The "Hello World" of Deep Learning : training a deep learning model to correctly classify hand-written digits

Libraries: Keras + TensorFlow

Steps:

1. Loading the Data Into Memory (with Keras)
2. Splitting it into training and validation data and thus labeling it accordingly
3. Preparing the data. (Data Pre-Processing)
   • Flatten- Convert 2d array to 1d
   • Normalize- The data is in the range of 0 to 255, Converting it into a range varying from 0 to 1
   • Categorize- makes data more understandable for the machine
4. Creating the model
   • Instantiating the model
   • Creating the input layer
   • Creating the hidden layer
   • Creating the output layer
   • Summarizing the model
   • Compiling the model
5. Training the model
6. Observing Accuracy