How to Create a Deep Learning Model and Deploy it in Web App: A demo with Keras and Flask for Malaria Detection
As medical diagnosis is a major application area of Machine Learning. Here we created a Machine Learning Model by applying a Convolution Neural Network thus obtaining an accuracy of 97.1% and minimal loss. Then we integrated the trained model in a Flask-based web app.
In order to make our model better, we have applied the concept of EarlyStopping, ReduceLROnPlateau, ModelCheckpoint. Talking about integrating the trained model with the web app we embedded the .h5 file with the flask backend and returned the obtained prediction and prediction percentage to the HTML template.
Code for training the model can be obtained from this link
The complete code of the web application for malaria detection along with the already trained model is available on my GitHub (Link Here)
