metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: 10k_test3_nli_finetuned_canine_c
results: []
10k_test3_nli_finetuned_canine_c
This model is a fine-tuned version of google/canine-c on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0990
- Accuracy: 0.3247
- F1 Weighted: 0.1591
- Precision Weighted: 0.1054
- Recall Weighted: 0.3247
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Weighted | Precision Weighted | Recall Weighted |
|---|---|---|---|---|---|---|---|
| 1.1017 | 1.0 | 312 | 1.1029 | 0.3247 | 0.1591 | 0.1054 | 0.3247 |
| 1.0999 | 2.0 | 625 | 1.0978 | 0.3533 | 0.1845 | 0.1248 | 0.3533 |
| 1.099 | 3.0 | 936 | 1.0990 | 0.3247 | 0.1591 | 0.1054 | 0.3247 |
Framework versions
- Transformers 4.27.2
- Pytorch 2.0.0+cu117
- Datasets 2.10.1
- Tokenizers 0.13.2