metadata
license: mit
base_model: VuongQuoc/checkpoints_28_9_microsoft_deberta_V2
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: checkpoints_28_9_microsoft_deberta_V2.1
results: []
checkpoints_28_9_microsoft_deberta_V2.1
This model is a fine-tuned version of VuongQuoc/checkpoints_28_9_microsoft_deberta_V2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5671
- Map@3: 0.875
- Accuracy: 0.795
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
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 | Accuracy |
|---|---|---|---|---|---|
| 0.3404 | 0.11 | 100 | 0.6107 | 0.8683 | 0.785 |
| 0.1782 | 0.21 | 200 | 0.8483 | 0.8392 | 0.74 |
| 0.1541 | 0.32 | 300 | 0.8127 | 0.8558 | 0.78 |
| 0.1423 | 0.43 | 400 | 0.7419 | 0.8517 | 0.765 |
| 0.2283 | 0.53 | 500 | 0.7557 | 0.8542 | 0.765 |
| 0.4409 | 0.64 | 600 | 0.6255 | 0.8733 | 0.795 |
| 0.6855 | 0.75 | 700 | 0.5831 | 0.87 | 0.795 |
| 0.6876 | 0.85 | 800 | 0.5710 | 0.875 | 0.795 |
| 0.6422 | 0.96 | 900 | 0.5671 | 0.875 | 0.795 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3