PyTorch Dataset Splitting: Train, Validation, and Test Sets

This code snippet is splitting the dataset into train, validation, and test sets. - The variable y represents the labels/targets of the dataset. - The train_test_split function from the torch library is used to split the indices of the dataset into train and test sets. - The train_size, val_size, and test_size parameters determine the proportions of the dataset allocated for each set. - The random_state parameter ensures that the split is reproducible. - If a validation set size is specified (val_size > 0), the train set is further split into train and validation sets using the same train_test_split function. - The stratify parameter ensures that the class distribution is maintained in the splits. - Finally, the train, validation, and test set indices are assigned to the train_nodes, val_nodes, and test_nodes variables, respectively.