<!-- 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. -->
hindi-bert-ner
This model is a fine-tuned version of bert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0337
- Precision: 0.8427
- Recall: 0.9013
- F1: 0.8710
- Accuracy: 0.9919
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.3231 | 1.0 | 882 | 0.1740 | 0.4463 | 0.5576 | 0.4957 | 0.9588 |
0.1442 | 2.0 | 1764 | 0.0758 | 0.6599 | 0.7765 | 0.7135 | 0.9811 |
0.0472 | 3.0 | 2646 | 0.0337 | 0.8427 | 0.9013 | 0.8710 | 0.9919 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.14.5
- Tokenizers 0.13.3