ "pytorch", "gpt2", "text-generation", "ko", "transformers", "license:cc-by-nc-sa-4.0" ]

text-generation

false

ttop324

null

ttop324/kogpt2jnovel

15

null

transformers

9,575

--- language: ko tags: - gpt2 license: cc-by-nc-sa-4.0 --- korean translated japan web novel finetuned from skt/kogpt2-base-v2

uclanlp/plbart-single_task-all-summarization

486692f974ac352601f3952a154d9aa9fa4bb7de

2022-03-02T07:28:07.000Z

[ "pytorch", "plbart", "text2text-generation", "transformers", "autotrain_compatible" ]

text2text-generation

false

uclanlp

null

uclanlp/plbart-single_task-all-summarization

15

null

transformers

9,576

Entry not found

ufal/byt5-small-multilexnorm2021-hr

c98026c96146a7d3d920dc64bf082f97f1900027

2021-10-20T12:27:34.000Z

[ "pytorch", "t5", "text2text-generation", "hr", "dataset:mc4", "dataset:wikipedia", "dataset:multilexnorm", "arxiv:2105.13626", "arxiv:1907.06292", "transformers", "lexical normalization", "license:apache-2.0", "autotrain_compatible" ]

text2text-generation

false

ufal

null

ufal/byt5-small-multilexnorm2021-hr

15

null

transformers

9,577

--- language: hr datasets: - mc4 - wikipedia - multilexnorm tags: - lexical normalization license: apache-2.0 --- # Fine-tuned ByT5-small for MultiLexNorm (Croatian version) ![model image](https://github.com/ufal/multilexnorm2021/raw/master/img/overall.png) This is the official release of the fine-tuned models for **the winning entry** to the [*W-NUT 2021: Multilingual Lexical Normalization (MultiLexNorm)* shared task](https://noisy-text.github.io/2021/multi-lexnorm.html), which evaluates lexical-normalization systems on 12 social media datasets in 11 languages. Our system is based on [ByT5](https://arxiv.org/abs/2105.13626), which we first pre-train on synthetic data and then fine-tune on authentic normalization data. It achieves the best performance by a wide margin in intrinsic evaluation, and also the best performance in extrinsic evaluation through dependency parsing. In addition to these fine-tuned models, we also release the source files on [GitHub](https://github.com/ufal/multilexnorm2021) and an interactive demo on [Google Colab](https://colab.research.google.com/drive/1rxpI8IlKk-D2crFqi2hdzbTBIezqgsCg?usp=sharing). ## How to use The model was *not* fine-tuned in a standard sentence-to-sentence setting – instead, it was tailored to the token-to-token definition of MultiLexNorm data. Please refer to [**the interactive demo on Colab notebook**](https://colab.research.google.com/drive/1rxpI8IlKk-D2crFqi2hdzbTBIezqgsCg?usp=sharing) to learn how to use these models. ## How to cite ```bibtex @inproceedings{wnut-ufal, title= "{ÚFAL} at {MultiLexNorm} 2021: Improving Multilingual Lexical Normalization by Fine-tuning {ByT5}", author = "Samuel, David and Straka, Milan", booktitle = "Proceedings of the 7th Workshop on Noisy User-generated Text (W-NUT 2021)", year = "2021", publisher = "Association for Computational Linguistics", address = "Punta Cana, Dominican Republic" } ``` ## ByT5 - Small ByT5 is a tokenizer-free version of [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and generally follows the architecture of [MT5](https://huggingface.co/google/mt5-small). ByT5 was only pre-trained on [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual) excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task. ByT5 works especially well on noisy text data,*e.g.*, `google/byt5-small` significantly outperforms [mt5-small](https://huggingface.co/google/mt5-small) on [TweetQA](https://arxiv.org/abs/1907.06292). Paper: [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626) Authors: *Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel*

unicamp-dl/ptt5-large-t5-vocab

c213b7615a0ecd776dfe6f2d95fcaca06fd03647

2021-06-23T14:32:15.000Z

[ "pytorch", "tf", "t5", "text2text-generation", "pt", "dataset:brWaC", "transformers", "tensorflow", "pt-br", "license:mit", "autotrain_compatible" ]

