import PIL
import PIL.Image
import PIL.ImageFont
import PIL.ImageOps
import PIL.ImageDraw
import pandas as pd
import numpy as np
import os,cv2
import matplotlib.pyplot as plt
import keras
import sklearn
from sklearn import preprocessing
from sklearn.utils import shuffle
from sklearn.cross_validation import train_test_split
from keras import backend as K
K.set_image_dim_ordering('th')
from keras.utils import np_utils
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation, Flatten
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras.optimizers import SGD,RMSprop,adam
import scipy
from skimage import io
%matplotlib inline

#img_rows=64
#img_cols=64
num_channel=1
num_epoch=10

labels =[]

image=[]

for i in range (50000):
    img = cv2.imread("E:/data/pic/%s.png"%('{0:06}'.format(i)))
    img=cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    #img_resize=cv2.resize(img,(180,670))
    image.append(img)


img_data = np.array(image)
img_data = img_data.astype('float32')
img_data /= 255
print (img_data.shape)

if num_channel==1:
        if K.image_dim_ordering()=='th':
            img_data= np.expand_dims(img_data, axis=1)
            print (img_data.shape)
        else:
            img_data= np.expand_dims(img_data, axis=4)
            print (img_data.shape)

else:
        if K.image_dim_ordering()=='th':
            img_data=np.rollaxis(img_data,3,1)
            print (img_data.shape)

#%%


# Assigning Labels

# Define the number of classes
num_classes = 5

num_of_samples = img_data.shape[0]
labels = np.ones((num_of_samples,),dtype='int64')
for i in range (50000):
    with open('E:/data/font/%s.meta'%('{0:06}'.format(i)), 'r') as f:
        temp = f.readline()
        first_line = temp.strip()
        if(first_line=='Alegreya-Regular'):

            labels[i] = 0

        elif(first_line=='Alegreya-Bold'):

            labels[i] = 1

        elif(first_line=='basic_sans_serif'):

            labels[i] = 2

        elif(first_line=='bold_sans_serif'):

            labels[i] = 3

        elif(first_line=='ChargerMoSp'):

            labels[i] = 4


names = ['Alegreya-Regular','Alegreya-Bold','basic_sans_serif','bold_sans_serif','ChargerMoSp']

# convert class labels to on-hot encoding
Y = np_utils.to_categorical(labels, num_classes)
#Shuffle the dataset
x,y = shuffle(img_data,Y, random_state=2)
# Split the dataset
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=2)

model = Sequential()
model.add(Convolution2D(16, kernel_size=(5, 5), strides=(1, 1),activation='relu',input_shape=(1,180,670)))
model.add(Convolution2D(16, kernel_size=(5, 5), strides=(1, 1),activation='relu'))
keras.layers.normalization.BatchNormalization(axis=-1, momentum=0.9, epsilon=0.001, center=True,scale=False, beta_initializer='zeros',gamma_initializer='ones', moving_mean_initializer='zeros',moving_variance_initializer='ones',beta_regularizer=None, gamma_regularizer=None,beta_constraint=None, gamma_constraint=None)
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Convolution2D(16, (5, 5), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
keras.layers.normalization.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001, center=True,scale=False, beta_initializer='zeros',gamma_initializer='ones', moving_mean_initializer='zeros',moving_variance_initializer='ones',beta_regularizer=None, gamma_regularizer=None, beta_constraint=None, gamma_constraint=None)
model.add(Convolution2D(16, (3, 3), activation='relu'))
keras.layers.normalization.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001, center=True,scale=False, beta_initializer='zeros',gamma_initializer='ones', moving_mean_initializer='zeros',moving_variance_initializer='ones',beta_regularizer=None, gamma_regularizer=None, beta_constraint=None, gamma_constraint=None)
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(32, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))

#sgd = SGD(lr=learning_rate, decay=decay, momentum=0.9, nesterov=False)
model.compile(loss=keras.losses.categorical_crossentropy,optimizer=keras.optimizers.SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=False),metrics=['accuracy'])

# Viewing model_configuration

model.summary()
model.get_config()
model.layers[0].get_config()
model.layers[0].input_shape
model.layers[0].output_shape
model.layers[0].get_weights()
np.shape(model.layers[0].get_weights()[0])
model.layers[0].trainable

# Training
hist = model.fit(X_train, y_train, batch_size=16, nb_epoch=num_epoch, verbose=1, validation_data=(X_test, y_test))

#hist = model.fit(X_train, y_train, batch_size=32, nb_epoch=20,verbose=1, validation_split=0.2)
# visualizing losses and accuracy
train_loss=hist.history['loss']
val_loss=hist.history['val_loss']
train_acc=hist.history['acc']
val_acc=hist.history['val_acc']
xc=range(num_epoch)

score = model.evaluate(X_test, y_test, verbose=0)
print('Test Loss:', score[0])
print('Test accuracy:', score[1])

test_image = X_test[0:1]
print (test_image.shape)

print(model.predict(test_image))
print(model.predict_classes(test_image))
print(y_test[0:1])
# Testing a new image
test_image = cv2.imread('C:/Users/Parvin.E/Desktop/data2/pic/000085.png')
test_image=cv2.cvtColor(test_image, cv2.COLOR_BGR2GRAY)
#test_image=cv2.resize(test_image,(64,64))
test_image = np.array(test_image)
test_image = test_image.astype('float32')
test_image /= 255
print (test_image.shape)

if num_channel==1:
    if K.image_dim_ordering()=='th':
        test_image= np.expand_dims(test_image, axis=0)
        test_image= np.expand_dims(test_image, axis=0)
        print (test_image.shape)
    else:
        test_image= np.expand_dims(test_image, axis=3)
        test_image= np.expand_dims(test_image, axis=0)
        print (test_image.shape)

else:
    if K.image_dim_ordering()=='th':
        test_image=np.rollaxis(test_image,2,0)
        test_image= np.expand_dims(test_image, axis=0)
        print (test_image.shape)
    else:
        test_image= np.expand_dims(test_image, axis=0)
        print (test_image.shape)
# Predicting the test image
print((model.predict(test_image)))
print(model.predict_classes(test_image))