metadata
library_name: transformers
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
model-index:
- name: bert-base-multilingual-cased-ukr
results: []
bert-base-multilingual-cased-ukr
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1717
- Accuracy: 0.8350
- F1 Binary: 0.4058
- Precision: 0.2966
- Recall: 0.6423
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: 3e-05
- train_batch_size: 64
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 36
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Binary | Precision | Recall |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 185 | 0.1355 | 0.6984 | 0.2684 | 0.1705 | 0.6308 |
| No log | 2.0 | 370 | 0.1206 | 0.7969 | 0.3428 | 0.2393 | 0.6038 |
| 0.12 | 3.0 | 555 | 0.1118 | 0.8111 | 0.3792 | 0.2664 | 0.6577 |
| 0.12 | 4.0 | 740 | 0.1717 | 0.8350 | 0.4058 | 0.2966 | 0.6423 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0