text2text-generation

false

unicamp-dl

null

unicamp-dl/ptt5-large-t5-vocab

15

null

transformers

9,578

--- language: pt license: mit tags: - t5 - pytorch - tensorflow - pt - pt-br datasets: - brWaC widget: - text: "Texto de exemplo em português" inference: false --- # Portuguese T5 (aka "PTT5") ## Introduction PTT5 is a T5 model pretrained in the BrWac corpus, a large collection of web pages in Portuguese, improving T5's performance on Portuguese sentence similarity and entailment tasks. It's available in three sizes (small, base and large) and two vocabularies (Google's T5 original and ours, trained on Portuguese Wikipedia). For further information or requests, please go to [PTT5 repository](https://github.com/unicamp-dl/PTT5). ## Available models | Model | Size | #Params | Vocabulary | | :-: | :-: | :-: | :-: | | [unicamp-dl/ptt5-small-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-small-t5-vocab) | small | 60M | Google's T5 | | [unicamp-dl/ptt5-base-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-base-t5-vocab) | base | 220M | Google's T5 | | [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) | large | 740M | Google's T5 | | [unicamp-dl/ptt5-small-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-small-portuguese-vocab) | small | 60M | Portuguese | | **[unicamp-dl/ptt5-base-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-base-portuguese-vocab)** **(Recommended)** | **base** | **220M** | **Portuguese** | | [unicamp-dl/ptt5-large-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-large-portuguese-vocab) | large | 740M | Portuguese | ## Usage ```python # Tokenizer from transformers import T5Tokenizer # PyTorch (bare model, baremodel + language modeling head) from transformers import T5Model, T5ForConditionalGeneration # Tensorflow (bare model, baremodel + language modeling head) from transformers import TFT5Model, TFT5ForConditionalGeneration model_name = 'unicamp-dl/ptt5-base-portuguese-vocab' tokenizer = T5Tokenizer.from_pretrained(model_name) # PyTorch model_pt = T5ForConditionalGeneration.from_pretrained(model_name) # TensorFlow model_tf = TFT5ForConditionalGeneration.from_pretrained(model_name) ``` # Citation If you use PTT5, please cite: @article{ptt5_2020, title={PTT5: Pretraining and validating the T5 model on Brazilian Portuguese data}, author={Carmo, Diedre and Piau, Marcos and Campiotti, Israel and Nogueira, Rodrigo and Lotufo, Roberto}, journal={arXiv preprint arXiv:2008.09144}, year={2020} }

vasudevgupta/bigbird-pegasus-large-pubmed

5cb34a36cccf14bb7bed607bade700d74e923fc2

2021-05-04T11:12:55.000Z

[ "pytorch", "bigbird_pegasus", "text2text-generation", "transformers", "autotrain_compatible" ]

text2text-generation

false

vasudevgupta

null

vasudevgupta/bigbird-pegasus-large-pubmed

15

null

transformers

9,579

Moved here: https://huggingface.co/google/bigbird-pegasus-large-pubmed

vukpetar/trocr-small-photomath

daa6f7cd6b80a9040ddb2ca4f15061652d2068cc

2021-12-27T19:41:43.000Z

[ "pytorch", "vision-encoder-decoder", "arxiv:2109.10282", "transformers" ]

