|
--- |
|
license: apache-2.0 |
|
base_model: google/bigbird-roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: frame_classification_bigbird |
|
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. --> |
|
|
|
# frame_classification_bigbird |
|
|
|
This model is a fine-tuned version of [google/bigbird-roberta-base](https://huggingface.co/google/bigbird-roberta-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8803 |
|
- Accuracy: 0.8991 |
|
- F1: 0.9396 |
|
- Precision: 0.9353 |
|
- Recall: 0.9440 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Accuracy | F1 | Validation Loss | Precision | Recall | |
|
|:-------------:|:-----:|:-----:|:--------:|:------:|:---------------:|:---------:|:------:| |
|
| 0.6673 | 1.0 | 1288 | 0.9270 | 0.9570 | 0.4955 | 0.9390 | 0.9757 | |
|
| 0.5913 | 2.0 | 2576 | 0.9099 | 0.9477 | 0.6212 | 0.9178 | 0.9795 | |
|
| 0.5858 | 3.0 | 3864 | 0.9270 | 0.9572 | 0.4327 | 0.9343 | 0.9813 | |
|
| 0.5384 | 4.0 | 5152 | 0.9317 | 0.9599 | 0.4998 | 0.9377 | 0.9832 | |
|
| 0.6131 | 5.0 | 6440 | 0.9255 | 0.9561 | 0.5642 | 0.9373 | 0.9757 | |
|
| 0.5834 | 6.0 | 7728 | 0.9239 | 0.9553 | 0.6238 | 0.9340 | 0.9776 | |
|
| 0.5023 | 7.0 | 9016 | 0.9208 | 0.9533 | 0.7194 | 0.9354 | 0.9720 | |
|
| 0.5271 | 8.0 | 10304 | 0.9177 | 0.9516 | 0.7188 | 0.9320 | 0.9720 | |
|
| 0.4755 | 9.0 | 11592 | 0.8618 | 0.9177 | 0.9514 | 0.9351 | 0.9683 | |
|
| 0.4173 | 10.0 | 12880 | 0.8803 | 0.8991 | 0.9396 | 0.9353 | 0.9440 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
