Handwritten digit recognition (Using Scikit-Learn)

Step 001: Loading the Dataset

importing the dataset

from sklearn.datasets import load_digits

digits = load_digits() following command to know the shape of the Dataset:

print(“Image Data Shape” , digits.data.shape)

There are 1797 images in the dataset

Step 002: Visualizing the images and labels in our Dataset

Here we are visualizing the first 5 images in the Dataset

import numpy as np

import matplotlib.pyplot as plt

plt.figure(figsize=(20,4))

for index, (image, label) in enumerate(zip(digits.data[0:5], digits.target[0:5])):

plt.subplot(1, 5, index + 1)

plt.imshow(np.reshape(image, (8,8)), cmap=plt.cm.gray)

plt.title(‘Training: %i\n’ % label, fontsize = 20)

Step 003: Splitting our Dataset into training and testing sets

from sklearn.model_selection import train_test_split

x_train, x_test, y_train, y_test = train_test_split(digits.data, digits.target, test_size=0.05, random_state=95)

Step 004: The Scikit-Learn 4-Step Modeling Pattern

Step 01. Importing the model we want to use.

from sklearn.linear_model import LogisticRegression

Step 02: Making an instance of the Model

logisticRegr = LogisticRegression()

Step 03: Training the Model

logisticRegr.fit(x_train, y_train)

Step 04. Predicting the labels of new data

predictions = logisticRegr.predict(x_test)

Step 05: Measuring the performance of our Model

Use accuracy_score

score = logisticRegr.score(x_test, y_test)

print(score)

Step 06: Confusion matrix

Using Seaborn for our confusion matrix.

import matplotlib.pyplot as plt

import seaborn as sns

from sklearn import metrics

cm = metrics.confusion_matrix(y_test, predictions)

plt.figure(figsize=(9,9))

sns.heatmap(cm, annot=True, fmt=”.3f”, linewidths=.5, square = True, cmap = ‘Pastel1’);

plt.ylabel(‘Actual label’);

plt.xlabel(‘Predicted label’);

all_sample_title = ‘Accuracy Score: {0}’.format(score)

plt.title(all_sample_title, size = 15);