checkpoints_28_9_microsoft_deberta_V2
This model is a fine-tuned version of microsoft/deberta-v3-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5675
- Map@3: 0.8842
- Accuracy: 0.815
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: 2
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 | Accuracy |
|---|---|---|---|---|---|
| 1.0011 | 0.11 | 100 | 0.8842 | 0.8258 | 0.74 |
| 0.8398 | 0.21 | 200 | 0.6978 | 0.8667 | 0.79 |
| 0.8414 | 0.32 | 300 | 0.6337 | 0.8625 | 0.795 |
| 0.7461 | 0.43 | 400 | 0.6609 | 0.8600 | 0.775 |
| 0.7131 | 0.53 | 500 | 0.6329 | 0.8758 | 0.805 |
| 0.6891 | 0.64 | 600 | 0.6157 | 0.8892 | 0.83 |
| 0.6969 | 0.75 | 700 | 0.5917 | 0.8808 | 0.805 |
| 0.6775 | 0.85 | 800 | 0.5698 | 0.8817 | 0.81 |
| 0.6534 | 0.96 | 900 | 0.5675 | 0.8842 | 0.815 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3