<!-- 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-post-finetune-gold-silver-div-2-data-indolem_sentiment-model-ha
This model is a fine-tuned version of haryoaw/scenario-normal-finetune-clf-data-indolem_sentiment-model-xlm-roberta-base on the indolem_sentiment dataset. It achieves the following results on the evaluation set:
- Loss: 5.6174
- Accuracy: 0.8471
- F1: 0.6995
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 | 5.7401 | 0.7794 | 0.6692 |
| No log | 1.75 | 200 | 5.0741 | 0.8471 | 0.7530 |
| No log | 2.63 | 300 | 6.5202 | 0.8020 | 0.7304 |
| No log | 3.51 | 400 | 6.9274 | 0.8496 | 0.7115 |
| 3.9457 | 4.39 | 500 | 9.3710 | 0.7920 | 0.7331 |
| 3.9457 | 5.26 | 600 | 4.7972 | 0.8747 | 0.7845 |
| 3.9457 | 6.14 | 700 | 4.9606 | 0.8622 | 0.7909 |
| 3.9457 | 7.02 | 800 | 8.8469 | 0.7970 | 0.5318 |
| 3.9457 | 7.89 | 900 | 7.5532 | 0.8346 | 0.7740 |
| 1.5051 | 8.77 | 1000 | 5.3765 | 0.8672 | 0.7535 |
| 1.5051 | 9.65 | 1100 | 6.2712 | 0.8396 | 0.6893 |
| 1.5051 | 10.53 | 1200 | 5.7429 | 0.8596 | 0.7586 |
| 1.5051 | 11.4 | 1300 | 5.9646 | 0.8546 | 0.7290 |
| 1.5051 | 12.28 | 1400 | 5.4026 | 0.8571 | 0.7373 |
| 0.8569 | 13.16 | 1500 | 4.5227 | 0.8647 | 0.7750 |
| 0.8569 | 14.04 | 1600 | 6.4925 | 0.8371 | 0.6829 |
| 0.8569 | 14.91 | 1700 | 5.3791 | 0.8596 | 0.7910 |
| 0.8569 | 15.79 | 1800 | 4.9327 | 0.8521 | 0.7378 |
| 0.8569 | 16.67 | 1900 | 5.1195 | 0.8596 | 0.7358 |
| 0.6002 | 17.54 | 2000 | 5.2174 | 0.8596 | 0.7667 |
| 0.6002 | 18.42 | 2100 | 5.2355 | 0.8647 | 0.7891 |
| 0.6002 | 19.3 | 2200 | 4.3233 | 0.8672 | 0.7969 |
| 0.6002 | 20.18 | 2300 | 4.6950 | 0.8622 | 0.7755 |
| 0.6002 | 21.05 | 2400 | 5.4194 | 0.8546 | 0.7820 |
| 0.4298 | 21.93 | 2500 | 4.7010 | 0.8647 | 0.7939 |
| 0.4298 | 22.81 | 2600 | 5.4841 | 0.8546 | 0.7339 |
| 0.4298 | 23.68 | 2700 | 5.6587 | 0.8496 | 0.7273 |
| 0.4298 | 24.56 | 2800 | 5.6259 | 0.8622 | 0.7534 |
| 0.4298 | 25.44 | 2900 | 5.3377 | 0.8521 | 0.7468 |
| 0.32 | 26.32 | 3000 | 5.0968 | 0.8471 | 0.7359 |
| 0.32 | 27.19 | 3100 | 5.4142 | 0.8421 | 0.7605 |
| 0.32 | 28.07 | 3200 | 6.0043 | 0.8396 | 0.7681 |
| 0.32 | 28.95 | 3300 | 5.0767 | 0.8546 | 0.7734 |
| 0.32 | 29.82 | 3400 | 5.4201 | 0.8546 | 0.7264 |
| 0.3306 | 30.7 | 3500 | 5.0005 | 0.8672 | 0.7665 |
| 0.3306 | 31.58 | 3600 | 6.0502 | 0.8446 | 0.7048 |
| 0.3306 | 32.46 | 3700 | 5.6174 | 0.8471 | 0.6995 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