null

false

vukpetar

null

vukpetar/trocr-small-photomath

15

null

transformers

9,580

## TrOCR (small-sized model, fine-tuned on Synthetic Math Expression Dataset) TrOCR model fine-tuned on the Synthetic Math Expression Dataset. It was introduced in the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Li et al. and first released in [this repository](https://github.com/microsoft/unilm/tree/master/trocr). Disclaimer: The team releasing TrOCR did not write a model card for this model so this model card has been written by the Hugging Face team. ## Model description The TrOCR model is an encoder-decoder model, consisting of an image Transformer as encoder, and a text Transformer as decoder. The image encoder was initialized from the weights of BEiT, while the text decoder was initialized from the weights of RoBERTa. Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder. Next, the Transformer text decoder autoregressively generates tokens. ## Intended uses & limitations You can use the raw model for optical character recognition (OCR) on single text-line images. See the model hub to look for fine-tuned versions on a task that interests you. ## How to use Here is how to use this model in PyTorch: ```python from transformers import VisionEncoderDecoderModel, AutoFeatureExtractor, AutoTokenizer from PIL import Image import requests # load image from the IAM database url = 'https://drive.google.com/uc?export=view&id=15dUjO44YDe1Agw_Qi8MyODRHpUFaCFw-' image = Image.open(requests.get(url, stream=True).raw).convert("RGB") feature_extractor = AutoFeatureExtractor.from_pretrained('vukpetar/trocr-small-photomath') tokenizer = AutoTokenizer.from_pretrained("vukpetar/trocr-small-photomath") model = VisionEncoderDecoderModel.from_pretrained('vukpetar/trocr-small-photomath') pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values generated_ids = model.generate(pixel_values) generated_text = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] ``` ## BibTeX entry and citation info @misc{li2021trocr, title={TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models}, author={Minghao Li and Tengchao Lv and Lei Cui and Yijuan Lu and Dinei Florencio and Cha Zhang and Zhoujun Li and Furu Wei}, year={2021}, eprint={2109.10282}, archivePrefix={arXiv}, primaryClass={cs.CL} }

yazdipour/text-to-sparql-t5-small

f485542939bc21807227d766ecbb8e47007c989d

2021-10-19T11:17:46.000Z

[ "pytorch", "tensorboard", "t5", "text2text-generation", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index", "autotrain_compatible" ]

text2text-generation

false

yazdipour

null

yazdipour/text-to-sparql-t5-small

15

null

transformers

9,581

--- license: apache-2.0 tags: - generated_from_trainer datasets: - null metrics: - f1 model-index: - name: text-to-sparql-t5-small-2021-10-19_10-17_lastDS results: - task: name: Sequence-to-sequence Language Modeling type: text2text-generation metrics: - name: F1 type: f1 value: 0.3129461705684662 ---  # text-to-sparql-t5-small-2021-10-19_10-17_lastDS This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2335 - Gen Len: 19.0 - P: 0.5580 - R: 0.0884 - F1: 0.3129 - Score: 5.9585 - Bleu-precisions: [90.11303396628615, 80.34125695971072, 73.81487011728768, 69.48796722990271] - Bleu-bp: 0.0763 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Gen Len | P | R | F1 | Score | Bleu-precisions | Bleu-bp | |:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:------:|:------:|:------:|:----------------------------------------------------------------------------:|:-------:| | 0.3166 | 1.0 | 4807 | 0.2335 | 19.0 | 0.5580 | 0.0884 | 0.3129 | 5.9585 | [90.11303396628615, 80.34125695971072, 73.81487011728768, 69.48796722990271] | 0.0763 | ### Framework versions - Transformers 4.10.0 - Pytorch 1.9.0+cu111 - Datasets 1.13.3 - Tokenizers 0.10.3

yuvraj/xSumm

6403af6f8f4eaf246bc94eef9d4ec1df88e2eca9

2020-12-11T22:05:01.000Z

[ "pytorch", "bart", "text2text-generation", "en", "transformers", "summarization", "extreme summarization", "autotrain_compatible" ]

summarization

false

yuvraj

null

yuvraj/xSumm

15

null

transformers

9,582

--- language: "en" tags: - summarization - extreme summarization --- ## Model description BartForConditionalGenerationModel for extreme summarization- creates a one line abstractive summary of a given article ## How to use PyTorch model available ```python from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline tokenizer = AutoTokenizer.from_pretrained("yuvraj/xSumm") model = AutoModelWithLMHead.from_pretrained("yuvraj/xSumm") xsumm = pipeline('summarization', model=model, tokenizer=tokenizer) xsumm("<text to be summarized>") ## Limitations and bias Trained on a small fraction of the xsumm training dataset

