{"repo":"horrible-dong/QTClassification","free":true,"listed":false,"github":"https://github.com/horrible-dong/QTClassification","clone":"git clone https://github.com/horrible-dong/QTClassification.git","description":"A lightweight and extensible toolbox for image classification and MORE","language":"Python","stars":20,"topics":["pytorch","toolbox","imagenet","resnet","deep-learning","image-classfication","template","vision-transformer","configs","amp"],"license":"Apache-2.0","category":"machine-learning","readme_excerpt":"QTClassification ======== A lightweight and extensible toolbox for image classification and MORE     Author: QIU Tian Affiliation: Zhejiang University 🛠️ Installation 📘 Documentation 🌱 Dataset Zoo 👀 Model Zoo English 简体中文 Installation The development environment of this project is python 3.8 & pytorch 1.13.1+cu117 . 1. Create your conda environment. 2. Enter your conda environment. 3. Install PyTorch. Or you can refer to PyTorch to install newer or older versions. Please note that if pytorch ≥ 1.13, then python ≥ 3.7.2 is required. 4. Install necessary dependencies. Getting Started For a quick experience, you can directly run the following commands: Training The cifar10 dataset and resnet50 pretrained weights will be automatically downloaded. Please keep the network accessible. The cifar10 dataset will be downloaded to ./data . The resnet50 pretrained weights will be downloaded to /.cache/torch/hub/checkpoints . During training, the config file, checkpoints, logs, and other outputs will be stored in ./runs/ tmp . Evaluation Using a config file (Recommended) You can also write the arguments into a config file (.py) and use --config / -c to import it. See configs. Training Evaluation The config arguments override or merge with the command-line arguments args pre-defined in main.py . Other examples Command-line arguments after --config xxx / -c xxx override the config arguments if the name is duplicated. For more details and advanced usage of config files, please r","default_branch":null,"files":null,"tree":[],"storefront":"/r/horrible-dong","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/horrible-dong/QTClassification/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}