|
--- |
|
license: mit |
|
base_model: xlm-roberta-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- hate_speech_filipino |
|
metrics: |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: scenario-non-kd-from-pre-finetune-div-2-data-hate_speech_filipino-model-xlm-robe |
|
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. --> |
|
|
|
# scenario-non-kd-from-pre-finetune-div-2-data-hate_speech_filipino-model-xlm-robe |
|
|
|
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the hate_speech_filipino dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0267 |
|
- Accuracy: 0.7750 |
|
- F1: 0.7538 |
|
|
|
## 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 |
|
- num_epochs: 6969 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:| |
|
| No log | 0.32 | 100 | 0.5927 | 0.6957 | 0.7102 | |
|
| No log | 0.64 | 200 | 0.5270 | 0.7372 | 0.7150 | |
|
| No log | 0.96 | 300 | 0.5211 | 0.7427 | 0.7347 | |
|
| No log | 1.28 | 400 | 0.5186 | 0.7628 | 0.7359 | |
|
| 0.5494 | 1.6 | 500 | 0.5580 | 0.7476 | 0.7544 | |
|
| 0.5494 | 1.92 | 600 | 0.4977 | 0.7462 | 0.6728 | |
|
| 0.5494 | 2.24 | 700 | 0.4961 | 0.7580 | 0.7335 | |
|
| 0.5494 | 2.56 | 800 | 0.4659 | 0.7776 | 0.7556 | |
|
| 0.5494 | 2.88 | 900 | 0.5498 | 0.7682 | 0.7341 | |
|
| 0.3965 | 3.19 | 1000 | 0.5319 | 0.7814 | 0.7666 | |
|
| 0.3965 | 3.51 | 1100 | 0.5999 | 0.7765 | 0.7671 | |
|
| 0.3965 | 3.83 | 1200 | 0.5410 | 0.7746 | 0.7599 | |
|
| 0.3965 | 4.15 | 1300 | 0.7042 | 0.7779 | 0.7638 | |
|
| 0.3965 | 4.47 | 1400 | 0.6954 | 0.7800 | 0.7513 | |
|
| 0.2696 | 4.79 | 1500 | 0.6503 | 0.7767 | 0.7516 | |
|
| 0.2696 | 5.11 | 1600 | 0.8617 | 0.7781 | 0.7654 | |
|
| 0.2696 | 5.43 | 1700 | 0.6982 | 0.7698 | 0.7586 | |
|
| 0.2696 | 5.75 | 1800 | 0.7800 | 0.7696 | 0.7618 | |
|
| 0.2696 | 6.07 | 1900 | 0.8446 | 0.7802 | 0.7657 | |
|
| 0.1866 | 6.39 | 2000 | 0.8388 | 0.7722 | 0.7566 | |
|
| 0.1866 | 6.71 | 2100 | 1.0689 | 0.7540 | 0.7584 | |
|
| 0.1866 | 7.03 | 2200 | 0.9079 | 0.7566 | 0.7596 | |
|
| 0.1866 | 7.35 | 2300 | 0.8218 | 0.7691 | 0.7618 | |
|
| 0.1866 | 7.67 | 2400 | 0.7888 | 0.7691 | 0.7177 | |
|
| 0.149 | 7.99 | 2500 | 0.8802 | 0.7760 | 0.7637 | |
|
| 0.149 | 8.31 | 2600 | 0.9296 | 0.7680 | 0.7700 | |
|
| 0.149 | 8.63 | 2700 | 0.9560 | 0.7795 | 0.7518 | |
|
| 0.149 | 8.95 | 2800 | 0.7176 | 0.7776 | 0.7542 | |
|
| 0.149 | 9.27 | 2900 | 1.2068 | 0.7724 | 0.7689 | |
|
| 0.1258 | 9.58 | 3000 | 1.0880 | 0.7727 | 0.7621 | |
|
| 0.1258 | 9.9 | 3100 | 0.8090 | 0.7746 | 0.7430 | |
|
| 0.1258 | 10.22 | 3200 | 1.1715 | 0.7753 | 0.7611 | |
|
| 0.1258 | 10.54 | 3300 | 1.2526 | 0.7715 | 0.7630 | |
|
| 0.1258 | 10.86 | 3400 | 1.1210 | 0.7791 | 0.7550 | |
|
| 0.1183 | 11.18 | 3500 | 1.0898 | 0.7821 | 0.7632 | |
|
| 0.1183 | 11.5 | 3600 | 1.1455 | 0.7795 | 0.7666 | |
|
| 0.1183 | 11.82 | 3700 | 1.1337 | 0.7729 | 0.7635 | |
|
| 0.1183 | 12.14 | 3800 | 1.2093 | 0.7814 | 0.7541 | |
|
| 0.1183 | 12.46 | 3900 | 1.0898 | 0.7793 | 0.7604 | |
|
| 0.0903 | 12.78 | 4000 | 1.1205 | 0.7743 | 0.7613 | |
|
| 0.0903 | 13.1 | 4100 | 1.0267 | 0.7750 | 0.7538 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.3 |
|
- Pytorch 2.0.1 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.13.3 |
|
|
