metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: t5_number_v7_new_data
results: []
t5_number_v7_new_data
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0108
- Cer: 0.6315
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: 4e-05
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.0163 | 1.0 | 858 | 0.0141 | 0.8245 |
| 0.0135 | 2.0 | 1716 | 0.0121 | 0.7102 |
| 0.012 | 3.0 | 2574 | 0.0115 | 0.6647 |
| 0.0115 | 4.0 | 3432 | 0.0111 | 0.6577 |
| 0.0115 | 5.0 | 4290 | 0.0110 | 0.6490 |
| 0.0106 | 6.0 | 5148 | 0.0108 | 0.6461 |
| 0.0103 | 7.0 | 6006 | 0.0108 | 0.6362 |
| 0.0103 | 8.0 | 6864 | 0.0108 | 0.6362 |
| 0.0101 | 9.0 | 7722 | 0.0108 | 0.6327 |
| 0.01 | 10.0 | 8580 | 0.0108 | 0.6315 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3