metadata
license: mit
base_model: MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: mDeBERTa-v3-base-xnli-multilingual-nli-2mil7-energy
results: []
mDeBERTa-v3-base-xnli-multilingual-nli-2mil7-energy
This model is a fine-tuned version of MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2358
- Accuracy: 0.9580
- Precision: 0.9583
- Recall: 0.9578
- F1: 0.9580
- Ratio: 0.4803
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- lr_scheduler_warmup_steps: 3
- num_epochs: 3
- label_smoothing_factor: 0.01
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Ratio |
|---|---|---|---|---|---|---|---|---|
| 0.5219 | 0.43 | 400 | 0.3524 | 0.8954 | 0.8972 | 0.8946 | 0.8951 | 0.4577 |
| 0.4069 | 0.86 | 800 | 0.3178 | 0.9249 | 0.9250 | 0.9246 | 0.9248 | 0.4809 |
| 0.2326 | 1.29 | 1200 | 0.3055 | 0.9355 | 0.9360 | 0.9351 | 0.9354 | 0.4740 |
| 0.2045 | 1.72 | 1600 | 0.2847 | 0.9455 | 0.9457 | 0.9453 | 0.9455 | 0.4803 |
| 0.1423 | 2.15 | 2000 | 0.2477 | 0.9555 | 0.9555 | 0.9556 | 0.9555 | 0.4903 |
| 0.0935 | 2.58 | 2400 | 0.2367 | 0.9599 | 0.9598 | 0.9600 | 0.9599 | 0.4922 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.7
- Tokenizers 0.13.3