OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Pytorch CIFAR100

Pytorch CIFAR100 is a Python/PyTorch repository with scripts for training and testing image-classification networks on CIFAR-100. It supports a range of CNN architectures, uses torchvision for the dataset in its documented workflow, and includes an example module for custom datasets.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Pytorch CIFAR100 — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a code-based CIFAR-100 experiment with a choice of supported CNN architectures and separate training and testing scripts.

Who it is for

  • People learning PyTorch image classification through CIFAR-100 experiments.
  • Practitioners comparing supported CNN architectures under a shared training workflow.
  • Developers seeking an example of a custom PyTorch dataset module.

Documented uses

  • Train a selected supported network on CIFAR-100.
  • Test a network using a saved weights file.
  • Monitor training with optional TensorBoard.
  • Adapt the example dataset module for custom dataset handling.
Before you choose

Requirements & limits

  • The README lists Python 3.6 and PyTorch 1.6.0+cu101 as its experiment environment.
  • TensorBoard 2.2.2 is optional.
  • The README lists many network arguments, including ResNet, VGG, DenseNet, MobileNet, ShuffleNet and Inception variants.
  • The README cautions that shared hyperparameters may not yield the best result for every network.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Train and test a CIFAR-100 model

Run the documented VGG16 training example and use the resulting weights with the test script.

Before you start

  • Python 3.6 and PyTorch 1.6.0+cu101, as listed in the README.
  • The repository and CIFAR-100 dataset supplied through torchvision.

Steps

  1. Enter the repository directory.

  2. Run the documented training command: python train.py -net vgg16 -gpu.

  3. Use the resulting weights file with test.py, specifying -net vgg16 and the weights-file path.

Expected outcome

The repository's training and testing scripts are configured to run a VGG16 CIFAR-100 experiment; the README says the best-accuracy weights are normally saved in the checkpoint folder with a “best” suffix.

Guide limitations

  • The README describes a particular experiment environment; it does not establish compatibility with other software or hardware configurations.
  • The GPU example requires a suitable GPU setup; the README does not provide a CPU-only example.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
weiaicunzai/pytorch-cifar100 · Official homepage title · retained in its source language
How we review sourcesSuggest a correction