metadata
library_name: transformers
license: mit
base_model: synapti/nci-ner-v2-stage-a
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
model-index:
- name: nci-ner-v2
results: []
nci-ner-v2
This model is a fine-tuned version of synapti/nci-ner-v2-stage-a on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5857
- Precision: 0.8651
- Recall: 0.9020
- F1: 0.8832
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: 5e-06
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 3
- label_smoothing_factor: 0.05
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 |
|---|---|---|---|---|---|---|
| 0.7627 | 0.0346 | 100 | 0.5908 | 0.8479 | 0.8877 | 0.8673 |
| 0.5656 | 0.0692 | 200 | 0.4720 | 0.8733 | 0.8316 | 0.8519 |
| 0.5146 | 0.1038 | 300 | 0.4855 | 0.7862 | 0.7272 | 0.7555 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.10.0+cu128
- Datasets 2.21.0
- Tokenizers 0.20.3