metadata
library_name: transformers
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- f1
- precision
- recall
model-index:
- name: roberta-base-multilingual-sentiment
results: []
roberta-base-multilingual-sentiment
This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4341
- F1: 0.8169
- Precision: 0.8176
- Recall: 0.8163
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: 5e-05
- train_batch_size: 512
- eval_batch_size: 512
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 2
- total_train_batch_size: 1024
- optimizer: Use adamw_torch_fused 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: 0.01
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall |
|---|---|---|---|---|---|---|
| 0.893 | 1.0 | 3074 | 0.4333 | 0.8038 | 0.8076 | 0.8022 |
| 0.8335 | 2.0 | 6148 | 0.4180 | 0.8150 | 0.8152 | 0.8149 |
| 0.7149 | 3.0 | 9222 | 0.4238 | 0.8162 | 0.8168 | 0.8158 |
| 0.7298 | 4.0 | 12296 | 0.4258 | 0.8168 | 0.8178 | 0.8160 |
| 0.6729 | 5.0 | 15370 | 0.4341 | 0.8169 | 0.8176 | 0.8163 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.4.0+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0