py315-pylingual-v3-segmenter
This model is a fine-tuned version of syssec-utd/py315-pylingual-v3-mlm on the syssec-utd/segmentation-py315-pylingual-v3-tokenized dataset. It achieves the following results on the evaluation set:
- Loss: 0.0078
- Precision: 0.9950
- Recall: 0.9947
- F1: 0.9948
- Accuracy: 0.9985
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: 2
- total_train_batch_size: 128
- total_eval_batch_size: 16
- 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 |
|---|---|---|---|---|---|---|---|
| 0.0144 | 1.0 | 40801 | 0.0085 | 0.9944 | 0.9934 | 0.9939 | 0.9982 |
| 0.0078 | 2.0 | 81602 | 0.0078 | 0.9950 | 0.9947 | 0.9948 | 0.9985 |
Framework versions
- Transformers 5.12.1
- Pytorch 2.12.0+cu130
- Datasets 5.0.0
- Tokenizers 0.22.2