zhiheng-huang/bert-large-uncased-whole-word-masking-embedding-relative-key-query

8e5156c80b48db5fbe0868ca18d4e4e462a896b0

2021-05-20T09:48:50.000Z

[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]

fill-mask

false

zhiheng-huang

null

zhiheng-huang/bert-large-uncased-whole-word-masking-embedding-relative-key-query

15

null

transformers

9,583

Entry not found

Davlan/xlm-roberta-base-masakhaner

643ee144abafa9c5fbe5f71f25d8a0118b6344a3

2022-02-25T15:23:22.000Z

[ "pytorch", "xlm-roberta", "token-classification", "am", "ha", "ig", "rw", "lg", "luo", "pcm", "sw", "wo", "yo", "multilingual", "dataset:masakhaner", "arxiv:2103.11811", "transformers", "autotrain_compatible" ]

token-classification

false

Davlan

null

Davlan/xlm-roberta-base-masakhaner

15

null

transformers

9,584

Hugging Face's logo --- language: - am - ha - ig - rw - lg - luo - pcm - sw - wo - yo - multilingual datasets: - masakhaner --- # xlm-roberta-base-masakhaner ## Model description **xlm-roberta-base-masakhaner** is the first **Named Entity Recognition** model for 10 African languages (Amharic, Hausa, Igbo, Kinyarwanda, Luganda, Nigerian Pidgin, Swahili, Wolof, and Yorùbá) based on a fine-tuned XLM-RoBERTa large model. It achieves the **state-of-the-art performance** for the NER task. It has been trained to recognize four types of entities: dates & times (DATE), location (LOC), organizations (ORG), and person (PER). Specifically, this model is a *xlm-roberta-large* model that was fine-tuned on an aggregation of African language datasets obtained from Masakhane [MasakhaNER](https://github.com/masakhane-io/masakhane-ner) dataset. ## Intended uses & limitations #### How to use You can use this model with Transformers *pipeline* for NER. ```python from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline tokenizer = AutoTokenizer.from_pretrained("Davlan/xlm-roberta-base-masakhaner") model = AutoModelForTokenClassification.from_pretrained("Davlan/xlm-roberta-base-masakhaner") nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Emir of Kano turban Zhang wey don spend 18 years for Nigeria" ner_results = nlp(example) print(ner_results) ``` #### Limitations and bias This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains. ## Training data This model was fine-tuned on 10 African NER datasets (Amharic, Hausa, Igbo, Kinyarwanda, Luganda, Nigerian Pidgin, Swahili, Wolof, and Yorùbá) Masakhane [MasakhaNER](https://github.com/masakhane-io/masakhane-ner) dataset The training dataset distinguishes between the beginning and continuation of an entity so that if there are back-to-back entities of the same type, the model can output where the second entity begins. As in the dataset, each token will be classified as one of the following classes: Abbreviation|Description -|- O|Outside of a named entity B-DATE |Beginning of a DATE entity right after another DATE entity I-DATE |DATE entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organisation right after another organisation I-ORG |Organisation B-LOC |Beginning of a location right after another location I-LOC |Location ## Training procedure This model was trained on a single NVIDIA V100 GPU with recommended hyperparameters from the [original MasakhaNER paper](https://arxiv.org/abs/2103.11811) which trained & evaluated the model on MasakhaNER corpus. ### BibTeX entry and citation info ``` @article{adelani21tacl, title = {Masakha{NER}: Named Entity Recognition for African Languages}, author = {David Ifeoluwa Adelani and Jade Abbott and Graham Neubig and Daniel D'souza and Julia Kreutzer and Constantine Lignos and Chester Palen-Michel and Happy Buzaaba and Shruti Rijhwani and Sebastian Ruder and Stephen Mayhew and Israel Abebe Azime and Shamsuddeen Muhammad and Chris Chinenye Emezue and Joyce Nakatumba-Nabende and Perez Ogayo and Anuoluwapo Aremu and Catherine Gitau and Derguene Mbaye and Jesujoba Alabi and Seid Muhie Yimam and Tajuddeen Gwadabe and Ignatius Ezeani and Rubungo Andre Niyongabo and Jonathan Mukiibi and Verrah Otiende and Iroro Orife and Davis David and Samba Ngom and Tosin Adewumi and Paul Rayson and Mofetoluwa Adeyemi and Gerald Muriuki and Emmanuel Anebi and Chiamaka Chukwuneke and Nkiruka Odu and Eric Peter Wairagala and Samuel Oyerinde and Clemencia Siro and Tobius Saul Bateesa and Temilola Oloyede and Yvonne Wambui and Victor Akinode and Deborah Nabagereka and Maurice Katusiime and Ayodele Awokoya and Mouhamadane MBOUP and Dibora Gebreyohannes and Henok Tilaye and Kelechi Nwaike and Degaga Wolde and Abdoulaye Faye and Blessing Sibanda and Orevaoghene Ahia and Bonaventure F. P. Dossou and Kelechi Ogueji and Thierno Ibrahima DIOP and Abdoulaye Diallo and Adewale Akinfaderin and Tendai Marengereke and Salomey Osei}, journal = {Transactions of the Association for Computational Linguistics (TACL)}, month = {}, url = {https://arxiv.org/abs/2103.11811}, year = {2021} } ```

