/module_files_cv

Primary LanguagePython

The repo contain generic code files for computer vision problems.

File Structure

  • main.py
    • get_data_loader() -> Returns train test data loader for the given dataset and batch size.
    • get_model() -> Returns requested model from model directory
    • set_model_config() -> It sets model config like optimizer, scheduler, type of device, learning rate.
    • train() -> Contains training logic of the model
    • test() -> Contains testing logic of the model
    • training_loop() -> Contains logic of training and testing the model for the given number of epochs.
  • utils
    • Albumentationtransform.py -> Contains logic for albumentation transformation.
    • generic_func.py -> Contains functions for downloading train and test set and plotting curves after model training.
    • gradcam.py -> Contains logic code for gradcam
    • utils_grad_cam.py -> Contains utils functions for gradcam
    • missclassified_images.py -> Contains function to plot miss classified images by inference on test loader.
  • models
    • resnet.py : Contains model code for resnet18 and resnet34