Torch_Template - A PyTorch template with commonly used models and tools
Project description
torch_template
A python package for commonly used pytorch models, data loaders and utils.
Installation
For pip
pip install torch-template
For source
Clone the repo, cd into it and run pip install .
command.
git clone https://github.com/misads/torch_template.git
cd torch_template
pip install .
For conda
source ~/anaconda3/bin/activate
conda activate <env>
python setup.py install
A configure file torch_template.egg-info
will be generated in the repo directory. Copy torch_template
and torch_template.egg-info
to your site-packages
folder.
Usage
Test if the package is successfully installed:
import torch_template as tt
from torch_template import torch_utils
Run:
#!-bash
tt-new
Enter your repo name, then a python project template will be created.
Documentation
The documentation webpage can be found here https://torch-template.readthedocs.io/en/latest/.
File structure
File structure
.
├── network
│ ├── base_model.py :Define models, losses and parameter updating
│ ├── norm.py :Normalizations
│ └── weights_init.py :weights init
│
├── dataloader/ :Define Dataloaders
├── model_zoo :Commonly used models
└── utils
└── torch_utils.py :PyTorch utils
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
torch-template-0.0.4.tar.gz
(36.9 kB
view hashes)
Built Distribution
Close
Hashes for torch_template-0.0.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | c5542d4f993fc2bbeedf9d1e16c9ad4f79a77263721d957602a4d1b73ef22643 |
|
MD5 | f3137d1b1157fcde11d3af51683b62b7 |
|
BLAKE2b-256 | ad10596371b3bf07f35dc9347d248087c67d47a56d6e39196b0548e8ec57cc64 |