electra-tagalog-base-uncased-discriminator-ner-v1
This model is a fine-tuned version of jcblaise/electra-tagalog-base-uncased-discriminator on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2224
- Precision: 0.9458
- Recall: 0.9221
- F1: 0.9338
- Accuracy: 0.9584
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 | 205 | 0.5797 | 0.7108 | 0.5981 | 0.6496 | 0.8354 |
No log | 2.0 | 410 | 0.3364 | 0.8057 | 0.8323 | 0.8188 | 0.9069 |
0.4976 | 3.0 | 615 | 0.2628 | 0.8653 | 0.8727 | 0.8689 | 0.9317 |
0.4976 | 4.0 | 820 | 0.2323 | 0.9173 | 0.8823 | 0.8994 | 0.9440 |
0.1335 | 5.0 | 1025 | 0.2212 | 0.9154 | 0.9102 | 0.9128 | 0.9473 |
0.1335 | 6.0 | 1230 | 0.2235 | 0.9335 | 0.9102 | 0.9217 | 0.9524 |
0.1335 | 7.0 | 1435 | 0.2207 | 0.9169 | 0.9221 | 0.9195 | 0.9541 |
0.0509 | 8.0 | 1640 | 0.2214 | 0.9414 | 0.9221 | 0.9316 | 0.9582 |
0.0509 | 9.0 | 1845 | 0.2266 | 0.9416 | 0.9255 | 0.9335 | 0.9571 |
0.0301 | 10.0 | 2050 | 0.2224 | 0.9458 | 0.9221 | 0.9338 | 0.9584 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
- Downloads last month
- 2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.