Dataset class.

data_feeders/_tut_sed_synthetic_2016.py

@@ -0,0 +1,67 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from pathlib import Path
+from typing import Tuple
+
+from numpy import ndarray
+from torch.utils.data import Dataset
+
+from tools import file_io
+
+__author__ = 'Konstantinos Drossos -- Tampere University'
+__docformat__ = 'reStructuredText'
+__all__ = ['TUTSEDSynthetic2016']
+
+
+class TUTSEDSynthetic2016(Dataset):
+ """TUT SED Synthetic 2016 dataset
+ """
+ def __init__(self,
+ root_dir: str,
+ split: str,
+ input_features_file_name: str,
+ target_values_input_name: str) \
+ -> None:
+ """TUT SED Synthetic 2016 dataset class. 
+ 
+ :param root_dir: The root directory for the dataset. 
+ :type root_dir: str
+ :param split: The split for the dataset (e.g. training).
+ :type split: str
+ :param input_features_file_name: Input features file name.
+ :type input_features_file_name: str
+ :param target_values_input_name: Target values file name.
+ :type target_values_input_name: str
+ """
+ super(TUTSEDSynthetic2016, self).__init__()
+ data_path = Path(root_dir, 'synthetic', split)
+
+ x_path = data_path.joinpath(input_features_file_name)
+ y_path = data_path.joinpath(target_values_input_name)
+
+ self.x: ndarray = file_io.load_numpy_object(x_path)
+ self.y: ndarray = file_io.load_numpy_object(y_path)
+
+ def __len__(self) \
+ -> int:
+ """Length method.
+
+ :return: Amount of examples.
+ :rtype: int
+ """
+ return self.x.shape[0]
+
+ def __getitem__(self,
+ item: int) \
+ -> Tuple[ndarray, ndarray]:
+ """Gets an example from the dataset.
+
+ :param item: Index of the example.
+ :type item: int
+ :return: Example from the dataset, input and target values.
+ :rtype: (numpy.ndarray, numpy.ndarray)
+ """
+ return self.x[item], self.y[item]
+
+# EOF