<!-- 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-kd-from-pre-finetune-silver-div-2-data-indolem_sentiment-model-xlm-robe
This model is a fine-tuned version of xlm-roberta-base on the indolem_sentiment dataset. It achieves the following results on the evaluation set:
- Loss: 6.2977
- Accuracy: 0.8271
- F1: 0.6497
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 | 0.88 | 100 | 10.0868 | 0.6617 | 0.6064 |
| No log | 1.75 | 200 | 4.8175 | 0.8095 | 0.6885 |
| No log | 2.63 | 300 | 7.1235 | 0.7820 | 0.7129 |
| No log | 3.51 | 400 | 5.2677 | 0.8296 | 0.6699 |
| 5.2778 | 4.39 | 500 | 4.2716 | 0.8471 | 0.7289 |
| 5.2778 | 5.26 | 600 | 6.7596 | 0.8020 | 0.5990 |
| 5.2778 | 6.14 | 700 | 5.5708 | 0.8296 | 0.6699 |
| 5.2778 | 7.02 | 800 | 8.7370 | 0.7769 | 0.4331 |
| 5.2778 | 7.89 | 900 | 5.6004 | 0.8346 | 0.7442 |
| 1.5264 | 8.77 | 1000 | 4.5021 | 0.8521 | 0.7511 |
| 1.5264 | 9.65 | 1100 | 4.6897 | 0.8697 | 0.7658 |
| 1.5264 | 10.53 | 1200 | 4.9296 | 0.8521 | 0.7424 |
| 1.5264 | 11.4 | 1300 | 6.1323 | 0.8195 | 0.6471 |
| 1.5264 | 12.28 | 1400 | 5.2312 | 0.8471 | 0.7382 |
| 0.7493 | 13.16 | 1500 | 4.9099 | 0.8521 | 0.7704 |
| 0.7493 | 14.04 | 1600 | 5.8251 | 0.8496 | 0.7273 |
| 0.7493 | 14.91 | 1700 | 4.3103 | 0.8471 | 0.7404 |
| 0.7493 | 15.79 | 1800 | 8.6020 | 0.7945 | 0.5176 |
| 0.7493 | 16.67 | 1900 | 5.9061 | 0.8421 | 0.6897 |
| 0.5237 | 17.54 | 2000 | 4.5787 | 0.8496 | 0.7273 |
| 0.5237 | 18.42 | 2100 | 4.7173 | 0.8546 | 0.7583 |
| 0.5237 | 19.3 | 2200 | 3.8765 | 0.8596 | 0.7724 |
| 0.5237 | 20.18 | 2300 | 5.0898 | 0.8496 | 0.7500 |
| 0.5237 | 21.05 | 2400 | 4.8015 | 0.8446 | 0.7156 |
| 0.3612 | 21.93 | 2500 | 4.2518 | 0.8521 | 0.7424 |
| 0.3612 | 22.81 | 2600 | 4.4654 | 0.8446 | 0.7459 |
| 0.3612 | 23.68 | 2700 | 4.9417 | 0.8471 | 0.7510 |
| 0.3612 | 24.56 | 2800 | 5.7947 | 0.8471 | 0.7240 |
| 0.3612 | 25.44 | 2900 | 4.2916 | 0.8797 | 0.8033 |
| 0.2448 | 26.32 | 3000 | 4.5161 | 0.8571 | 0.7489 |
| 0.2448 | 27.19 | 3100 | 5.1830 | 0.8596 | 0.7667 |
| 0.2448 | 28.07 | 3200 | 5.1865 | 0.8346 | 0.7462 |
| 0.2448 | 28.95 | 3300 | 4.3948 | 0.8571 | 0.7692 |
| 0.2448 | 29.82 | 3400 | 5.1436 | 0.8446 | 0.7156 |
| 0.2034 | 30.7 | 3500 | 5.2358 | 0.8471 | 0.7550 |
| 0.2034 | 31.58 | 3600 | 4.2826 | 0.8571 | 0.7654 |
| 0.2034 | 32.46 | 3700 | 5.2970 | 0.8396 | 0.7091 |
| 0.2034 | 33.33 | 3800 | 5.5713 | 0.8521 | 0.7424 |
| 0.2034 | 34.21 | 3900 | 5.3156 | 0.8571 | 0.7299 |
| 0.1679 | 35.09 | 4000 | 6.0690 | 0.8421 | 0.7549 |
| 0.1679 | 35.96 | 4100 | 5.5894 | 0.8421 | 0.7273 |
| 0.1679 | 36.84 | 4200 | 5.4521 | 0.8496 | 0.7619 |
| 0.1679 | 37.72 | 4300 | 5.2742 | 0.8521 | 0.7424 |
| 0.1679 | 38.6 | 4400 | 6.2977 | 0.8271 | 0.6497 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
