Demo Tutorial to Build a Convolutional Neural Network with keras+google drive and google colaboratory (exploring Google's colab tool - https://colab.research.google.com/notebooks/welcome.ipynb)
This guide is meant to be a simple tutorial which lets you take images classes, train a model with keras+tensorflow and then predict a new image with your "intelligent" model.
Here is further explanation in a short medium post: https://medium.com/@chibuk/machine-learning-for-the-everyday-person-part-3-your-first-neural-network-744528595c65
You can use this python script to get sample training images from a website: https://github.com/ibbad/image-scrapper
To have this notebook work you will need to:
Have a google drive folder; copy this notebook to it
- Put the notebook and training images in a separate folder e.g. drive/Simple CNN Tutorial/
- Images: https://drive.google.com/drive/folders/1RcWFv-PDtGK4n1JgkAgL77fB-N0I4oMP?usp=sharing
- To run this notebook select each cell and click shift+enter on your keyboard, wait until the cell is finished running before going to the next
Below outline the standard import and installation settings for openCV and tensorflow