Deploy your machine learning models with Flask in 5 minutes ! After you’ve finished all the R&D involved in the machine learning part, it’s nice to be able to deploy your model and make it available as a service. Flask is a python library that allows you to create http servers easily. You can use it to make your ML model available through HTTP requests. https://medium.com/@omaraflak/flask-api-for-keras-87c06da174e8
I recently published a post on Medium which goes through the math of machine learning all the way down to the code (in Python). If you understand the basic principles then you can implement any kind of layers you ever dreamt of (FullyConnected, Convolutional, MaxPooling, Dropout, etc). https://medium.com/@omaraflak/math-neural-network-from-scratch-in-python-d6da9f29ce65 You can also directly checkout the code on Github ! https://github.com/OmarAflak/Medium-Python-Neural-Network Cheers !