<!-- 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-scratch-gold-silver-data-indolem_sentiment-model-xlm-roberta-ba
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: 8.4298
- Accuracy: 0.8120
- F1: 0.6231
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 | 10.3302 | 0.6717 | 0.5498 |
| No log | 1.75 | 200 | 9.1280 | 0.7318 | 0.6323 |
| No log | 2.63 | 300 | 8.9793 | 0.7519 | 0.6551 |
| No log | 3.51 | 400 | 8.0335 | 0.8221 | 0.6758 |
| 6.8324 | 4.39 | 500 | 11.7796 | 0.7494 | 0.6732 |
| 6.8324 | 5.26 | 600 | 11.2177 | 0.7419 | 0.6709 |
| 6.8324 | 6.14 | 700 | 8.1424 | 0.8195 | 0.7049 |
| 6.8324 | 7.02 | 800 | 9.1449 | 0.8095 | 0.6381 |
| 6.8324 | 7.89 | 900 | 8.5068 | 0.8020 | 0.6393 |
| 2.421 | 8.77 | 1000 | 9.1862 | 0.8120 | 0.6606 |
| 2.421 | 9.65 | 1100 | 8.1178 | 0.8120 | 0.6835 |
| 2.421 | 10.53 | 1200 | 7.8290 | 0.8246 | 0.6983 |
| 2.421 | 11.4 | 1300 | 9.2880 | 0.8120 | 0.6154 |
| 2.421 | 12.28 | 1400 | 9.8865 | 0.7970 | 0.5318 |
| 1.6481 | 13.16 | 1500 | 10.6401 | 0.8120 | 0.5714 |
| 1.6481 | 14.04 | 1600 | 9.7094 | 0.8045 | 0.6214 |
| 1.6481 | 14.91 | 1700 | 7.8479 | 0.8371 | 0.7234 |
| 1.6481 | 15.79 | 1800 | 9.5132 | 0.8271 | 0.6634 |
| 1.6481 | 16.67 | 1900 | 10.2200 | 0.7870 | 0.6886 |
| 1.4967 | 17.54 | 2000 | 8.3184 | 0.8271 | 0.6820 |
| 1.4967 | 18.42 | 2100 | 8.2454 | 0.8195 | 0.6697 |
| 1.4967 | 19.3 | 2200 | 8.1251 | 0.8246 | 0.6392 |
| 1.4967 | 20.18 | 2300 | 8.7633 | 0.8145 | 0.6022 |
| 1.4967 | 21.05 | 2400 | 8.4419 | 0.8070 | 0.6351 |
| 0.9451 | 21.93 | 2500 | 9.2928 | 0.7995 | 0.5238 |
| 0.9451 | 22.81 | 2600 | 9.8403 | 0.8045 | 0.5568 |
| 0.9451 | 23.68 | 2700 | 8.7286 | 0.8095 | 0.5778 |
| 0.9451 | 24.56 | 2800 | 8.0295 | 0.8371 | 0.7257 |
| 0.9451 | 25.44 | 2900 | 8.9716 | 0.7920 | 0.6982 |
| 0.931 | 26.32 | 3000 | 9.6346 | 0.8070 | 0.5838 |
| 0.931 | 27.19 | 3100 | 8.8305 | 0.8070 | 0.6244 |
| 0.931 | 28.07 | 3200 | 8.7564 | 0.8145 | 0.6864 |
| 0.931 | 28.95 | 3300 | 8.7550 | 0.8195 | 0.6327 |
| 0.931 | 29.82 | 3400 | 9.4704 | 0.8170 | 0.5922 |
| 0.7372 | 30.7 | 3500 | 8.3372 | 0.8045 | 0.61 |
| 0.7372 | 31.58 | 3600 | 8.8144 | 0.8221 | 0.6816 |
| 0.7372 | 32.46 | 3700 | 8.5465 | 0.8246 | 0.6818 |
| 0.7372 | 33.33 | 3800 | 7.8328 | 0.8170 | 0.6812 |
| 0.7372 | 34.21 | 3900 | 8.3860 | 0.8271 | 0.7113 |
| 0.729 | 35.09 | 4000 | 8.9969 | 0.7945 | 0.5495 |
| 0.729 | 35.96 | 4100 | 9.1585 | 0.8070 | 0.6244 |
| 0.729 | 36.84 | 4200 | 8.3759 | 0.8120 | 0.6606 |
| 0.729 | 37.72 | 4300 | 8.4298 | 0.8120 | 0.6231 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
