upload everything

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+optimizer.pt filter=lfs diff=lfs merge=lfs -text
+pytorch_model.bin filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,69 @@
+---
+language: en
+widget:
+- text: What is Night of the Living Dead? \n Night of the Living Dead is a 1968 American independent horror film , directed by George A. Romero , starring Duane Jones and Judith O'Dea . George A. Romero George A. Romero Duane Jones Duane Jones Judith O'Dea Judith O'Dea independent Independent film horror film horror film.
+---
+
+# Domain-adapted QA Model From ZeroFEC 
+
+ZeroFEC is a faithful and interpetable factual error correction framework introduced in the paper [Zero-shot Faithful Factual Error Correction](https://aclanthology.org/2023.acl-long.311/). It involves a QA component, which is a UnifiedQA model continue fine-tuned on two additional biomedical QA datasets. The associated code is released in [this](https://github.com/khuangaf/ZeroFEC) repository.
+
+### How to use
+Using Huggingface pipeline abstraction:
+```python
+from transformers import pipeline
+
+nlp = pipeline("text2text-generation", model='khhuang/zerofec-daqa-t5-base', tokenizer='khhuang/zerofec-daqa-t5-base')
+    
+QUESTION = "What is Night of the Living Dead?"
+CONTEXT = "Night of the Living Dead is a 1968 American independent horror film , directed by George A. Romero , starring Duane Jones and Judith O'Dea . George A. Romero George A. Romero Duane Jones Duane Jones Judith O'Dea Judith O'Dea independent Independent film horror film horror film."
+
+def format_inputs(context: str, question: str):
+    return f"{question} \n {context}"
+
+text = format_inputs(CONTEXT, QUESTION)
+
+nlp(text)
+# should output [{'generated_text': 'a 1968 american independent horror film'}]
+```
+
+Using the pre-trained model directly:
+```python
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+
+tokenizer = AutoTokenizer.from_pretrained('khhuang/zerofec-daqa-t5-base')
+model = AutoModelForSeq2SeqLM.from_pretrained('khhuang/zerofec-daqa-t5-base')
+
+QUESTION = "What is Night of the Living Dead?"
+CONTEXT = "Night of the Living Dead is a 1968 American independent horror film , directed by George A. Romero , starring Duane Jones and Judith O'Dea . George A. Romero George A. Romero Duane Jones Duane Jones Judith O'Dea Judith O'Dea independent Independent film horror film horror film."
+
+def format_inputs(context: str, question: str):
+    return f"{question} \n {context}"
+
+text = format_inputs(CONTEXT, QUESTION)
+
+
+input_ids = tokenizer(text, return_tensors="pt").input_ids
+generated_ids = model.generate(input_ids, max_length=32, num_beams=4)
+output = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
+print(output)
+# should output "a 1968 american independent horror film"
+```
+
+### Citation
+```
+@inproceedings{huang-etal-2023-zero,
+    title = "Zero-shot Faithful Factual Error Correction",
+    author = "Huang, Kung-Hsiang  and
+      Chan, Hou Pong  and
+      Ji, Heng",
+    booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
+    month = jul,
+    year = "2023",
+    address = "Toronto, Canada",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/2023.acl-long.311",
+    doi = "10.18653/v1/2023.acl-long.311",
+    pages = "5660--5676",
+}
+```

config.json

@@ -0,0 +1,59 @@
+{
+  "_name_or_path": "allenai/unifiedqa-v2-t5-base-1251000",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "d_ff": 3072,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "relu",
+  "gradient_checkpointing": false,
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 200,
+      "min_length": 30,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4,
+      "prefix": "summarize: "
+    },
+    "translation_en_to_de": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to German: "
+    },
+    "translation_en_to_fr": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to French: "
+    },
+    "translation_en_to_ro": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to Romanian: "
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.19.2",
+  "use_cache": true,
+  "vocab_size": 32128
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da2cd3167734405f65ae2f756238da68b925b3f3d5aab21ea6b1c2781ef9070c
+size 1783379133

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fd9a05c6323a6ddb5ad922529d506b6e86591d73d0aafe711abe59e3178040f
+size 891700799

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:673b71f54c0c70f4e0bd565013e8a7f3456fc29723c29c988dac718275580bbc
+size 14503

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca2359fd6bb85576e36c95dd2d4356c709ca53c6847e7aab7798843fd12175a2
+size 559

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a836542adc98c03c89bf40db3637ba926035990d136f6780cd5043d550fe9a5d
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
+size 791656

tokenizer_config.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "t5-base", "tokenizer_class": "T5Tokenizer"}

