Computer Vision

CIFAR-10 Image classification

This is an implementation from the book Deep Learning for Vision System¶

Here I've implemented a robust CNN with BatchNormalization, regularization and many more concepts.

The exact keras implementation is here. Step-by-step overview

1. Import dependencies
2. get the data ready for training

   1. Download the data from the Keras library
   2. Split the data into train, validate and test datasets
   3. Normalize the data
   4. One-hot encoding the labels

3. Build the model architecture

   1. Deeper neural network to increase the learningccapacity
   2. dropout layers
   3. L2 regularization to our CONV layers
   4. Batch normalization layers

4. Train the model
5. Evaluate the model
6. Plot the learning curve