https://github.com/marsggbo/torchline

Easy to use Pytorch
https://github.com/marsggbo/torchline

demo pipeline-framework pytorch pytorch-lightning torchline

Last synced: 4 days ago
JSON representation

Easy to use Pytorch

• Host: GitHub
• URL: https://github.com/marsggbo/torchline
• Owner: marsggbo
• License: mit
• Created: 2019-12-11T05:59:52.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-11-10T07:00:31.000Z (over 4 years ago)
• Last Synced: 2024-12-12T08:02:29.882Z (2 months ago)
• Topics: demo, pipeline-framework, pytorch, pytorch-lightning, torchline
• Language: Python
• Size: 188 KB
• Stars: 71
• Watchers: 6
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# torchline v0.3.0.4

> Easy to use Pytorch

> 

> Only one configure file is enough!

> 

> You can change anything you want just in only one configure file.

# Dependences

- Python>=3.6

- Pytorch>=1.3.1

- torchvision>=0.4.0,<0.5.0

- yacs==0.1.6

- pytorch-lightning<=0.7.6

# Install

- Before you install `torchline`, please make sure you have installed the above libraries.

- You can use `torchline` both in Linux and Windows.

```bash

pip install torchline

```

# Run demo

## train model with GPU0 and GPU 1

```python

cd projects/cifar10_demo

python main.py --config_file cifar10.yaml trainer.gpus [0,1]

```

## debug，add command line `trainer.fast_dev_run True`

```python

cd projects/cifar10_demo

python main.py --config_file cifar10.yaml trainer.gpus [0] trainer.fast_dev_run True

```

CIFAR demo uses ResNet50，which is trained for 72 epochs and achieved the best result (94.39% validation accuracy) at the epoch 54.

# Thanks

- [AutoML](https://zhuanlan.zhihu.com/automl)

- [pytorch-lightning](https://github.com/PyTorchLightning/pytorch-lightning)