ghadeermobasher/BC5CDR-Chem2-imbalanced-BiomedNLP-PubMedBERT-base-uncased-abstract

a95c14a635a2938902dc6d864e6b1fb147e1faa9

2022-03-01T06:00:08.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC5CDR-Chem2-imbalanced-BiomedNLP-PubMedBERT-base-uncased-abstract

15

null

transformers

9,585

Entry not found

ghadeermobasher/Model_org

ec0a976a03a68831924a915e003e3cbe8eee4ee6

2022-03-01T21:25:54.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_org

15

null

transformers

9,586

Entry not found

ghadeermobasher/Model_imb

e01715bfd053cf8e19121f5b09131f2e2394a1ee

2022-03-01T21:26:51.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_imb

15

null

transformers

9,587

Entry not found

ghadeermobasher/Model_imb_1

1a90a396146a91d18e13ceae7a50bb42962c0250

2022-03-02T04:10:59.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_imb_1

15

null

transformers

9,588

Entry not found

ghadeermobasher/Model_org_1

9925b3108d651f35452c969d7289aca4d75f6ab1

2022-03-02T04:16:51.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_org_1

15

null

transformers

9,589

Entry not found

ghadeermobasher/Model_imb_2

4fdc981bc5a81f8d794ab339886b2630f0d7b089

2022-03-02T11:34:17.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_imb_2

15

null

transformers

9,590

Entry not found

ghadeermobasher/Model_co_imb

d4dcc81191114c756cac56ba59b7babe3d471a85

2022-03-01T23:08:18.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/Model_co_imb

15

null

transformers

9,591

Entry not found

ActivationAI/distilbert-base-uncased-finetuned-emotion

dbf4470880ff3b73f22975241cd309bdf8e2195f

2022-03-02T03:40:08.000Z

[ "pytorch", "tensorboard", "distilbert", "text-classification", "dataset:emotion", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]

text-classification

false

ActivationAI

null

ActivationAI/distilbert-base-uncased-finetuned-emotion

15

null

transformers

9,592

--- license: apache-2.0 tags: - generated_from_trainer datasets: - emotion metrics: - accuracy - f1 model-index: - name: distilbert-base-uncased-finetuned-emotion results: - task: name: Text Classification type: text-classification dataset: name: emotion type: emotion args: default metrics: - name: Accuracy type: accuracy value: 0.928 - name: F1 type: f1 value: 0.9280065074208208 ---  # distilbert-base-uncased-finetuned-emotion This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset. It achieves the following results on the evaluation set: - Loss: 0.2128 - Accuracy: 0.928 - F1: 0.9280 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 64 - eval_batch_size: 64 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:| | 0.8151 | 1.0 | 250 | 0.3043 | 0.907 | 0.9035 | | 0.24 | 2.0 | 500 | 0.2128 | 0.928 | 0.9280 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.10.0+cu111 - Datasets 1.16.1 - Tokenizers 0.10.3

ghadeermobasher/BC4-original-PubmedBert

96d1e5c0b00169c4644306ddfd91da3dbe509f69

2022-03-03T02:53:18.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-original-PubmedBert

