group3_non_all_zero_notEqualWeights
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: 2.3167
- Precision: 0.0476
- Recall: 0.2642
- F1: 0.0807
- Accuracy: 0.9145
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 | 1.3844 | 0.0068 | 0.2579 | 0.0133 | 0.6506 |
No log | 2.0 | 110 | 1.1245 | 0.0107 | 0.2342 | 0.0205 | 0.7285 |
No log | 3.0 | 165 | 1.2261 | 0.0103 | 0.2120 | 0.0196 | 0.7286 |
No log | 4.0 | 220 | 1.1828 | 0.0099 | 0.1693 | 0.0188 | 0.7551 |
No log | 5.0 | 275 | 1.2474 | 0.0141 | 0.2152 | 0.0265 | 0.8008 |
No log | 6.0 | 330 | 1.4395 | 0.0264 | 0.2516 | 0.0478 | 0.8601 |
No log | 7.0 | 385 | 1.5667 | 0.0253 | 0.2278 | 0.0456 | 0.8614 |
No log | 8.0 | 440 | 1.6080 | 0.0286 | 0.2468 | 0.0512 | 0.8756 |
No log | 9.0 | 495 | 1.7798 | 0.0289 | 0.2358 | 0.0515 | 0.8849 |
0.6462 | 10.0 | 550 | 1.9265 | 0.0364 | 0.2579 | 0.0638 | 0.8933 |
0.6462 | 11.0 | 605 | 2.0633 | 0.0347 | 0.2468 | 0.0608 | 0.8911 |
0.6462 | 12.0 | 660 | 2.2610 | 0.0458 | 0.2690 | 0.0783 | 0.9138 |
0.6462 | 13.0 | 715 | 2.1700 | 0.0435 | 0.2595 | 0.0745 | 0.9044 |
0.6462 | 14.0 | 770 | 2.3153 | 0.0480 | 0.2690 | 0.0814 | 0.9127 |
0.6462 | 15.0 | 825 | 2.3167 | 0.0476 | 0.2642 | 0.0807 | 0.9145 |
Framework versions
- Transformers 4.30.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.13.3
- Downloads last month
- 6
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.