This is an ongoing project for Skin Disease Analysis using machine learning.
Currently the repository is in training phase we are using both labeled and unlabelled data.
-
'part-1-pre-process': This script handles dataset loading, conducts necessary image preprocessing, and divides the data into training, validation, and test sets
-
'part-2-data-visuals': This code is designed to display the distribution of different types of skin lesions across the training, validation, and test datasets.
-
'part-3-extra-images': Creating more images for classes that have too few in our dataset.
-
'model.py': The code is used to build our Xception model for analysis.
-
'evaluate.py': script to evaluate our model for fine-tuning and deeper insights. It includes the confusion matrix, accuracy and loss histograms, and a classification report.
-
'predict.py': Code for prediction a batch of images from a directory, using our model.