15

null

transformers

9,593

Entry not found

ghadeermobasher/BC4-original-PubmedBert_small

f72da94ac4f4208739edfaca46482bd43873f66f

2022-03-02T11:07:21.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-original-PubmedBert_small

15

null

transformers

9,594

Entry not found

ghadeermobasher/BC4-modified-PubmedBert_small

c7ffad75406d8918820a2a813bca1e9e6a013c60

2022-03-02T11:07:25.000Z

[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]

token-classification

false

ghadeermobasher

null

ghadeermobasher/BC4-modified-PubmedBert_small

15

null

transformers

9,595

Entry not found

ivanlau/distil-bert-uncased-finetuned-github-issues

0d35383caff319649c8996504a0f8b5b0a33dea4

2022-03-04T10:16:47.000Z

[ "pytorch", "distilbert", "text-classification", "dataset:ticket tagger", "transformers", "model-index" ]

text-classification

false

ivanlau

null

ivanlau/distil-bert-uncased-finetuned-github-issues

15

null

transformers

9,596

--- datasets: - ticket tagger metrics: - accuracy model-index: - name: distil-bert-uncased-finetuned-github-issues results: - task: name: Text Classification type: text-classification dataset: name: ticket tagger type: ticket tagger args: full metrics: - name: Accuracy type: accuracy value: 0.7862 --- # Model Description This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) and fine-tuning it on the [github ticket tagger dataset](https://tickettagger.blob.core.windows.net/datasets/dataset-labels-top3-30k-real.txt). It classifies issue into 3 common categories: Bug, Enhancement, Questions. It achieves the following results on the evaluation set: - Accuracy: 0.7862 ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-5 - train_batch_size: 16 - optimizer: AdamW with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 0 - num_epochs: 5 ### Codes https://github.com/IvanLauLinTiong/IntelliLabel

l3cube-pune/marathi-albert-v2

76ef3b6421baf9bb747e102310594550f4627587

2022-06-26T15:13:43.000Z

[ "pytorch", "albert", "fill-mask", "mr", "dataset:L3Cube-MahaCorpus", "arxiv:2202.01159", "transformers", "license:cc-by-4.0", "autotrain_compatible" ]

fill-mask

false

l3cube-pune

null

l3cube-pune/marathi-albert-v2

15

1

transformers

9,597

--- license: cc-by-4.0 language: mr datasets: - L3Cube-MahaCorpus --- ## MahaAlBERT MahaAlBERT is a Marathi AlBERT model trained on L3Cube-MahaCorpus and other publicly available Marathi monolingual datasets. [dataset link] (https://github.com/l3cube-pune/MarathiNLP) More details on the dataset, models, and baseline results can be found in our [paper] (https://arxiv.org/abs/2202.01159) ``` @InProceedings{joshi:2022:WILDRE6, author = {Joshi, Raviraj}, title = {L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resources}, booktitle = {Proceedings of The WILDRE-6 Workshop within the 13th Language Resources and Evaluation Conference}, month = {June}, year = {2022}, address = {Marseille, France}, publisher = {European Language Resources Association}, pages = {97--101} } ```

timothyshi/bart-large-cnn-finetuned-booksum-chapter

24ef62670565e5ca800a0c4365d7db48bea3f494

2022-03-07T05:13:01.000Z

[ "pytorch", "tensorboard", "bart", "text2text-generation", "transformers", "generated_from_trainer", "license:mit", "model-index", "autotrain_compatible" ]

text2text-generation

false

timothyshi

null

timothyshi/bart-large-cnn-finetuned-booksum-chapter

15

1

transformers

9,598

--- license: mit tags: - generated_from_trainer metrics: - rouge model-index: - name: bart-large-cnn-finetuned-booksum-chapter results: [] ---  # bart-large-cnn-finetuned-booksum-chapter This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on the None dataset. It achieves the following results on the evaluation set: - Loss: 3.1373 - Rouge1: 18.1222 - Rouge2: 3.5783 - Rougel: 13.4084 - Rougelsum: 13.5832 - Gen Len: 63.5121 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:| | 2.5297 | 1.0 | 23094 | 3.1373 | 18.1222 | 3.5783 | 13.4084 | 13.5832 | 63.5121 | ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.11.6

