py314-pylingual-v4-segmenter
This model is a fine-tuned version of syssec-utd/py314-pylingual-v4-mlm on the syssec-utd/segmentation-py314-pylingual-v4-tokenized dataset. It achieves the following results on the evaluation set:
- Loss: 0.0022
- Precision: 0.9949
- Recall: 0.9972
- F1: 0.9961
- Accuracy: 0.9992
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: 48
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 3
- total_train_batch_size: 144
- 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 |
|---|---|---|---|---|---|---|---|
| 0.0069 | 1.0 | 42896 | 0.0022 | 0.9940 | 0.9968 | 0.9954 | 0.9991 |
| 0.0039 | 2.0 | 85792 | 0.0022 | 0.9949 | 0.9972 | 0.9961 | 0.9992 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.1+cu128
- Datasets 4.4.1
- Tokenizers 0.22.1