Trying out various machine learning and deep learning on X-ray dataset
The dataset consists of images of X-rays from different sources divided into 2 different sets viz. Covid and Non-Covid(Normal) Here we used training set of around 186 images and testing set of 92 images.
The distribution of the images where somewhat equal in both the classes.
- Scikit-Learn
- Tensorflow/Keras
- Matplotlib
- Jupyter Notebook
- Logistic Regression
- Gaussian Naive-bayes
- Support Vector Classifier
- Vanilla CNN(Basic CNN)
- CNN with Batch Normalization and Dropout
The dataset was loaded and all the above algorithms were tested on the data to find out which method is better.
The results are shown in the notebook along with supporting metrics like Training and Testing accuracy, F1 score, recall, confusion matrix, etc.
The jupyter notebook can be downloaded along with the dataset to test the performance on your machine.
Note: GPU support will be required for better performance