STL-Full-FT-mt5-base-sentiment
This model is a fine-tuned version of google/mt5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1774
- Accuracy: 0.6047
- Macro F1: 0.5807
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 300
- num_epochs: 10
- label_smoothing_factor: 0.05
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 |
|---|---|---|---|---|---|
| 12.7883 | 1.0 | 189 | 3.3935 | 0.3766 | 0.1378 |
| 1.4209 | 2.0 | 378 | 1.2932 | 0.4389 | 0.3079 |
| 1.3251 | 3.0 | 567 | 1.2075 | 0.5748 | 0.5294 |
| 1.2981 | 4.0 | 756 | 1.1993 | 0.6047 | 0.5993 |
| 1.2665 | 5.0 | 945 | 1.1933 | 0.5786 | 0.5598 |
| 1.2545 | 6.0 | 1134 | 1.1868 | 0.6035 | 0.5850 |
| 1.2489 | 7.0 | 1323 | 1.1774 | 0.6047 | 0.5807 |
Framework versions
- Transformers 5.16.1
- Pytorch 2.11.0+cu128
- Datasets 4.8.5
- Tokenizers 0.23.1
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for tadiecool29/STL-Full-FT-mt5-base-sentiment
Base model
google/mt5-base