32 lines
1.0 KiB
Markdown
32 lines
1.0 KiB
Markdown
|
## MNIST Examples
|
||
|
|
||
|
Here are two MNIST classifiers implemented in PyTorch.
|
||
|
The first one is implemented in pure PyTorch, but isn't easy to scale.
|
||
|
The second one is using [Lightning Lite](https://pytorch-lightning.readthedocs.io/en/stable/starter/lightning_lite.html) to accelerate and scale the model.
|
||
|
|
||
|
#### 1. Image Classifier with Vanilla PyTorch
|
||
|
|
||
|
Trains a simple CNN over MNIST using vanilla PyTorch. It only supports singe GPU training.
|
||
|
|
||
|
```bash
|
||
|
# CPU
|
||
|
python image_classifier_1_pytorch.py
|
||
|
```
|
||
|
|
||
|
______________________________________________________________________
|
||
|
|
||
|
#### 2. Image Classifier with Lightning Lite
|
||
|
|
||
|
This script shows you how to scale the pure PyTorch code to enable GPU and multi-GPU training using [Lightning Lite](https://pytorch-lightning.readthedocs.io/en/stable/starter/lightning_lite.html).
|
||
|
|
||
|
```bash
|
||
|
# CPU
|
||
|
lightning run model image_classifier_2_lite.py
|
||
|
|
||
|
# GPU (CUDA or M1 Mac)
|
||
|
lightning run model image_classifier_2_lite.py --accelerator=gpu
|
||
|
|
||
|
# Multiple GPUs
|
||
|
lightning run model image_classifier_2_lite.py --accelerator=gpu --devices=4
|
||
|
```
|