metadata
license: apache-2.0
base_model: t5-base
tags:
- generated_from_trainer
model-index:
- name: Question_Generation_ComQ_onT5base
results: []
Question_Generation_ComQ_onT5base
This model is a fine-tuned version of t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4638
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7508 | 0.32 | 1000 | 0.5682 |
| 0.6081 | 0.64 | 2000 | 0.5212 |
| 0.5745 | 0.96 | 3000 | 0.4933 |
| 0.4907 | 1.27 | 4000 | 0.4872 |
| 0.4859 | 1.59 | 5000 | 0.4697 |
| 0.4786 | 1.91 | 6000 | 0.4630 |
| 0.4288 | 2.23 | 7000 | 0.4782 |
| 0.4264 | 2.55 | 8000 | 0.4671 |
| 0.4272 | 2.87 | 9000 | 0.4586 |
| 0.3975 | 3.19 | 10000 | 0.4642 |
| 0.3787 | 3.5 | 11000 | 0.4645 |
| 0.3937 | 3.82 | 12000 | 0.4534 |
| 0.3672 | 4.14 | 13000 | 0.4640 |
| 0.3529 | 4.46 | 14000 | 0.4628 |
| 0.3564 | 4.78 | 15000 | 0.4570 |
| 0.3509 | 5.1 | 16000 | 0.4618 |
| 0.3236 | 5.42 | 17000 | 0.4635 |
| 0.3324 | 5.73 | 18000 | 0.4592 |
| 0.3289 | 6.05 | 19000 | 0.4572 |
| 0.3094 | 6.37 | 20000 | 0.4637 |
| 0.3098 | 6.69 | 21000 | 0.4638 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0