metadata
library_name: transformers
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: xlm-roberta-base-finetuned
results: []
xlm-roberta-base-finetuned
This model is a fine-tuned version of xlm-roberta-base on the 'Contradictory, My Dear Watson' kaggle competition dataset. It achieves the following results on the evaluation set:
- Loss: 0.7526
- Accuracy: 0.7075
- F1: 0.7080
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 1.0 | 76 | 1.0440 | 0.5004 | 0.5007 |
| 1.03 | 2.0 | 152 | 0.8601 | 0.6370 | 0.6374 |
| 1.03 | 3.0 | 228 | 0.7761 | 0.6712 | 0.6727 |
| 0.7667 | 4.0 | 304 | 0.7526 | 0.6951 | 0.6967 |
| 0.7667 | 5.0 | 380 | 0.7526 | 0.7075 | 0.7080 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0