metadata
library_name: transformers
license: mit
base_model: lotusbro/x5-ner
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: x5-ner-weighted
results: []
x5-ner-weighted
This model is a fine-tuned version of lotusbro/x5-ner on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8575
- Precision: 0.9465
- Recall: 0.9594
- F1: 0.9529
- Accuracy: 0.9505
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
- 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: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1246 | 1.0 | 3066 | 0.8009 | 0.9367 | 0.9480 | 0.9423 | 0.9441 |
| 0.1415 | 2.0 | 6132 | 0.6241 | 0.9399 | 0.9521 | 0.9460 | 0.9466 |
| 0.1218 | 3.0 | 9198 | 0.5923 | 0.9426 | 0.9534 | 0.9480 | 0.9475 |
| 0.0877 | 4.0 | 12264 | 0.7352 | 0.9420 | 0.9537 | 0.9478 | 0.9478 |
| 0.0725 | 5.0 | 15330 | 0.6434 | 0.9443 | 0.9518 | 0.9480 | 0.9493 |
| 0.0337 | 6.0 | 18396 | 0.7521 | 0.9478 | 0.9562 | 0.9520 | 0.9509 |
| 0.0287 | 7.0 | 21462 | 0.8421 | 0.9403 | 0.9540 | 0.9471 | 0.9460 |
| 0.0125 | 8.0 | 24528 | 0.8045 | 0.9497 | 0.9591 | 0.9544 | 0.9501 |
| 0.0105 | 9.0 | 27594 | 0.8138 | 0.9462 | 0.9597 | 0.9529 | 0.9517 |
| 0.0044 | 10.0 | 30660 | 0.8575 | 0.9465 | 0.9594 | 0.9529 | 0.9505 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.7.1+cu118
- Datasets 3.6.0
- Tokenizers 0.21.4