

In [1]:

    
import keras
from keras import layers, models
import matplotlib.pyplot as plt

(x_train, y_train), (x_test, y_test) = keras.datasets.cifar10.load_data()
x_mean, x_std = x_train.mean(), x_train.std()
x_train = (x_train - x_mean) / x_std
x_test = (x_test - x_mean) / x_std
y_train = keras.utils.to_categorical(y_train)
y_test = keras.utils.to_categorical(y_test)
x_train.shape
x_train = x_train.reshape(-1, 32*32*3)
x_test = x_test.reshape(-1, 32*32*3)









    



Using TensorFlow backend.

Here starts the questions :

What are the classes ?
In other words, in y_train (or y_test) what does 0, 1, 2.... 9 corresponds to ?



In [2]:

    
# Answer

Visualize the data

Using plt.imshow() visualize few images (2 to 5)



In [3]:

    
# Answer

Well.. Train the model

Do what ever you can to train a classifier that have score above 53% accuracy (in the test set) (Always explain what and why you're doing this or that)
Non-justified answer is like no answer... If you've been through an iterative process, descibe the steps that led you to this answer.



In [4]:

    
# Answer

Show some quantitative results

Is your model efficient ? What is your classification error, your recall and precision. Describe in a sentense what do these results mean given you dataset.



In [5]:

    
# Answer

Delivery

Make sure that I can run the notebook with "restart & run all" before delivering