afozbek

document.querySelectorAll("a").forEach(link=>link.setAttribute("target", "_blank"))

afozbek

# Datasetimizi parçalayarak test datası oluşturacağız.
# Datamızın 1/3 ünü test datası olarak oluşturuyoruz 
# Yani 30 satırımızın 20 si train datası olarak ayrılacak
# Ve 10 satırımız da test datası olucak.

X_train, X_test, y_train, y_test = train_test_split(x, y, test_size = 1/3, random_state = 123, shuffle=1)

afozbek

history = model.fit(X_train, y_train, validation_split=0.1, epochs=10, batch_size=200, 
                    verbose=1, shuffle=1)

afozbek

import requests
from PIL import Image

url = 'https://www.researchgate.net/profile/Jose_Sempere/publication/221258631/figure/fig1/AS:305526891139075@1449854695342/Handwritten-digit-2.png'
response = requests.get(url, stream = True)
img = Image.open(response.raw)

plt.imshow(img)

afozbek

assert(X_train.shape[0] == y_train.shape[0]), "The number of images is not equal .."
assert(X_test.shape[0] == y_test.shape[0]), "The number of images is not equal .."
assert(X_train.shape[1:] == (28, 28)), "The dimension of the images are not 28x28"
assert(X_test.shape[1:] == (28, 28)), "The dimension of the images are not 28x28"

afozbek

num_of_samples = []

cols = 5 #We will select 5 random images
num_of_classes = 10 #each digit total: 10

fig, axs = plt.subplots(nrows=num_of_classes, ncols=cols,
                       figsize=(5, 10))
fig.tight_layout()
for i in range(cols):
  for j in range(num_of_classes):

afozbek

def create_model():
  model = Sequential()
  model.add(Dense(10, input_dim = num_pixels, #num_pixels: 784
                  activation = 'relu'))
  model.add(Dense(30, activation='relu'))
  model.add(Dense(10, activation='relu'))
  model.add(Dense(num_of_classes, activation='softmax'))
  model.compile(Adam(lr=0.01), #lr: learning rate
                loss='categorical_crossentropy', #loss function
               metrics=['accuracy'])

afozbek

def create_model():
  model = Sequential()
  model.add(Dense(10, input_dim = num_pixels, #num_pixels: 784
                  activation = 'relu'))
  model.add(Dense(30, activation='relu'))
  model.add(Dense(10, activation='relu'))
  model.add(Dense(num_of_classes, activation='softmax'))
  model.compile(Adam(lr=0.01), #lr: learning rate
                loss='categorical_crossentropy', #loss function
               metrics=['accuracy'])