<!-- 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. -->
electra-10-epoch-sentiment
This model is a fine-tuned version of google/electra-small-discriminator on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 1.0656
- Accuracy: 0.6736
- Precision: 0.6781
- Recall: 0.6736
- Micro-avg-recall: 0.6736
- Micro-avg-precision: 0.6736
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | Micro-avg-recall | Micro-avg-precision |
|---|---|---|---|---|---|---|---|---|
| 0.6088 | 1.0 | 2851 | 0.6945 | 0.6925 | 0.6972 | 0.6925 | 0.6925 | 0.6925 |
| 0.6718 | 2.0 | 5702 | 0.7151 | 0.6921 | 0.6940 | 0.6921 | 0.6921 | 0.6921 |
| 0.5468 | 3.0 | 8553 | 0.6952 | 0.6968 | 0.6970 | 0.6968 | 0.6968 | 0.6968 |
| 0.4752 | 4.0 | 11404 | 0.7832 | 0.6795 | 0.6858 | 0.6795 | 0.6795 | 0.6795 |
| 0.502 | 5.0 | 14255 | 0.8268 | 0.6844 | 0.6878 | 0.6844 | 0.6844 | 0.6844 |
| 0.3661 | 6.0 | 17106 | 0.9138 | 0.6714 | 0.6794 | 0.6714 | 0.6714 | 0.6714 |
| 0.4026 | 7.0 | 19957 | 0.9372 | 0.6821 | 0.6850 | 0.6821 | 0.6821 | 0.6821 |
| 0.2269 | 8.0 | 22808 | 1.0094 | 0.6742 | 0.6812 | 0.6742 | 0.6742 | 0.6742 |
| 0.2442 | 9.0 | 25659 | 1.0349 | 0.6745 | 0.6782 | 0.6745 | 0.6745 | 0.6745 |
| 0.2338 | 10.0 | 28510 | 1.0656 | 0.6736 | 0.6781 | 0.6736 | 0.6736 | 0.6736 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