datarpit/toy

9ffd6ea55f01c2da71bfd7f7a3c6c5a3f5472cdb

2022-03-10T00:06:22.000Z

[ "pytorch", "distilbert", "question-answering", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index", "autotrain_compatible" ]

question-answering

false

datarpit

null

datarpit/toy

15

null

transformers

9,599

--- license: apache-2.0 tags: - generated_from_trainer model-index: - name: toy results: [] ---  # toy This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2124 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:-----:|:---------------:| | 0.4798 | 1.0 | 231 | 0.2252 | | 0.3378 | 2.0 | 462 | 0.1777 | | 0.1024 | 3.0 | 693 | 0.1586 | | 0.0736 | 4.0 | 924 | 0.1664 | | 0.1237 | 5.0 | 1155 | 0.1692 | | 0.1049 | 6.0 | 1386 | 0.1818 | | 0.0239 | 7.0 | 1617 | 0.2127 | | 0.0036 | 8.0 | 1848 | 0.1888 | | 0.0051 | 9.0 | 2079 | 0.2061 | | 0.0003 | 10.0 | 2310 | 0.1905 | | 0.0005 | 11.0 | 2541 | 0.2011 | | 0.0003 | 12.0 | 2772 | 0.1928 | | 0.0029 | 13.0 | 3003 | 0.2563 | | 0.0002 | 14.0 | 3234 | 0.2076 | | 0.0002 | 15.0 | 3465 | 0.1980 | | 0.0001 | 16.0 | 3696 | 0.2013 | | 0.0001 | 17.0 | 3927 | 0.2089 | | 0.0001 | 18.0 | 4158 | 0.1984 | | 0.0001 | 19.0 | 4389 | 0.2017 | | 0.0001 | 20.0 | 4620 | 0.2013 | | 0.0001 | 21.0 | 4851 | 0.2142 | | 0.0001 | 22.0 | 5082 | 0.1943 | | 0.0001 | 23.0 | 5313 | 0.2003 | | 0.0 | 24.0 | 5544 | 0.2015 | | 0.0001 | 25.0 | 5775 | 0.2031 | | 0.0002 | 26.0 | 6006 | 0.2600 | | 0.0022 | 27.0 | 6237 | 0.2269 | | 0.0 | 28.0 | 6468 | 0.2125 | | 0.0 | 29.0 | 6699 | 0.2172 | | 0.0 | 30.0 | 6930 | 0.2185 | | 0.0 | 31.0 | 7161 | 0.2004 | | 0.0 | 32.0 | 7392 | 0.2077 | | 0.0 | 33.0 | 7623 | 0.2333 | | 0.0003 | 34.0 | 7854 | 0.2102 | | 0.0 | 35.0 | 8085 | 0.2095 | | 0.0 | 36.0 | 8316 | 0.2030 | | 0.0 | 37.0 | 8547 | 0.2038 | | 0.0 | 38.0 | 8778 | 0.2062 | | 0.0 | 39.0 | 9009 | 0.2080 | | 0.0 | 40.0 | 9240 | 0.2083 | | 0.0 | 41.0 | 9471 | 0.2063 | | 0.0 | 42.0 | 9702 | 0.2146 | | 0.0 | 43.0 | 9933 | 0.2168 | | 0.0 | 44.0 | 10164 | 0.2112 | | 0.0 | 45.0 | 10395 | 0.2109 | | 0.0 | 46.0 | 10626 | 0.2116 | | 0.0 | 47.0 | 10857 | 0.2122 | | 0.0 | 48.0 | 11088 | 0.2122 | | 0.0 | 49.0 | 11319 | 0.2124 | | 0.0 | 50.0 | 11550 | 0.2124 | ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0 - Datasets 1.18.3 - Tokenizers 0.11.6