# This Python 3 environment comes with many helpful analytics libraries installed
# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python
# For example, here's several helpful packages to load in 

import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)

# Input data files are available in the "../input/" directory.
# For example, running this (by clicking run or pressing Shift+Enter) will list the files in the input directory

import os
print(os.listdir("../input"))

# Any results you write to the current directory are saved as output.


from __future__ import print_function
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten ,BatchNormalization
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K

batch_size = 128
num_classes = 10
epochs = 12

# Load the data
train = pd.read_csv("../input/digits-dataset/train.csv")
test = pd.read_csv("../input/digits-dataset/test.csv")

Y_train = train["label"]
X_train = train.drop(labels = ["label"],axis = 1) 

del train 

import matplotlib.pyplot as plt
%matplotlib inline

import seaborn as sns
g = sns.countplot(Y_train)




# Reshape image in 3 dimensions (height = 28px, width = 28px , canal = 1)
X_train = X_train.values.reshape(-1,28,28,1)
test = test.values.reshape(-1,28,28,1)

# input image dimensions

img_rows, img_cols = 28, 28

if K.image_data_format() == 'channels_first':

    X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)

    test = test.reshape(test.shape[0], 1, img_rows, img_cols)

    input_shape = (1, img_rows, img_cols)

else:

    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)

    test = test.reshape(test.shape[0], img_rows, img_cols, 1)

    input_shape = (img_rows, img_cols, 1)
    
    
    X_train = X_train.astype('float32')

test = test.astype('float32')

X_train /= 255

test /= 255

print('x_train shape:', X_train.shape)

print(X_train.shape[0], 'train samples')

print(test.shape[0], 'test samples')

# convert class vectors to binary class matrices

Y_train = keras.utils.to_categorical(Y_train, num_classes)

#x_test = keras.utils.to_categorical(y_test, num_classes)



model = Sequential()

model.add(Conv2D(32, kernel_size=(3, 3),

                 activation='relu',

                 input_shape=input_shape))

model.add(BatchNormalization())

model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(BatchNormalization())

model.add(MaxPooling2D(pool_size=(2, 2)))

model.add(Dropout(0.25))

model.add(Flatten())

model.add(Dense(128, activation='relu'))

model.add(Dropout(0.5))

model.add(Dense(num_classes, activation='softmax'))



model.compile(loss=keras.losses.categorical_crossentropy,

              optimizer=keras.optimizers.Adadelta(),

              metrics=['accuracy'])
              
              
              
              
history=model.fit(X_train, Y_train,
          batch_size=250,
          epochs=12,
          verbose=1,
         )


prediction=model.predict(test)
results=prediction
results = np.argmax(results,axis = 1)


results = pd.Series(results,name="Label")




submission = pd.concat([pd.Series(range(1,28001),name = "ImageId"),results],axis = 1)

submission.to_csv("cnn_mnist_datagen.csv",index=False)