metadata
license: cc-by-4.0
base_model: EMBEDDIA/crosloengual-bert
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: fine_tuned_boolq_bert_croslo
results: []
fine_tuned_boolq_bert_croslo
This model is a fine-tuned version of EMBEDDIA/crosloengual-bert on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3270
- Accuracy: 0.8333
- F1: 0.8243
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 400
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.5567 | 4.1667 | 50 | 0.5262 | 0.7222 | 0.6523 |
| 0.1098 | 8.3333 | 100 | 0.8949 | 0.8333 | 0.8243 |
| 0.0031 | 12.5 | 150 | 1.2237 | 0.7778 | 0.7778 |
| 0.0011 | 16.6667 | 200 | 1.2641 | 0.7778 | 0.7778 |
| 0.0008 | 20.8333 | 250 | 1.2343 | 0.8333 | 0.8243 |
| 0.0007 | 25.0 | 300 | 1.2852 | 0.8333 | 0.8243 |
| 0.0005 | 29.1667 | 350 | 1.3133 | 0.8333 | 0.8243 |
| 0.0005 | 33.3333 | 400 | 1.3270 | 0.8333 | 0.8243 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0
- Datasets 2.19.0
- Tokenizers 0.19.1