metadata
library_name: transformers
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
datasets:
- turkish_ner
metrics:
- f1
- precision
- recall
- accuracy
model-index:
- name: turkish-ner-fold-bBERT5
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: turkish_ner
type: turkish_ner
config: default
split: train
args: default
metrics:
- name: F1
type: f1
value: 0.6423017107309488
- name: Precision
type: precision
value: 0.6300533943554538
- name: Recall
type: recall
value: 0.655035685963521
- name: Accuracy
type: accuracy
value: 0.9049387731414311
turkish-ner-fold-bBERT5
This model is a fine-tuned version of xlm-roberta-base on the turkish_ner dataset. It achieves the following results on the evaluation set:
- Loss: 0.3151
- F1: 0.6423
- Precision: 0.6301
- Recall: 0.6550
- Accuracy: 0.9049
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.4095 | 1.0 | 500 | 0.3264 | 0.5578 | 0.5579 | 0.5577 | 0.8747 |
| 0.2751 | 2.0 | 1000 | 0.2901 | 0.6305 | 0.6227 | 0.6385 | 0.8927 |
| 0.2126 | 3.0 | 1500 | 0.2820 | 0.6337 | 0.6356 | 0.6318 | 0.8996 |
| 0.1636 | 4.0 | 2000 | 0.2948 | 0.6507 | 0.6514 | 0.6501 | 0.9038 |
| 0.1275 | 5.0 | 2500 | 0.3136 | 0.6524 | 0.6445 | 0.6604 | 0.9023 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0