metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: xlm-roberta-base-kyrgyzNER
results: []
xlm-roberta-base-kyrgyzNER
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3229
- Precision: 0.7042
- Recall: 0.6871
- F1: 0.6956
- Accuracy: 0.9119
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: 1e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 1
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 800
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 88 | 3.8135 | 0.0001 | 0.0009 | 0.0002 | 0.0002 |
| No log | 2.0 | 176 | 1.3867 | 0.0 | 0.0 | 0.0 | 0.7162 |
| No log | 3.0 | 264 | 0.9276 | 0.1684 | 0.0357 | 0.0590 | 0.7678 |
| No log | 4.0 | 352 | 0.6467 | 0.4470 | 0.2771 | 0.3421 | 0.8420 |
| No log | 5.0 | 440 | 0.5200 | 0.6394 | 0.5282 | 0.5785 | 0.8792 |
| 1.647 | 6.0 | 528 | 0.4383 | 0.6712 | 0.5918 | 0.6290 | 0.8927 |
| 1.647 | 7.0 | 616 | 0.3847 | 0.6724 | 0.6439 | 0.6578 | 0.9028 |
| 1.647 | 8.0 | 704 | 0.3586 | 0.6857 | 0.6575 | 0.6713 | 0.9061 |
| 1.647 | 9.0 | 792 | 0.3422 | 0.6786 | 0.6717 | 0.6751 | 0.9070 |
| 1.647 | 10.0 | 880 | 0.3229 | 0.7042 | 0.6871 | 0.6956 | 0.9119 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2