py314-pylingual-v10-segmenter
This model is a fine-tuned version of syssec-utd/py314-pylingual-v10-mlm on the syssec-utd/segmentation-py314-pylingual-v10-tokenized dataset. It achieves the following results on the evaluation set:
- Loss: 0.0162
- Precision: 0.9886
- Recall: 0.9893
- F1: 0.9890
- Accuracy: 0.9972
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: 64
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 3
- total_train_batch_size: 192
- total_eval_batch_size: 24
- 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: linear
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0237 | 1.0 | 29337 | 0.0196 | 0.9873 | 0.9884 | 0.9879 | 0.9969 |
| 0.0126 | 2.0 | 58674 | 0.0162 | 0.9886 | 0.9893 | 0.9890 | 0.9972 |
Framework versions
- Transformers 5.17.0
- Pytorch 2.14.0+cu130
- Datasets 5.0.1
- Tokenizers 0.23.2