metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: taskA-DeBERTa-large-1.0.0
results: []
taskA-DeBERTa-large-1.0.0
This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1001
- Accuracy: 0.7924
- Precision: 0.5855
- Recall: 0.5821
- F1: 0.5838
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: 4e-06
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.5231 | 0.39 | 500 | 0.5201 | 0.7967 | 0.7519 | 0.2795 | 0.4076 |
| 0.4515 | 0.78 | 1000 | 0.5449 | 0.8032 | 0.7937 | 0.2882 | 0.4228 |
| 0.4128 | 1.16 | 1500 | 0.6890 | 0.8190 | 0.6805 | 0.5216 | 0.5905 |
| 0.4025 | 1.55 | 2000 | 0.9337 | 0.7787 | 0.5481 | 0.6571 | 0.5976 |
| 0.4251 | 1.94 | 2500 | 0.8829 | 0.7981 | 0.6070 | 0.5476 | 0.5758 |
| 0.2864 | 2.33 | 3000 | 1.1001 | 0.7924 | 0.5855 | 0.5821 | 0.5838 |
| 0.3186 | 2.71 | 3500 | 1.1268 | 0.7794 | 0.5504 | 0.6455 | 0.5942 |
| 0.2679 | 3.1 | 4000 | 1.0378 | 0.8039 | 0.6093 | 0.6023 | 0.6058 |
| 0.1893 | 3.49 | 4500 | 1.1135 | 0.7996 | 0.6062 | 0.5677 | 0.5863 |
| 0.2037 | 3.88 | 5000 | 1.1569 | 0.7945 | 0.5886 | 0.5937 | 0.5911 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2