split_data

src/main.py

@@ -4,3 +4,13 @@ from utils import *
 if __name__ == "__main__":
     # Load CIFAR data
     data, labels = read_cifar(r"../data/cifar-10-batches-py")
+    
+    print(f"Data shape: {data.shape}, Labels shape: {labels.shape}\n")
+    
+    # Split the dataset
+    data_train, labels_train, data_test, labels_test = split_dataset(data, labels, 0.8)
+    
+    print(f"Split the dataset with a {0.8} split factor:")
+    print(f" - Training data shape: {data_train.shape}, Training labels shape: {labels_train.shape}")
+    print(f" - Testing data shape: {data_test.shape}, Testing labels shape: {labels_test.shape}")
+    
\ No newline at end of file

src/utils/__init__.py

@@ -3,6 +3,7 @@
 ROOT = r"../../.."
 
 from .read_cifar import *
+from .split_data import *
 
 
 

src/utils/split_data.py

+# Author: Pierre CAU 
+# Date: 2024
+
+import numpy as np
+
+def split_dataset(data, labels, split):
+    """
+    Split the dataset into a training set and a test set.
+    
+    Parameters
+    ----------
+    data : np.ndarray
+        Array of data samples.
+    labels : np.ndarray
+        Array of labels corresponding to the data samples.
+    split : float
+        A float between 0 and 1 which determines the split factor of the training set with respect to the test set.
+    
+    Returns
+    -------
+    data_train : np.ndarray
+        Training data.
+    labels_train : np.ndarray
+        Corresponding labels for the training data.
+    data_test : np.ndarray
+        Testing data.
+    labels_test : np.ndarray
+        Corresponding labels for the testing data.
+    """
+    assert data.shape[0] == labels.shape[0], "Data and labels must have the same number of samples"
+    assert 0 < split < 1, "Split must be a float between 0 and 1"
+    
+    # Shuffle the data and labels in unison
+    indices = np.arange(data.shape[0])
+    np.random.shuffle(indices)
+    data = data[indices]
+    labels = labels[indices]
+    
+    # Calculate the split index
+    split_index = int(data.shape[0] * split)
+    
+    # Split the data and labels
+    data_train = data[:split_index]
+    labels_train = labels[:split_index]
+    data_test = data[split_index:]
+    labels_test = labels[split_index:]
+    
+    return data_train, labels_train, data_test, labels_test
+
+# Example usage
+if __name__ == "__main__":
+    # Example data and labels
+    data = np.random.rand(100, 10)
+    labels = np.random.randint(0, 2, 100)
+
+    print(f"Data shape: {data.shape}, Labels shape: {labels.shape}")
+    
+    # Split the dataset
+    data_train, labels_train, data_test, labels_test = split_dataset(data, labels, 0.8)
+    
+    print(f"Training data shape: {data_train.shape}, Training labels shape: {labels_train.shape}")
+    print(f"Testing data shape: {data_test.shape}, Testing labels shape: {labels_test.shape}")
\ No newline at end of file