metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: 1_microsoft_deberta_V1.0
results: []
1_microsoft_deberta_V1.0
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: 1.0849
- Map@3: 0.7725
- Accuracy: 0.665
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: 25
- total_train_batch_size: 50
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 60
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 | Accuracy |
|---|---|---|---|---|---|
| 1.6098 | 0.01 | 10 | 1.6090 | 0.6108 | 0.475 |
| 1.6057 | 0.02 | 20 | 1.6027 | 0.7375 | 0.625 |
| 1.5615 | 0.03 | 30 | 1.4516 | 0.7458 | 0.64 |
| 1.2061 | 0.03 | 40 | 1.2130 | 0.7292 | 0.595 |
| 1.1028 | 0.04 | 50 | 1.0947 | 0.765 | 0.65 |
| 1.0682 | 0.05 | 60 | 1.0849 | 0.7725 | 0.665 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3