google-electra-base-discriminator-english-sentweet-derogatory
This model is a fine-tuned version of google/electra-base-discriminator on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6258
- Accuracy: 0.8160
- Precision: 0.8241
- Recall: 0.8246
- F1: 0.8160
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: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 81 | 0.4813 | 0.7674 | 0.7723 | 0.7740 | 0.7673 |
| No log | 2.0 | 162 | 0.4645 | 0.8056 | 0.8382 | 0.8217 | 0.8046 |
| No log | 3.0 | 243 | 0.4874 | 0.8160 | 0.8258 | 0.8253 | 0.8160 |
| No log | 4.0 | 324 | 0.4160 | 0.8125 | 0.8250 | 0.8229 | 0.8125 |
| No log | 5.0 | 405 | 0.5105 | 0.8125 | 0.8147 | 0.8178 | 0.8123 |
| No log | 6.0 | 486 | 0.6258 | 0.8160 | 0.8241 | 0.8246 | 0.8160 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.1+cu117
- Datasets 2.6.1
- Tokenizers 0.11.0