Omniglot-CNN-Demo

In this repository, we implement several deep neural networks for the Omniglot challenge.

Requirements and Structures

The required environments are as follows.

• torch-1.11.0
• torchinfo-1.7.1
• scikit-learn-1.2.0
• Pillow-8.4.0

The structure of our project is as follows.

• codes: contain all the codes.
  • main.py: the entrance of our project.
  • train.py: define the class to train the model.
  • model.py: define all the models.
  • utils.py: load the data for training and evaluations.
  • config.yaml: store the configurations for model training.
• omniglot_resized: contain the data.
• output: the path to save models and evaluation results.
• report.pdf: a brief report of details of our implementations, experimental results and analysis.

Usage

To run our codes, change to the codes directory and use the following command

python main.py --config = config.yaml