<!-- 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-silver-data-indolem_sentiment-model-xlm-roberta-base
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: 6.9924
- Accuracy: 0.7920
- F1: 0.6719
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 | 12.8614 | 0.5965 | 0.5589 |
| No log | 1.75 | 200 | 7.3545 | 0.8095 | 0.6122 |
| No log | 2.63 | 300 | 6.7410 | 0.8221 | 0.6537 |
| No log | 3.51 | 400 | 6.6248 | 0.8221 | 0.6635 |
| 6.3557 | 4.39 | 500 | 7.7312 | 0.8020 | 0.6973 |
| 6.3557 | 5.26 | 600 | 6.8030 | 0.8020 | 0.6827 |
| 6.3557 | 6.14 | 700 | 7.2582 | 0.8145 | 0.6337 |
| 6.3557 | 7.02 | 800 | 7.2238 | 0.8045 | 0.6977 |
| 6.3557 | 7.89 | 900 | 7.5518 | 0.7794 | 0.6716 |
| 1.9555 | 8.77 | 1000 | 9.5252 | 0.7519 | 0.6551 |
| 1.9555 | 9.65 | 1100 | 7.4447 | 0.8246 | 0.65 |
| 1.9555 | 10.53 | 1200 | 9.5618 | 0.8020 | 0.5212 |
| 1.9555 | 11.4 | 1300 | 6.6724 | 0.8221 | 0.6979 |
| 1.9555 | 12.28 | 1400 | 6.6052 | 0.8246 | 0.6930 |
| 1.4165 | 13.16 | 1500 | 6.8108 | 0.8145 | 0.6838 |
| 1.4165 | 14.04 | 1600 | 7.1792 | 0.8346 | 0.6916 |
| 1.4165 | 14.91 | 1700 | 8.4259 | 0.7744 | 0.6715 |
| 1.4165 | 15.79 | 1800 | 7.0160 | 0.8321 | 0.6996 |
| 1.4165 | 16.67 | 1900 | 7.1200 | 0.8070 | 0.6667 |
| 1.0272 | 17.54 | 2000 | 9.0392 | 0.7494 | 0.6552 |
| 1.0272 | 18.42 | 2100 | 7.4422 | 0.8195 | 0.6364 |
| 1.0272 | 19.3 | 2200 | 6.7059 | 0.8095 | 0.6833 |
| 1.0272 | 20.18 | 2300 | 7.8892 | 0.8120 | 0.6544 |
| 1.0272 | 21.05 | 2400 | 8.2249 | 0.8095 | 0.5824 |
| 0.9378 | 21.93 | 2500 | 7.2143 | 0.8145 | 0.63 |
| 0.9378 | 22.81 | 2600 | 7.0931 | 0.8095 | 0.6481 |
| 0.9378 | 23.68 | 2700 | 7.7609 | 0.8170 | 0.6096 |
| 0.9378 | 24.56 | 2800 | 7.0980 | 0.8195 | 0.6505 |
| 0.9378 | 25.44 | 2900 | 5.8459 | 0.8321 | 0.7309 |
| 0.8777 | 26.32 | 3000 | 8.9860 | 0.7945 | 0.4875 |
| 0.8777 | 27.19 | 3100 | 7.9930 | 0.8145 | 0.6105 |
| 0.8777 | 28.07 | 3200 | 7.2410 | 0.8045 | 0.6667 |
| 0.8777 | 28.95 | 3300 | 7.1432 | 0.8170 | 0.6368 |
| 0.8777 | 29.82 | 3400 | 7.1618 | 0.8246 | 0.6535 |
| 0.6685 | 30.7 | 3500 | 7.2110 | 0.8045 | 0.6455 |
| 0.6685 | 31.58 | 3600 | 6.5962 | 0.8296 | 0.6699 |
| 0.6685 | 32.46 | 3700 | 7.9600 | 0.8170 | 0.6256 |
| 0.6685 | 33.33 | 3800 | 7.0565 | 0.7995 | 0.6774 |
| 0.6685 | 34.21 | 3900 | 6.8412 | 0.8120 | 0.6575 |
| 0.5701 | 35.09 | 4000 | 8.5415 | 0.8095 | 0.5730 |
| 0.5701 | 35.96 | 4100 | 7.5700 | 0.8145 | 0.6146 |
| 0.5701 | 36.84 | 4200 | 6.8730 | 0.8095 | 0.6807 |
| 0.5701 | 37.72 | 4300 | 6.9817 | 0.8070 | 0.6385 |
| 0.5701 | 38.6 | 4400 | 6.9924 | 0.7920 | 0.6719 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
