Tf-CoronaXray
Model summary:
3 Convolution layers accompanied by max pooling layers with flatten to compress the parameters into a column vector containing all the contents of the larger vectors. Sigmoid used for the final binary classification with ReLU as hidden layer activation function. Mini-batches of 10 images were trained and validated every epoch (Total 15 epochs). The dataset was obtained from Kaggle (dated Feb. 2020).
Model Test and validation data accuracy and loss visualised:
