|
--- |
|
base_model: vinai/phobert-base-v2 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- recall |
|
- precision |
|
model-index: |
|
- name: cls-comment-phobert-base-v2-v3.2.1 |
|
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. --> |
|
|
|
# cls-comment-phobert-base-v2-v3.2.1 |
|
|
|
This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2873 |
|
- Accuracy: 0.9323 |
|
- F1 Score: 0.9262 |
|
- Recall: 0.9217 |
|
- Precision: 0.9320 |
|
|
|
## 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: 1e-05 |
|
- train_batch_size: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- training_steps: 4000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall | Precision | |
|
|:-------------:|:-------:|:----:|:---------------:|:--------:|:--------:|:------:|:---------:| |
|
| 1.8947 | 0.8696 | 100 | 1.6875 | 0.4001 | 0.0832 | 0.1437 | 0.1464 | |
|
| 1.5395 | 1.7391 | 200 | 1.2897 | 0.5849 | 0.2356 | 0.2632 | 0.2752 | |
|
| 1.1205 | 2.6087 | 300 | 0.8468 | 0.7999 | 0.5833 | 0.5810 | 0.5890 | |
|
| 0.82 | 3.4783 | 400 | 0.6537 | 0.8369 | 0.6179 | 0.6355 | 0.6062 | |
|
| 0.6232 | 4.3478 | 500 | 0.5371 | 0.8538 | 0.6337 | 0.6518 | 0.7525 | |
|
| 0.5148 | 5.2174 | 600 | 0.4651 | 0.8728 | 0.7299 | 0.7211 | 0.7549 | |
|
| 0.4204 | 6.0870 | 700 | 0.4010 | 0.8869 | 0.7654 | 0.7712 | 0.8914 | |
|
| 0.3421 | 6.9565 | 800 | 0.3648 | 0.9051 | 0.8714 | 0.8588 | 0.8941 | |
|
| 0.2841 | 7.8261 | 900 | 0.3240 | 0.9182 | 0.9007 | 0.9038 | 0.8978 | |
|
| 0.2319 | 8.6957 | 1000 | 0.3025 | 0.9204 | 0.9061 | 0.8976 | 0.9175 | |
|
| 0.205 | 9.5652 | 1100 | 0.2986 | 0.9209 | 0.9099 | 0.9086 | 0.9123 | |
|
| 0.1783 | 10.4348 | 1200 | 0.3047 | 0.9206 | 0.9104 | 0.9207 | 0.9025 | |
|
| 0.1587 | 11.3043 | 1300 | 0.2758 | 0.9296 | 0.9203 | 0.9177 | 0.9233 | |
|
| 0.1286 | 12.1739 | 1400 | 0.2927 | 0.9266 | 0.9144 | 0.9199 | 0.9101 | |
|
| 0.1221 | 13.0435 | 1500 | 0.2821 | 0.9318 | 0.9245 | 0.9194 | 0.9309 | |
|
| 0.1087 | 13.9130 | 1600 | 0.2789 | 0.9293 | 0.9160 | 0.9237 | 0.9090 | |
|
| 0.0982 | 14.7826 | 1700 | 0.2834 | 0.9291 | 0.9196 | 0.9213 | 0.9188 | |
|
| 0.089 | 15.6522 | 1800 | 0.2828 | 0.9299 | 0.9202 | 0.9261 | 0.9152 | |
|
| 0.0795 | 16.5217 | 1900 | 0.2737 | 0.9331 | 0.9244 | 0.9239 | 0.9253 | |
|
| 0.0684 | 17.3913 | 2000 | 0.2873 | 0.9323 | 0.9262 | 0.9217 | 0.9320 | |
|
| 0.0673 | 18.2609 | 2100 | 0.2904 | 0.9320 | 0.9252 | 0.9184 | 0.9333 | |
|
| 0.0571 | 19.1304 | 2200 | 0.3166 | 0.9293 | 0.9222 | 0.9210 | 0.9251 | |
|
| 0.0561 | 20.0 | 2300 | 0.2922 | 0.9318 | 0.9221 | 0.9298 | 0.9150 | |
|
| 0.0511 | 20.8696 | 2400 | 0.2993 | 0.9315 | 0.9191 | 0.9303 | 0.9088 | |
|
| 0.0442 | 21.7391 | 2500 | 0.3201 | 0.9266 | 0.9162 | 0.9280 | 0.9060 | |
|
| 0.0447 | 22.6087 | 2600 | 0.3155 | 0.9282 | 0.9137 | 0.9282 | 0.9010 | |
|
| 0.0415 | 23.4783 | 2700 | 0.3018 | 0.9334 | 0.9226 | 0.9270 | 0.9185 | |
|
| 0.0359 | 24.3478 | 2800 | 0.3192 | 0.9299 | 0.9177 | 0.9308 | 0.9063 | |
|
| 0.0369 | 25.2174 | 2900 | 0.3064 | 0.9337 | 0.9211 | 0.9286 | 0.9141 | |
|
| 0.0296 | 26.0870 | 3000 | 0.3110 | 0.9329 | 0.9237 | 0.9279 | 0.9198 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.1 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
