|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: swin-food102 |
|
results: [] |
|
datasets: |
|
- juliensimon/food102 |
|
--- |
|
|
|
# swin-food102 |
|
|
|
This model is a fine-tuned version of [juliensimon/autotrain-food101-1471154053](https://huggingface.co/juliensimon/autotrain-food101-1471154053) on the [food102](https://huggingface.co/datasets/juliensimon/food102) dataset, namely the [food101](https://huggingface.co/datasets/food101) dataset with an extra class generated with a Stable Diffusion model. |
|
|
|
A detailed walk-through is available on [YouTube](https://youtu.be/sIe0eo3fYQ4). |
|
|
|
The achieves the following results on the evaluation set: |
|
- Loss: 0.2510 |
|
- Accuracy: 0.9338 |
|
|
|
## 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: 1e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 1.1648 | 1.0 | 597 | 0.3118 | 0.9218 | |
|
| 0.31 | 2.0 | 1194 | 0.2606 | 0.9322 | |
|
| 0.2488 | 3.0 | 1791 | 0.2510 | 0.9338 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.23.1 |
|
- Pytorch 1.12.1+cu102 |
|
- Datasets 2.4.0 |
|
- Tokenizers 0.13.1 |
|
|