metadata
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- precision
- recall
model-index:
- name: lex-cross-encoder-mbert-10neg
results: []
lex-cross-encoder-mbert-10neg
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4360
- Precision: 0.6020
- Recall: 0.8593
- F2: 0.7917
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: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 128
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F2 |
|---|---|---|---|---|---|---|
| 0.4572 | 1.0 | 2317 | 0.4705 | 0.4735 | 0.8620 | 0.7405 |
| 0.4283 | 2.0 | 4634 | 0.4515 | 0.4774 | 0.9124 | 0.7718 |
| 0.4115 | 3.0 | 6951 | 0.4485 | 0.4796 | 0.9201 | 0.7773 |
| 0.4021 | 4.0 | 9268 | 0.4387 | 0.5217 | 0.9068 | 0.7902 |
| 0.3918 | 5.0 | 11585 | 0.4466 | 0.6111 | 0.8242 | 0.7705 |
| 0.3879 | 6.0 | 13902 | 0.4337 | 0.5783 | 0.8767 | 0.7947 |
| 0.383 | 7.0 | 16219 | 0.4336 | 0.5633 | 0.8907 | 0.7980 |
| 0.3781 | 8.0 | 18536 | 0.4354 | 0.5929 | 0.8660 | 0.7930 |
| 0.3767 | 9.0 | 20853 | 0.4353 | 0.5980 | 0.8636 | 0.7931 |
| 0.3712 | 10.0 | 23170 | 0.4360 | 0.6020 | 0.8593 | 0.7917 |
Framework versions
- Transformers 4.39.1
- Pytorch 2.5.1+cu121
- Datasets 3.6.0
- Tokenizers 0.15.2