alicia 0.1.0

A CLI to download, train, test, predict and compare an image classifiers.

AlicIA

Install and usage

pip install alicia
alicia --help

If you just want to see a quick showcase of the tool, download and run showcase.sh https://github.com/aemonge/alicia/raw/main/docs/showcase.sh

Features

To see the full list of features, and option please refer to alicia –help

• Download common torchvision datasets.

• Select different transforms to train.

• Train, test and predict using different custom-made and torch-vision models (SqueezeNet, AlexNet, MNASNet)

• Get information about each model.

• Compare models training speed, accuracy, and meta information.

• Tested with MNIST, FashionMNIST, Flowers102.

• View results in the console, or with matplotlib.

References

Useful links found and used while developing this

• https://medium.com/analytics-vidhya/creating-a-custom-dataset-and-dataloader-in-pytorch-76f210a1df5d

• https://stackoverflow.com/questions/51911749/what-is-the-difference-between-torch-tensor-and-torch-tensor

• https://deepai.org/dataset/mnist

• https://medium.com/fenwicks/tutorial-1-mnist-the-hello-world-of-deep-learning-abd252c47709