metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: edos-2023-baseline-bert-base-multilingual-uncased-label_vector
results: []
edos-2023-baseline-bert-base-multilingual-uncased-label_vector
This model is a fine-tuned version of bert-base-multilingual-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0875
- F1: 0.3778
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: 46
- eval_batch_size: 46
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 2.0532 | 1.69 | 100 | 1.7911 | 0.1191 |
| 1.6923 | 3.39 | 200 | 1.4958 | 0.2426 |
| 1.48 | 5.08 | 300 | 1.2994 | 0.2989 |
| 1.3375 | 6.78 | 400 | 1.1620 | 0.3514 |
| 1.2391 | 8.47 | 500 | 1.0875 | 0.3778 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2