metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- fever
metrics:
- f1
base_model: distilbert-base-cased
model-index:
- name: destilbert_fever_nli
results: []
destilbert_fever_nli
This model is a fine-tuned version of distilbert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.5463
- F1: 0.6747
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | F1 |
---|---|---|---|---|
No log | 1.0 | 235 | 1.2711 | 0.6671 |
No log | 2.0 | 470 | 1.8000 | 0.6538 |
0.1341 | 3.0 | 705 | 1.6965 | 0.6770 |
0.1341 | 4.0 | 940 | 1.8415 | 0.6619 |
0.068 | 5.0 | 1175 | 1.7477 | 0.6682 |
0.068 | 6.0 | 1410 | 2.2007 | 0.6695 |
0.0435 | 7.0 | 1645 | 2.3327 | 0.6705 |
0.0435 | 8.0 | 1880 | 2.3927 | 0.6729 |
0.015 | 9.0 | 2115 | 2.4978 | 0.6721 |
0.015 | 10.0 | 2350 | 2.5463 | 0.6747 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2