|
--- |
|
license: mit |
|
base_model: roberta-large |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- precision |
|
- recall |
|
- accuracy |
|
model-index: |
|
- name: finetuned_robert |
|
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. --> |
|
|
|
# finetuned_robert |
|
|
|
This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the topic-keyword inclusion dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2694 |
|
- F1: 0.9041 |
|
- Precision: 0.8354 |
|
- Recall: 0.9851 |
|
- Accuracy: 0.9067 |
|
|
|
## 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 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 6 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:|:--------:| |
|
| 0.7067 | 0.28 | 10 | 0.6843 | 0.0 | 0.0 | 0.0 | 0.5533 | |
|
| 0.7087 | 0.56 | 20 | 0.6786 | 0.0 | 0.0 | 0.0 | 0.5533 | |
|
| 0.6887 | 0.83 | 30 | 0.6543 | 0.7241 | 0.8571 | 0.6269 | 0.7867 | |
|
| 0.6773 | 1.11 | 40 | 0.6069 | 0.816 | 0.8793 | 0.7612 | 0.8467 | |
|
| 0.6073 | 1.39 | 50 | 0.4951 | 0.7711 | 0.6465 | 0.9552 | 0.7467 | |
|
| 0.5731 | 1.67 | 60 | 0.3976 | 0.8219 | 0.7595 | 0.8955 | 0.8267 | |
|
| 0.4806 | 1.94 | 70 | 0.3487 | 0.8421 | 0.8485 | 0.8358 | 0.86 | |
|
| 0.4685 | 2.22 | 80 | 0.5218 | 0.7811 | 0.6471 | 0.9851 | 0.7533 | |
|
| 0.4243 | 2.5 | 90 | 0.8471 | 0.7322 | 0.5776 | 1.0 | 0.6733 | |
|
| 0.3692 | 2.78 | 100 | 0.3453 | 0.8514 | 0.7778 | 0.9403 | 0.8533 | |
|
| 0.4633 | 3.06 | 110 | 0.2813 | 0.8611 | 0.8052 | 0.9254 | 0.8667 | |
|
| 0.3334 | 3.33 | 120 | 0.3090 | 0.8514 | 0.7778 | 0.9403 | 0.8533 | |
|
| 0.3167 | 3.61 | 130 | 0.3531 | 0.8497 | 0.7558 | 0.9701 | 0.8467 | |
|
| 0.2615 | 3.89 | 140 | 0.2679 | 0.8873 | 0.84 | 0.9403 | 0.8933 | |
|
| 0.2672 | 4.17 | 150 | 0.2528 | 0.8889 | 0.8312 | 0.9552 | 0.8933 | |
|
| 0.2103 | 4.44 | 160 | 0.2905 | 0.8649 | 0.7901 | 0.9552 | 0.8667 | |
|
| 0.2208 | 4.72 | 170 | 0.2992 | 0.8649 | 0.7901 | 0.9552 | 0.8667 | |
|
| 0.2267 | 5.0 | 180 | 0.2911 | 0.8859 | 0.8049 | 0.9851 | 0.8867 | |
|
| 0.1623 | 5.28 | 190 | 0.2355 | 0.9014 | 0.8533 | 0.9552 | 0.9067 | |
|
| 0.2148 | 5.56 | 200 | 0.2200 | 0.9091 | 0.8553 | 0.9701 | 0.9133 | |
|
| 0.1537 | 5.83 | 210 | 0.2694 | 0.9041 | 0.8354 | 0.9851 | 0.9067 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.0.dev0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
