group3_non_all_zero
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0497
- Precision: 0.0638
- Recall: 0.2421
- F1: 0.1009
- Accuracy: 0.9339
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: 3e-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: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 55 | 1.1877 | 0.0140 | 0.25 | 0.0265 | 0.7339 |
| No log | 2.0 | 110 | 0.9789 | 0.0219 | 0.2041 | 0.0395 | 0.8081 |
| No log | 3.0 | 165 | 1.0274 | 0.0385 | 0.2437 | 0.0665 | 0.8703 |
| No log | 4.0 | 220 | 1.1138 | 0.0225 | 0.1820 | 0.0401 | 0.8343 |
| No log | 5.0 | 275 | 1.1690 | 0.0335 | 0.2184 | 0.0581 | 0.8702 |
| No log | 6.0 | 330 | 1.3425 | 0.0421 | 0.2310 | 0.0712 | 0.8972 |
| No log | 7.0 | 385 | 1.5089 | 0.0445 | 0.2342 | 0.0748 | 0.9079 |
| No log | 8.0 | 440 | 1.5614 | 0.0466 | 0.2453 | 0.0783 | 0.9119 |
| No log | 9.0 | 495 | 1.7200 | 0.0534 | 0.2453 | 0.0876 | 0.9220 |
| 0.5787 | 10.0 | 550 | 1.7086 | 0.0447 | 0.2453 | 0.0756 | 0.9098 |
| 0.5787 | 11.0 | 605 | 1.8784 | 0.0553 | 0.2342 | 0.0895 | 0.9263 |
| 0.5787 | 12.0 | 660 | 1.9659 | 0.0589 | 0.2421 | 0.0947 | 0.9299 |
| 0.5787 | 13.0 | 715 | 1.9472 | 0.0600 | 0.2437 | 0.0963 | 0.9297 |
| 0.5787 | 14.0 | 770 | 2.0058 | 0.0605 | 0.2373 | 0.0964 | 0.9310 |
| 0.5787 | 15.0 | 825 | 2.0497 | 0.0638 | 0.2421 | 0.1009 | 0.9339 |
Framework versions
- Transformers 4.30.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.13.3
- Downloads last month
- 3
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.