metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: tinygram
results: []
tinygram
This model is a fine-tuned version of google/t5-efficient-tiny-nh32 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4318
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 8
- 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 |
|---|---|---|---|
| 1.4129 | 1.0 | 1875 | 0.7042 |
| 0.8989 | 2.0 | 3750 | 0.5567 |
| 0.7689 | 3.0 | 5625 | 0.5098 |
| 0.7041 | 4.0 | 7500 | 0.4821 |
| 0.6824 | 5.0 | 9375 | 0.4634 |
| 0.6456 | 6.0 | 11250 | 0.4511 |
| 0.6241 | 7.0 | 13125 | 0.4440 |
| 0.6194 | 8.0 | 15000 | 0.4364 |
| 0.609 | 9.0 | 16875 | 0.4328 |
| 0.6042 | 10.0 | 18750 | 0.4318 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2