|
--- |
|
license: mit |
|
base_model: FacebookAI/roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: roberta_base_ledgar |
|
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. --> |
|
|
|
# roberta_base_ledgar |
|
|
|
This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6144 |
|
- Accuracy: 0.8449 |
|
- F1 Macro: 0.7396 |
|
- F1 Micro: 0.8449 |
|
|
|
## 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 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:| |
|
| 3.1205 | 0.11 | 100 | 2.7420 | 0.5459 | 0.2772 | 0.5459 | |
|
| 2.0491 | 0.21 | 200 | 1.8506 | 0.6623 | 0.3967 | 0.6623 | |
|
| 1.6304 | 0.32 | 300 | 1.4552 | 0.722 | 0.4953 | 0.722 | |
|
| 1.3418 | 0.43 | 400 | 1.2101 | 0.7574 | 0.5559 | 0.7574 | |
|
| 1.2156 | 0.53 | 500 | 1.0701 | 0.7728 | 0.5868 | 0.7728 | |
|
| 1.0994 | 0.64 | 600 | 0.9578 | 0.7922 | 0.6223 | 0.7922 | |
|
| 0.9857 | 0.75 | 700 | 0.8957 | 0.7968 | 0.6274 | 0.7968 | |
|
| 0.9507 | 0.85 | 800 | 0.8474 | 0.7999 | 0.6368 | 0.7999 | |
|
| 0.8734 | 0.96 | 900 | 0.7990 | 0.814 | 0.6675 | 0.814 | |
|
| 0.7802 | 1.07 | 1000 | 0.7788 | 0.8128 | 0.6606 | 0.8128 | |
|
| 0.7869 | 1.17 | 1100 | 0.7537 | 0.8178 | 0.6742 | 0.8178 | |
|
| 0.8341 | 1.28 | 1200 | 0.7309 | 0.8232 | 0.6881 | 0.8232 | |
|
| 0.7372 | 1.39 | 1300 | 0.7157 | 0.8219 | 0.6876 | 0.8219 | |
|
| 0.661 | 1.49 | 1400 | 0.7058 | 0.8224 | 0.6941 | 0.8224 | |
|
| 0.6932 | 1.6 | 1500 | 0.6944 | 0.8258 | 0.6981 | 0.8258 | |
|
| 0.7305 | 1.71 | 1600 | 0.6807 | 0.8292 | 0.7058 | 0.8292 | |
|
| 0.6952 | 1.81 | 1700 | 0.6627 | 0.8291 | 0.7066 | 0.8291 | |
|
| 0.6583 | 1.92 | 1800 | 0.6509 | 0.8322 | 0.7086 | 0.8322 | |
|
| 0.6157 | 2.03 | 1900 | 0.6487 | 0.8321 | 0.7120 | 0.8321 | |
|
| 0.5817 | 2.13 | 2000 | 0.6429 | 0.8347 | 0.7164 | 0.8347 | |
|
| 0.6002 | 2.24 | 2100 | 0.6375 | 0.836 | 0.7202 | 0.836 | |
|
| 0.5786 | 2.35 | 2200 | 0.6344 | 0.8401 | 0.7318 | 0.8401 | |
|
| 0.595 | 2.45 | 2300 | 0.6276 | 0.8382 | 0.7221 | 0.8382 | |
|
| 0.564 | 2.56 | 2400 | 0.6197 | 0.8416 | 0.7353 | 0.8416 | |
|
| 0.5404 | 2.67 | 2500 | 0.6157 | 0.8438 | 0.7412 | 0.8438 | |
|
| 0.5706 | 2.77 | 2600 | 0.6162 | 0.8418 | 0.7368 | 0.8418 | |
|
| 0.5419 | 2.88 | 2700 | 0.6148 | 0.844 | 0.7383 | 0.844 | |
|
| 0.5631 | 2.99 | 2800 | 0.6144 | 0.8449 | 0.7396 | 0.8449 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.0.dev0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
