Objective here is to explore the concepts of deep learning and implementing it from scratch in a structured way. We will test the network with our default handwritten dataset.
About : In this part of this exercise, we will create a Layer and Sequential API for building neural network at ease. We will also implement SGD with momentum and popular activation functions
Link of Work: Code
About : In this exercise, we will try to explore different optimizers, extend those and implement it as part of the Layered API and we also test those with our default dataset
Link of Work: Code
About : In this session, we will discuss dropout in detail and compare it with ensemble or bagging techniques. Extend Dropout as part of Layered API.
Link of Work: Code
About : In this session, we will extend our layer API to support batch normalization.
Link of Work: Code
About : In this session, we will create a wrapper which accepts the input as a csv and return back numercalized data which would be used for building a language model in upcoming sessions.
Link of Work: Code