my_ner_model
This model is a fine-tuned version of distilbert-base-uncased on the wnut_17 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3321
- Precision: 0.5759
- Recall: 0.4254
- F1: 0.4893
- Accuracy: 0.9489
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 213 | 0.2766 | 0.5657 | 0.2873 | 0.3811 | 0.9399 |
No log | 2.0 | 426 | 0.2634 | 0.6267 | 0.3438 | 0.4440 | 0.9447 |
0.1828 | 3.0 | 639 | 0.3173 | 0.6354 | 0.2697 | 0.3787 | 0.9432 |
0.1828 | 4.0 | 852 | 0.3102 | 0.6018 | 0.3670 | 0.4560 | 0.9470 |
0.0475 | 5.0 | 1065 | 0.3047 | 0.5914 | 0.3957 | 0.4742 | 0.9478 |
0.0475 | 6.0 | 1278 | 0.3226 | 0.5927 | 0.4059 | 0.4818 | 0.9481 |
0.0475 | 7.0 | 1491 | 0.3109 | 0.5709 | 0.4291 | 0.4899 | 0.9486 |
0.0212 | 8.0 | 1704 | 0.3609 | 0.6200 | 0.3855 | 0.4754 | 0.9474 |
0.0212 | 9.0 | 1917 | 0.3236 | 0.5587 | 0.4365 | 0.4901 | 0.9486 |
0.0117 | 10.0 | 2130 | 0.3321 | 0.5759 | 0.4254 | 0.4893 | 0.9489 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.14.1
- Downloads last month
- 89
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for yadvi/my_ner_model
Base model
distilbert/distilbert-base-uncasedDataset used to train yadvi/my_ner_model
Evaluation results
- Precision on wnut_17test set self-reported0.576
- Recall on wnut_17test set self-reported0.425
- F1 on wnut_17test set self-reported0.489
- Accuracy on wnut_17test set self-reported0.949