<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
pos-ner-tagging-v3
This model is a fine-tuned version of om-ashish-soni/pos-ner-tagging-v2 on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6356
- Precision: 0.9339
- Recall: 0.9374
- F1: 0.9357
- Accuracy: 0.9273
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 439 | 0.6415 | 0.9341 | 0.9367 | 0.9354 | 0.9265 |
0.0078 | 2.0 | 878 | 0.6372 | 0.9327 | 0.9363 | 0.9345 | 0.9259 |
0.006 | 3.0 | 1317 | 0.6283 | 0.9338 | 0.9373 | 0.9356 | 0.9274 |
0.0036 | 4.0 | 1756 | 0.6356 | 0.9339 | 0.9374 | 0.9357 | 0.9273 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3