checkpoints_30_9_microsoft_deberta_V1.0_384
This model is a fine-tuned version of VuongQuoc/checkpoints_30_9_microsoft_deberta_V1.0_384 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.5746
- Map@3: 0.7625
- Accuracy: 0.655
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-07
- train_batch_size: 2
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 1200
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 | Accuracy |
|---|---|---|---|---|---|
| 1.6081 | 0.05 | 100 | 1.6083 | 0.7092 | 0.585 |
| 1.6107 | 0.11 | 200 | 1.6078 | 0.7375 | 0.625 |
| 1.6077 | 0.16 | 300 | 1.6070 | 0.7517 | 0.65 |
| 1.6097 | 0.21 | 400 | 1.6055 | 0.7542 | 0.645 |
| 1.6083 | 0.27 | 500 | 1.6030 | 0.7650 | 0.65 |
| 1.6006 | 0.32 | 600 | 1.5989 | 0.7733 | 0.665 |
| 1.5932 | 0.37 | 700 | 1.5927 | 0.7742 | 0.66 |
| 1.5881 | 0.43 | 800 | 1.5858 | 0.7742 | 0.665 |
| 1.578 | 0.48 | 900 | 1.5800 | 0.7708 | 0.66 |
| 1.5717 | 0.53 | 1000 | 1.5763 | 0.7658 | 0.655 |
| 1.5677 | 0.59 | 1100 | 1.5748 | 0.7625 | 0.655 |
| 1.5666 | 0.64 | 1200 | 1.5746 | 0.7625 | 0.655 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3