trainer_state.json

@@ -0,0 +1,253 @@
+{
+  "best_metric": 64.0,
+  "best_model_checkpoint": "pubmedqa_unifiedqa/checkpoint-1100",
+  "epoch": 39.283185840707965,
+  "global_step": 1100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.0,
+      "loss": 5.2557,
+      "step": 1
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 1.44e-06,
+      "loss": 5.5451,
+      "step": 100
+    },
+    {
+      "epoch": 3.57,
+      "eval_bleu": 3.515852791538636,
+      "eval_gen_len": 2.44,
+      "eval_loss": 5.634477615356445,
+      "eval_meteor": 0.25,
+      "eval_rouge1": 50.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 50.0,
+      "eval_rougeLsum": 50.0,
+      "eval_runtime": 3.8236,
+      "eval_samples_per_second": 13.077,
+      "eval_steps_per_second": 1.831,
+      "step": 100
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 2.925e-06,
+      "loss": 4.9008,
+      "step": 200
+    },
+    {
+      "epoch": 7.14,
+      "eval_bleu": 3.4801542313344793,
+      "eval_gen_len": 2.44,
+      "eval_loss": 4.566105842590332,
+      "eval_meteor": 0.24,
+      "eval_rouge1": 48.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 48.0,
+      "eval_rougeLsum": 48.0,
+      "eval_runtime": 1.6121,
+      "eval_samples_per_second": 31.016,
+      "eval_steps_per_second": 4.342,
+      "step": 200
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 2.9425e-06,
+      "loss": 3.9884,
+      "step": 300
+    },
+    {
+      "epoch": 10.71,
+      "eval_bleu": 5.48568366540206,
+      "eval_gen_len": 2.24,
+      "eval_loss": 3.5699760913848877,
+      "eval_meteor": 0.28,
+      "eval_rouge1": 56.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 56.0,
+      "eval_rougeLsum": 56.0,
+      "eval_runtime": 1.5002,
+      "eval_samples_per_second": 33.328,
+      "eval_steps_per_second": 4.666,
+      "step": 300
+    },
+    {
+      "epoch": 14.28,
+      "learning_rate": 2.88125e-06,
+      "loss": 3.2254,
+      "step": 400
+    },
+    {
+      "epoch": 14.28,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.9374730587005615,
+      "eval_meteor": 0.3,
+      "eval_rouge1": 60.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 60.0,
+      "eval_rougeLsum": 60.0,
+      "eval_runtime": 1.3736,
+      "eval_samples_per_second": 36.402,
+      "eval_steps_per_second": 5.096,
+      "step": 400
+    },
+    {
+      "epoch": 17.85,
+      "learning_rate": 2.81875e-06,
+      "loss": 2.6924,
+      "step": 500
+    },
+    {
+      "epoch": 17.85,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.5620052814483643,
+      "eval_meteor": 0.29,
+      "eval_rouge1": 58.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 58.0,
+      "eval_rougeLsum": 58.0,
+      "eval_runtime": 1.3734,
+      "eval_samples_per_second": 36.405,
+      "eval_steps_per_second": 5.097,
+      "step": 500
+    },
+    {
+      "epoch": 21.42,
+      "learning_rate": 2.75625e-06,
+      "loss": 2.4122,
+      "step": 600
+    },
+    {
+      "epoch": 21.42,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.3537566661834717,
+      "eval_meteor": 0.28,
+      "eval_rouge1": 56.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 56.0,
+      "eval_rougeLsum": 56.0,
+      "eval_runtime": 1.3768,
+      "eval_samples_per_second": 36.317,
+      "eval_steps_per_second": 5.084,
+      "step": 600
+    },
+    {
+      "epoch": 24.99,
+      "learning_rate": 2.695e-06,
+      "loss": 2.2719,
+      "step": 700
+    },
+    {
+      "epoch": 24.99,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.212905168533325,
+      "eval_meteor": 0.29,
+      "eval_rouge1": 58.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 58.0,
+      "eval_rougeLsum": 58.0,
+      "eval_runtime": 1.3724,
+      "eval_samples_per_second": 36.432,
+      "eval_steps_per_second": 5.101,
+      "step": 700
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 2.6324999999999998e-06,
+      "loss": 2.1539,
+      "step": 800
+    },
+    {
+      "epoch": 28.57,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.123441219329834,
+      "eval_meteor": 0.29,
+      "eval_rouge1": 58.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 58.0,
+      "eval_rougeLsum": 58.0,
+      "eval_runtime": 1.3738,
+      "eval_samples_per_second": 36.397,
+      "eval_steps_per_second": 5.096,
+      "step": 800
+    },
+    {
+      "epoch": 32.14,
+      "learning_rate": 2.57125e-06,
+      "loss": 2.1149,
+      "step": 900
+    },
+    {
+      "epoch": 32.14,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.0748918056488037,
+      "eval_meteor": 0.29,
+      "eval_rouge1": 58.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 58.0,
+      "eval_rougeLsum": 58.0,
+      "eval_runtime": 1.3748,
+      "eval_samples_per_second": 36.37,
+      "eval_steps_per_second": 5.092,
+      "step": 900
+    },
+    {
+      "epoch": 35.71,
+      "learning_rate": 2.509375e-06,
+      "loss": 2.0404,
+      "step": 1000
+    },
+    {
+      "epoch": 35.71,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.0326242446899414,
+      "eval_meteor": 0.31,
+      "eval_rouge1": 62.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 62.0,
+      "eval_rougeLsum": 62.0,
+      "eval_runtime": 1.3725,
+      "eval_samples_per_second": 36.429,
+      "eval_steps_per_second": 5.1,
+      "step": 1000
+    },
+    {
+      "epoch": 39.28,
+      "learning_rate": 2.446875e-06,
+      "loss": 1.9918,
+      "step": 1100
+    },
+    {
+      "epoch": 39.28,
+      "eval_bleu": 0.0,
+      "eval_gen_len": 2.0,
+      "eval_loss": 2.012605667114258,
+      "eval_meteor": 0.32,
+      "eval_rouge1": 64.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 64.0,
+      "eval_rougeLsum": 64.0,
+      "eval_runtime": 1.3788,
+      "eval_samples_per_second": 36.265,
+      "eval_steps_per_second": 5.077,
+      "step": 1100
+    }
+  ],
+  "max_steps": 5000,
+  "num_train_epochs": 179,
+  "total_flos": 2.153031549530931e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f9334b33e22eb3319970cb9153ba10497580ada38c636d4807cece434bae65b
+size 3311