metadata
tags:
- generated_from_trainer
datasets:
- scientific_papers
model-index:
- name: reformer-big_patent-wikipedia-arxiv-16384
results: []
reformer-big_patent-wikipedia-arxiv-16384
This model is a fine-tuned version of robingeibel/reformer-big_patent-wikipedia-arxiv-16384 on the scientific_papers dataset. It achieves the following results on the evaluation set:
- Loss: 5.3635
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: 2.5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.3833 | 1.0 | 52747 | 5.3635 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1