MNIST Handwritten Digits Recognition - Image_Classification.ipynb

Kernel: Python 3

Dataset Information

This dataset allows you to study, analyze and recognize elements in the images. That’s exactly how your camera detects your face, using image recognition! It’s a digit recognition problem. This data set has 49,000 images of 28 X 28 size, totalling 49 MB.

Import Modules

# !pip install tensorflow-gpu keras

import pandas as pd
import numpy as np
from tqdm.notebook import tqdm
from keras.preprocessing.image import img_to_array, load_img
import tensorflow as tf
import matplotlib.pyplot as plt
%matplotlib inline
import warnings

warnings.filterwarnings('ignore')

Unzip the train data|

# !unzip Train_UQcUa52.zip

Load the data

df = pd.read_csv('train.csv')
df.head()

!pwd

/content

image_path = 'Images/train/'

X = np.array([img_to_array(load_img(image_path+df['filename'][i], target_size=(28,28,1), grayscale=True))
              for i in tqdm(range(df.shape[0]))
              ]).astype('float32')

HBox(children=(FloatProgress(value=0.0, max=49000.0), HTML(value='')))

y = df['label']

print(X.shape, y.shape)

(49000, 28, 28, 1) (49000,)

Exploratory Data Analysis

image_index = 0
print(y[image_index])
plt.imshow(X[image_index].reshape(28,28), cmap='Greys')

4

<matplotlib.image.AxesImage at 0x7f813cbf4e48>

image_index = 10
print(y[image_index])
plt.imshow(X[image_index].reshape(28,28), cmap='Greys')

2

<matplotlib.image.AxesImage at 0x7f813cb8e668>

image_index = 100
print(y[image_index])
plt.imshow(X[image_index].reshape(28,28), cmap='Greys')

7

<matplotlib.image.AxesImage at 0x7f813c629c50>

Train-Test Split

from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42, stratify=np.array(y))

Normalization

# x_train[0]

x_train /= 255
x_test /= 255

# x_train[0]

Model Creation

input_shape = (28,28,1)
output_class = 10

from keras.models import Sequential
from keras.layers import Dense, Conv2D, Dropout, Flatten, MaxPooling2D

# define the model
model = Sequential()
model.add(Conv2D(28, kernel_size=(3,3), input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Flatten())
model.add(Dense(128, activation=tf.nn.relu))
model.add(Dropout(0.3))
model.add(Dense(output_class, activation=tf.nn.softmax))

model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics='accuracy')

# train the model
model.fit(x=x_train, y=y_train, batch_size=32, epochs=30, validation_data=(x_test, y_test))

Epoch 1/30
1149/1149 [==============================] - 10s 3ms/step - loss: 0.4816 - accuracy: 0.8475 - val_loss: 0.1202 - val_accuracy: 0.9637
Epoch 2/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.1336 - accuracy: 0.9605 - val_loss: 0.0848 - val_accuracy: 0.9743
Epoch 3/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0863 - accuracy: 0.9732 - val_loss: 0.0807 - val_accuracy: 0.9742
Epoch 4/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0685 - accuracy: 0.9783 - val_loss: 0.0734 - val_accuracy: 0.9788
Epoch 5/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0543 - accuracy: 0.9825 - val_loss: 0.0690 - val_accuracy: 0.9809
Epoch 6/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0461 - accuracy: 0.9844 - val_loss: 0.0684 - val_accuracy: 0.9808
Epoch 7/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0360 - accuracy: 0.9873 - val_loss: 0.0743 - val_accuracy: 0.9798
Epoch 8/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0318 - accuracy: 0.9884 - val_loss: 0.0733 - val_accuracy: 0.9811
Epoch 9/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0319 - accuracy: 0.9891 - val_loss: 0.0658 - val_accuracy: 0.9838
Epoch 10/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0242 - accuracy: 0.9919 - val_loss: 0.0728 - val_accuracy: 0.9827
Epoch 11/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0218 - accuracy: 0.9926 - val_loss: 0.0815 - val_accuracy: 0.9818
Epoch 12/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0286 - accuracy: 0.9895 - val_loss: 0.0766 - val_accuracy: 0.9829
Epoch 13/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0199 - accuracy: 0.9928 - val_loss: 0.0762 - val_accuracy: 0.9820
Epoch 14/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0239 - accuracy: 0.9918 - val_loss: 0.0754 - val_accuracy: 0.9836
Epoch 15/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0160 - accuracy: 0.9938 - val_loss: 0.0865 - val_accuracy: 0.9820
Epoch 16/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0196 - accuracy: 0.9935 - val_loss: 0.0842 - val_accuracy: 0.9822
Epoch 17/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0152 - accuracy: 0.9951 - val_loss: 0.0825 - val_accuracy: 0.9828
Epoch 18/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0155 - accuracy: 0.9943 - val_loss: 0.0889 - val_accuracy: 0.9817
Epoch 19/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0207 - accuracy: 0.9930 - val_loss: 0.0886 - val_accuracy: 0.9822
Epoch 20/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0122 - accuracy: 0.9955 - val_loss: 0.0958 - val_accuracy: 0.9822
Epoch 21/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0135 - accuracy: 0.9957 - val_loss: 0.0986 - val_accuracy: 0.9824
Epoch 22/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0166 - accuracy: 0.9949 - val_loss: 0.0987 - val_accuracy: 0.9824
Epoch 23/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0153 - accuracy: 0.9949 - val_loss: 0.0917 - val_accuracy: 0.9832
Epoch 24/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0147 - accuracy: 0.9950 - val_loss: 0.0967 - val_accuracy: 0.9838
Epoch 25/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0112 - accuracy: 0.9957 - val_loss: 0.1057 - val_accuracy: 0.9816
Epoch 26/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0134 - accuracy: 0.9959 - val_loss: 0.1024 - val_accuracy: 0.9830
Epoch 27/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0085 - accuracy: 0.9968 - val_loss: 0.1256 - val_accuracy: 0.9795
Epoch 28/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0127 - accuracy: 0.9958 - val_loss: 0.1099 - val_accuracy: 0.9832
Epoch 29/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0136 - accuracy: 0.9952 - val_loss: 0.1043 - val_accuracy: 0.9824
Epoch 30/30
1149/1149 [==============================] - 4s 3ms/step - loss: 0.0132 - accuracy: 0.9959 - val_loss: 0.1162 - val_accuracy: 0.9827

<tensorflow.python.keras.callbacks.History at 0x7f80dc057278>

Testing the model

image_index = 10
# print("Original output:",y_test[image_index])
plt.imshow(x_test[image_index].reshape(28,28), cmap='Greys')
pred = model.predict(x_test[image_index].reshape(1,28,28,1))
print("Predicted output:", pred.argmax())

Predicted output: 1

image_index = 100
# print("Original output:",y_test[image_index])
plt.imshow(x_test[image_index].reshape(28,28), cmap='Greys')
pred = model.predict(x_test[image_index].reshape(1,28,28,1))
print("Predicted output:", pred.argmax())

Predicted output: 8