TorchZQ: A simple PyTorch experiment runner.
Project description
TorchZQ: A simple PyTorch experiment runner.
Zouqi (『走起』 in Chinese) means "let's go". When you zouqi your experiment, the experiment will go with you.
Installation
Install from PyPI:
pip install torchzq
Install the latest version:
pip install git+https://github.com/enhuiz/torchzq@master
Run an Example
Training
$ zouqi example/config/mnist.yml train
Testing
$ zouqi example/config/mnist.yml test
TensorBoard
$ tensorboard --logdir logs
Supported Features
- Model checkpoints
- Logging
- Gradient accumulation
- Configuration file
- Configuration file inheritance
- TensorBoard
- (c)GAN training (WGAN-GP)
- FP16
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
torchzq-1.0.5.dev1.tar.gz
(10.6 kB
view hashes)
Built Distribution
Close
Hashes for torchzq-1.0.5.dev1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 845a657fd0c7ce0f74e7c57a76a7bc70e850ac3b9ee6796877754d098627ecf7 |
|
MD5 | 7ebaa8769110f408cc0b37ba76f2afef |
|
BLAKE2b-256 | 78b2f9484540a8f4b1b0bf6484f8618d0c5e82f37c33514f5a697fd89eed90fa |