notebook.community

Edit and run 





In [0]:



    


import tensorflow as tf
from tensorflow import keras



    











In [0]:



    


mnist = keras.datasets.fashion_mnist



    











In [0]:



    


mnist.load_data()



    











In [0]:



    


(train_images,train_labels),(test_images,test_labels) = mnist.load_data()



    











In [5]:



    


train_images.shape



    


















    
Out[5]:








(60000, 28, 28)

















In [6]:



    


test_images.shape



    


















    
Out[6]:








(10000, 28, 28)

















In [0]:



    


import numpy as np
import matplotlib.pyplot as plt



    











In [0]:



    


class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat', 
               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']



    











In [11]:



    


plt.figure()
plt.imshow(train_images[0])
plt.colorbar()
plt.grid(False)



    


















    
























In [12]:



    


train_images



    


















    
Out[12]:








array([[[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]],

       [[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]],

       [[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]],

       ...,

       [[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]],

       [[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]],

       [[0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        ...,
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0],
        [0, 0, 0, ..., 0, 0, 0]]], dtype=uint8)

















In [14]:



    


train_images.data



    


















    
Out[14]:








<memory at 0x7f811b938138>

















In [0]:



    


train_images = train_images/255.0
test_images = test_images/255.0



    











In [16]:



    


plt.figure()
plt.imshow(train_images[0])
plt.colorbar()
plt.grid(False)



    


















    
























In [17]:



    


plt.figure(figsize=(10,10))
for i in range(25):
    plt.subplot(5,5,i+1)
    plt.xticks([])
    plt.yticks([])
    plt.grid(False)
    plt.imshow(train_images[i], cmap=plt.cm.binary)
    plt.xlabel(class_names[train_labels[i]])



    


















    
























In [0]:



    


model = keras.Sequential([
    
    keras.layers.Flatten(input_shape=(28,28)),
    keras.layers.Dense(128,activation=tf.nn.relu),
    keras.layers.Dense(10, activation=tf.nn.softmax)
])



    











In [0]:



    


model.compile(loss='sparse_categorical_crossentropy',
             optimizer=tf.train.AdamOptimizer(),
             metrics=['accuracy'])



    











In [22]:



    


model.fit(train_images, train_labels,  epochs=10)



    


















    






Epoch 1/10
60000/60000 [==============================] - 6s 108us/step - loss: 0.4961 - acc: 0.8265
Epoch 2/10
60000/60000 [==============================] - 5s 78us/step - loss: 0.3731 - acc: 0.8647
Epoch 3/10
60000/60000 [==============================] - 5s 78us/step - loss: 0.3372 - acc: 0.8771
Epoch 4/10
60000/60000 [==============================] - 5s 77us/step - loss: 0.3146 - acc: 0.8862
Epoch 5/10
60000/60000 [==============================] - 5s 76us/step - loss: 0.2949 - acc: 0.8909
Epoch 6/10
60000/60000 [==============================] - 5s 78us/step - loss: 0.2800 - acc: 0.8962
Epoch 7/10
60000/60000 [==============================] - 5s 77us/step - loss: 0.2674 - acc: 0.9008
Epoch 8/10
60000/60000 [==============================] - 5s 78us/step - loss: 0.2566 - acc: 0.9036
Epoch 9/10
60000/60000 [==============================] - 5s 77us/step - loss: 0.2464 - acc: 0.9074
Epoch 10/10
60000/60000 [==============================] - 5s 79us/step - loss: 0.2387 - acc: 0.9119










    
Out[22]:








<tensorflow.python.keras.callbacks.History at 0x7f8118a6ac50>

















In [23]:



    


test_loss, test_acc = model.evaluate(test_images, test_labels)



    


















    






10000/10000 [==============================] - 0s 35us/step

















In [24]:



    


print(test_acc)



    


















    






0.8852

















In [0]:



    


pred = model.predict(test_images)



    











In [26]:



    


pred[0]



    


















    
Out[26]:








array([3.9710821e-07, 1.5823820e-10, 1.9410448e-08, 4.0665621e-10,
       1.2048154e-08, 1.2978223e-03, 9.2292964e-08, 5.7965014e-03,
       4.7191915e-07, 9.9290472e-01], dtype=float32)

















In [28]:



    


np.argmax(pred[2])



    


















    
Out[28]:








1

















In [29]:



    


test_labels[2]



    


















    
Out[29]:








1

















In [0]:



    


def plot_image(i, predictions_array, true_label, img):
  predictions_array, true_label, img = predictions_array[i], true_label[i], img[i]
  plt.grid(False)
  plt.xticks([])
  plt.yticks([])
  
  plt.imshow(img, cmap=plt.cm.binary)

  predicted_label = np.argmax(predictions_array)
  if predicted_label == true_label:
    color = 'blue'
  else:
    color = 'red'
  
  plt.xlabel("{} {:2.0f}% ({})".format(class_names[predicted_label],
                                100*np.max(predictions_array),
                                class_names[true_label]),
                                color=color)

def plot_value_array(i, predictions_array, true_label):
  predictions_array, true_label = predictions_array[i], true_label[i]
  plt.grid(False)
  plt.xticks([])
  plt.yticks([])
  thisplot = plt.bar(range(10), predictions_array, color="#777777")
  plt.ylim([0, 1]) 
  predicted_label = np.argmax(predictions_array)
 
  thisplot[predicted_label].set_color('red')
  thisplot[true_label].set_color('blue')



    











In [32]:



    


i = 2
plt.figure(figsize=(6,3))
plt.subplot(1,2,1)
plot_image(i, pred, test_labels, test_images)
plt.subplot(1,2,2)
plot_value_array(i, pred,  test_labels)



    


















    
























In [34]:



    


num_rows = 5
num_cols = 3
num_images = num_rows*num_cols
plt.figure(figsize=(2*2*num_cols, 2*num_rows))
for i in range(num_images):
  plt.subplot(num_rows, 2*num_cols, 2*i+1)
  plot_image(i, pred, test_labels, test_images)
  plt.subplot(num_rows, 2*num_cols, 2*i+2)
  plot_value_array(i, pred, test_labels)



    


















    
























In [0]:

Content source: Pritam-N/Linx

Similar notebooks:

notebook.community | gallery | about