metadata
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: GPT2_v5
results: []
GPT2_v5
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7670
- Precision: 0.7725
- Recall: 0.8367
- F1: 0.4733
- Accuracy: 0.7646
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 1.2212 | 1.0 | 1012 | 0.7874 | 0.7557 | 0.7560 | 0.4041 | 0.7150 |
| 0.7162 | 2.0 | 2024 | 0.7007 | 0.7495 | 0.8714 | 0.4855 | 0.7647 |
| 0.6241 | 3.0 | 3036 | 0.6799 | 0.7681 | 0.8532 | 0.4804 | 0.7702 |
| 0.5545 | 4.0 | 4048 | 0.6997 | 0.7635 | 0.8658 | 0.4814 | 0.7714 |
| 0.4963 | 5.0 | 5060 | 0.7186 | 0.7696 | 0.8470 | 0.4764 | 0.7669 |
| 0.449 | 6.0 | 6072 | 0.7436 | 0.7711 | 0.8382 | 0.4731 | 0.7644 |
| 0.4182 | 7.0 | 7084 | 0.7670 | 0.7725 | 0.8367 | 0.4733 | 0.7646 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.1.0
- Tokenizers 0.12.1