|
--- |
|
base_model: avichr/heBERT_sentiment_analysis |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: heBERT_sentiment_analysis |
|
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. --> |
|
|
|
# heBERT_sentiment_analysis |
|
|
|
This model is a fine-tuned version of [avichr/heBERT_sentiment_analysis](https://huggingface.co/avichr/heBERT_sentiment_analysis) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4046 |
|
- Accuracy: 0.8563 |
|
- F1: 0.8554 |
|
- Precision: 0.8551 |
|
- Recall: 0.8567 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 2.5497 | 0.0820 | 50 | 0.7625 | 0.7076 | 0.6676 | 0.7357 | 0.7090 | |
|
| 0.6821 | 0.1639 | 100 | 0.5465 | 0.7898 | 0.7818 | 0.7916 | 0.7905 | |
|
| 0.5475 | 0.2459 | 150 | 0.4972 | 0.8155 | 0.8074 | 0.8198 | 0.8161 | |
|
| 0.5397 | 0.3279 | 200 | 0.4992 | 0.8213 | 0.8168 | 0.8210 | 0.8217 | |
|
| 0.5328 | 0.4098 | 250 | 0.4999 | 0.8101 | 0.7988 | 0.8213 | 0.8109 | |
|
| 0.5052 | 0.4918 | 300 | 0.4983 | 0.8108 | 0.8131 | 0.8191 | 0.8107 | |
|
| 0.4627 | 0.5738 | 350 | 0.4597 | 0.8291 | 0.8290 | 0.8317 | 0.8291 | |
|
| 0.4565 | 0.6557 | 400 | 0.4328 | 0.8362 | 0.8332 | 0.8356 | 0.8366 | |
|
| 0.4187 | 0.7377 | 450 | 0.4416 | 0.8402 | 0.8361 | 0.8413 | 0.8406 | |
|
| 0.4611 | 0.8197 | 500 | 0.4655 | 0.8308 | 0.8306 | 0.8304 | 0.8311 | |
|
| 0.436 | 0.9016 | 550 | 0.4246 | 0.8335 | 0.8343 | 0.8350 | 0.8338 | |
|
| 0.4118 | 0.9836 | 600 | 0.4179 | 0.8429 | 0.8418 | 0.8425 | 0.8431 | |
|
| 0.4019 | 1.0656 | 650 | 0.4041 | 0.8436 | 0.8426 | 0.8426 | 0.8438 | |
|
| 0.419 | 1.1475 | 700 | 0.4238 | 0.8418 | 0.8405 | 0.8403 | 0.8422 | |
|
| 0.3656 | 1.2295 | 750 | 0.4046 | 0.8485 | 0.8486 | 0.8486 | 0.8487 | |
|
| 0.3559 | 1.3115 | 800 | 0.4032 | 0.8414 | 0.8415 | 0.8414 | 0.8417 | |
|
| 0.3529 | 1.3934 | 850 | 0.4229 | 0.8489 | 0.8452 | 0.8498 | 0.8494 | |
|
| 0.3247 | 1.4754 | 900 | 0.4198 | 0.8491 | 0.8463 | 0.8491 | 0.8494 | |
|
| 0.3435 | 1.5574 | 950 | 0.4254 | 0.8395 | 0.8413 | 0.8468 | 0.8394 | |
|
| 0.3519 | 1.6393 | 1000 | 0.4414 | 0.8447 | 0.8427 | 0.8472 | 0.8448 | |
|
| 0.3763 | 1.7213 | 1050 | 0.4097 | 0.8534 | 0.8521 | 0.8534 | 0.8536 | |
|
| 0.3739 | 1.8033 | 1100 | 0.3926 | 0.8523 | 0.8487 | 0.8540 | 0.8527 | |
|
| 0.3807 | 1.8852 | 1150 | 0.3896 | 0.8528 | 0.8515 | 0.8514 | 0.8531 | |
|
| 0.3655 | 1.9672 | 1200 | 0.3897 | 0.8526 | 0.8504 | 0.8535 | 0.8528 | |
|
| 0.3125 | 2.0492 | 1250 | 0.4199 | 0.8539 | 0.8540 | 0.8545 | 0.8540 | |
|
| 0.2851 | 2.1311 | 1300 | 0.4107 | 0.8550 | 0.8541 | 0.8554 | 0.8551 | |
|
| 0.2907 | 2.2131 | 1350 | 0.4093 | 0.8527 | 0.8532 | 0.8539 | 0.8528 | |
|
| 0.2619 | 2.2951 | 1400 | 0.4085 | 0.8585 | 0.8572 | 0.8573 | 0.8588 | |
|
| 0.2619 | 2.3770 | 1450 | 0.4166 | 0.8490 | 0.8501 | 0.8523 | 0.8491 | |
|
| 0.2863 | 2.4590 | 1500 | 0.4046 | 0.8548 | 0.8538 | 0.8536 | 0.8550 | |
|
| 0.2604 | 2.5410 | 1550 | 0.4143 | 0.8550 | 0.8544 | 0.8541 | 0.8552 | |
|
| 0.2778 | 2.6230 | 1600 | 0.4021 | 0.8532 | 0.8540 | 0.8551 | 0.8533 | |
|
| 0.2579 | 2.7049 | 1650 | 0.4071 | 0.8505 | 0.8518 | 0.8542 | 0.8505 | |
|
| 0.2734 | 2.7869 | 1700 | 0.4069 | 0.8565 | 0.8566 | 0.8567 | 0.8567 | |
|
| 0.2873 | 2.8689 | 1750 | 0.4006 | 0.8567 | 0.8563 | 0.8563 | 0.8569 | |
|
| 0.2568 | 2.9508 | 1800 | 0.3998 | 0.8567 | 0.8568 | 0.8570 | 0.8569 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.42.4 |
|
- Pytorch 2.3.1+cu121 |
|
- Tokenizers 0.19.1 |
|
|
