<!-- 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_weight_reset-data-indolem_sentiment-model-xlmr_base_trained
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: 8.6553
- Accuracy: 0.8170
- F1: 0.6138
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 | 9.3113 | 0.6842 | 0.5563 |
| No log | 1.75 | 200 | 7.3649 | 0.7744 | 0.6154 |
| No log | 2.63 | 300 | 6.5426 | 0.8045 | 0.6389 |
| No log | 3.51 | 400 | 7.5605 | 0.7845 | 0.6446 |
| 6.3642 | 4.39 | 500 | 8.7190 | 0.7644 | 0.6594 |
| 6.3642 | 5.26 | 600 | 8.0002 | 0.8070 | 0.6010 |
| 6.3642 | 6.14 | 700 | 7.5141 | 0.8271 | 0.6667 |
| 6.3642 | 7.02 | 800 | 7.0104 | 0.8271 | 0.6849 |
| 6.3642 | 7.89 | 900 | 8.4964 | 0.8120 | 0.5946 |
| 2.1694 | 8.77 | 1000 | 9.0989 | 0.7895 | 0.5116 |
| 2.1694 | 9.65 | 1100 | 8.9281 | 0.7794 | 0.6944 |
| 2.1694 | 10.53 | 1200 | 8.6285 | 0.7895 | 0.5333 |
| 2.1694 | 11.4 | 1300 | 7.4566 | 0.7970 | 0.6920 |
| 2.1694 | 12.28 | 1400 | 7.8976 | 0.8045 | 0.7023 |
| 1.5914 | 13.16 | 1500 | 8.1554 | 0.7945 | 0.6917 |
| 1.5914 | 14.04 | 1600 | 8.1453 | 0.8120 | 0.7104 |
| 1.5914 | 14.91 | 1700 | 8.6040 | 0.8221 | 0.6816 |
| 1.5914 | 15.79 | 1800 | 8.9231 | 0.8020 | 0.5864 |
| 1.5914 | 16.67 | 1900 | 8.9577 | 0.7744 | 0.6959 |
| 1.2984 | 17.54 | 2000 | 7.3876 | 0.8321 | 0.6968 |
| 1.2984 | 18.42 | 2100 | 8.5137 | 0.8145 | 0.6263 |
| 1.2984 | 19.3 | 2200 | 8.7818 | 0.8170 | 0.6218 |
| 1.2984 | 20.18 | 2300 | 8.4247 | 0.8120 | 0.5946 |
| 1.2984 | 21.05 | 2400 | 8.9892 | 0.7970 | 0.5263 |
| 0.8187 | 21.93 | 2500 | 9.5395 | 0.7870 | 0.4848 |
| 0.8187 | 22.81 | 2600 | 8.8902 | 0.8045 | 0.5761 |
| 0.8187 | 23.68 | 2700 | 7.3568 | 0.8296 | 0.6822 |
| 0.8187 | 24.56 | 2800 | 7.5821 | 0.8145 | 0.6373 |
| 0.8187 | 25.44 | 2900 | 7.6924 | 0.8095 | 0.6238 |
| 0.7843 | 26.32 | 3000 | 7.8809 | 0.8195 | 0.6364 |
| 0.7843 | 27.19 | 3100 | 8.6553 | 0.8170 | 0.6138 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
