group2_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: 2.3325
- Precision: 0.0395
- Recall: 0.182
- F1: 0.0649
- Accuracy: 0.8597
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 | 43 | 1.5592 | 0.0020 | 0.124 | 0.0040 | 0.3311 |
No log | 2.0 | 86 | 1.2689 | 0.0104 | 0.14 | 0.0193 | 0.6247 |
No log | 3.0 | 129 | 1.1742 | 0.0110 | 0.172 | 0.0206 | 0.6614 |
No log | 4.0 | 172 | 1.3716 | 0.0147 | 0.178 | 0.0271 | 0.6468 |
No log | 5.0 | 215 | 1.3265 | 0.0177 | 0.178 | 0.0323 | 0.7203 |
No log | 6.0 | 258 | 1.5835 | 0.0217 | 0.176 | 0.0386 | 0.7574 |
No log | 7.0 | 301 | 1.6678 | 0.0249 | 0.174 | 0.0435 | 0.7952 |
No log | 8.0 | 344 | 1.9432 | 0.0387 | 0.18 | 0.0636 | 0.8551 |
No log | 9.0 | 387 | 1.9371 | 0.0306 | 0.188 | 0.0526 | 0.7962 |
No log | 10.0 | 430 | 2.0129 | 0.0305 | 0.182 | 0.0523 | 0.8187 |
No log | 11.0 | 473 | 2.1952 | 0.0402 | 0.192 | 0.0664 | 0.8595 |
0.5993 | 12.0 | 516 | 2.1873 | 0.0369 | 0.182 | 0.0614 | 0.8512 |
0.5993 | 13.0 | 559 | 2.2653 | 0.0394 | 0.18 | 0.0646 | 0.8583 |
0.5993 | 14.0 | 602 | 2.3001 | 0.0397 | 0.184 | 0.0653 | 0.8553 |
0.5993 | 15.0 | 645 | 2.3325 | 0.0395 | 0.182 | 0.0649 | 0.8597 |
Framework versions
- Transformers 4.30.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.13.3
- Downloads last month
- 8
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.