This repository contains a Python code that demonstrates digit classification using Support Vector Machines (SVM). The code utilizes the Digits dataset from the scikit-learn library and includes functionality for training an SVM classifier, evaluating its performance, and making predictions on test images.
The Digits dataset consists of a collection of 8x8 images of handwritten digits (0-9). Each image is represented as a matrix of grayscale pixel values. The dataset is included in the scikit-learn library and can be loaded using the datasets.load_digits()
function.
This will execute the SVM digit classification script, which performs the following steps:
- Loads the Digits dataset.
- Flattens the 8x8 images into a one-dimensional array.
- Splits the dataset into training and testing sets (75% training, 25% testing).
- Trains an SVM classifier using the training data.
- Predicts the labels for the testing data.
- Generates a confusion matrix to evaluate the classifier's performance.
- Displays the confusion matrix plot.
- Prints the accuracy score of the classifier.
- Displays the prediction for a test image.
-
The confusion matrix plot provides a visual representation of the classifier's performance, showing the number of correctly and incorrectly classified instances for each digit.
-
The accuracy score indicates the percentage of correctly classified instances in the testing set.
Here is an example of the output generated by the code: