metadata
license: mit
base_model: BigTMiami/micro_base_help_tapt_pretrain_model
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: micro_base_help_class_tapt_seed_3
results: []
micro_base_help_class_tapt_seed_3
This model is a fine-tuned version of BigTMiami/micro_base_help_tapt_pretrain_model on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3588
- Accuracy: 0.8398
- F1 Macro: 0.6881
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: 16
- eval_batch_size: 16
- seed: 3
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|---|---|---|---|---|---|
| 0.3463 | 1.0 | 313 | 0.3531 | 0.8586 | 0.6005 |
| 0.303 | 2.0 | 626 | 0.3318 | 0.8616 | 0.5976 |
| 0.2288 | 3.0 | 939 | 0.3732 | 0.8288 | 0.6841 |
| 0.1718 | 4.0 | 1252 | 0.6695 | 0.8586 | 0.5697 |
| 0.1466 | 5.0 | 1565 | 0.7902 | 0.855 | 0.6347 |
| 0.0853 | 6.0 | 1878 | 0.9930 | 0.847 | 0.6702 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2