metadata
library_name: transformers
license: apache-2.0
base_model: distilbert/distilbert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: multilingual_dbert_linsearch_only_abstract
results: []
multilingual_dbert_linsearch_only_abstract
This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1210
- Accuracy: 0.6499
- F1 Macro: 0.5631
- Precision Macro: 0.5636
- Recall Macro: 0.5672
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
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- 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: cosine
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Precision Macro | Recall Macro |
|---|---|---|---|---|---|---|---|
| 2.4454 | 1.0 | 1233 | 1.4217 | 0.5971 | 0.4617 | 0.5194 | 0.4599 |
| 1.3605 | 2.0 | 2466 | 1.1851 | 0.6360 | 0.5358 | 0.5534 | 0.5358 |
| 1.1562 | 3.0 | 3699 | 1.1435 | 0.6424 | 0.5511 | 0.5580 | 0.5552 |
| 1.0514 | 4.0 | 4932 | 1.1216 | 0.6487 | 0.5621 | 0.5628 | 0.5673 |
| 0.9556 | 4.9962 | 6160 | 1.1210 | 0.6499 | 0.5631 | 0.5636 | 0.5672 |
Framework versions
- Transformers 4.50.1
- Pytorch 2.5.1+cu121
- Datasets 3.4.1
- Tokenizers 0.21.1