MLOPS Project

This Project is mainly build in order to join two industry in our world in this time, Yes i am talking about the Medical Industry and Technical Industry. In this project i've made a Deep Learning Model with the help of arounf 6000 - 7000 images for training downloaded from various data sources on the internet.

My Deep Learning Model is predicing at the rate of accuracy of 95 % for detecting Pneumonia Patients using X-Ray Images. I've also uploaded the model after training in this Repository as a File Named "pneumonia_model.h5". You don't need to train this model again, just load this model in your application and start using in the Medical Services.

Some tools of DevOps industry are also used in this project for performing following task.

This Repository contains the script of Automating te job of Hyper Parameter Tuning in Deep Learning. Basically in this project I've integrated several tools for Automating this Job, i've used tools like Docker, Jenkins, Deep Learning, Git, Regular Expressions.

Jenkins is the best tool for performing this task because it is working as an intermediater between two environments and it will automatically tune the hyper parameters in the Code like Jenkins will automatically add CRP layers in our CNN code as well as it can change the values of parameters in the Layers like Number of Filters, kernel_size, Number of Epochs in the file.

This project is another example that how our Technologies available in the market can help different other industries to build.