metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: checkpoints_28_9_microsoft_deberta_V4
results: []
checkpoints_28_9_microsoft_deberta_V4
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: 1.2854
- Map@3: 0.5483
- Accuracy: 0.435
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: 0.2
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 | Accuracy |
|---|---|---|---|---|---|
| 1.2365 | 0.11 | 100 | 1.0631 | 0.7583 | 0.64 |
| 0.8608 | 0.21 | 200 | 0.7329 | 0.8383 | 0.75 |
| 0.8527 | 0.32 | 300 | 0.6985 | 0.8575 | 0.78 |
| 0.744 | 0.43 | 400 | 0.6498 | 0.8625 | 0.785 |
| 0.7686 | 0.53 | 500 | 0.7450 | 0.8575 | 0.765 |
| 1.4098 | 0.64 | 600 | 1.3030 | 0.5575 | 0.4 |
| 1.4246 | 0.75 | 700 | 1.3018 | 0.5575 | 0.435 |
| 1.3987 | 0.85 | 800 | 1.2906 | 0.5450 | 0.41 |
| 1.4121 | 0.96 | 900 | 1.2854 | 0.5483 | 0.435 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3