scikit-network 0.30.0__cp38-cp38-win_amd64.whl → 0.32.1__cp38-cp38-win_amd64.whl

sknetwork/data/load.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-Created on November 15, 2019
+Created in November 2019
 @author: Quentin Lutz <qlutz@enst.fr>
 """
 
@@ -19,12 +19,15 @@ import numpy as np
 from scipy import sparse
 
 from sknetwork.data.parse import from_csv, load_labels, load_header, load_metadata
-from sknetwork.utils import Bunch
+from sknetwork.data.base import Bunch
 from sknetwork.utils.check import is_square
-from sknetwork.utils.verbose import Log
+from sknetwork.log import Log
 
 NETSET_URL = 'https://netset.telecom-paris.fr'
 
+# former name of Dataset
+Bunch = Bunch
+
 
 def is_within_directory(directory, target):
     """Utility function."""
@@ -132,7 +135,7 @@ def load_netset(name: Optional[str] = None, data_home: Optional[Union[str, Path]
     if not data_path.exists():
         name_npz = name + '_npz.tar.gz'
         try:
-            logger.print('Downloading', name, 'from NetSet...')
+            logger.print_log('Downloading', name, 'from NetSet...')
             urlretrieve(folder_npz + name_npz, data_netset / name_npz)
         except HTTPError:
             raise ValueError('Invalid dataset: ' + name + '.'
@@ -141,11 +144,11 @@ def load_netset(name: Optional[str] = None, data_home: Optional[Union[str, Path]
         except ConnectionResetError:  # pragma: no cover
             raise RuntimeError("Could not reach Netset.")
         with tarfile.open(data_netset / name_npz, 'r:gz') as tar_ref:
-            logger.print('Unpacking archive...')
+            logger.print_log('Unpacking archive...')
             safe_extract(tar_ref, data_path)
 
     files = [file for file in listdir(data_path)]
-    logger.print('Parsing files...')
+    logger.print_log('Parsing files...')
     for file in files:
         file_components = file.split('.')
         if len(file_components) == 2:
@@ -159,7 +162,7 @@ def load_netset(name: Optional[str] = None, data_home: Optional[Union[str, Path]
                     dataset[file_name] = pickle.load(f)
 
     clean_data_home(data_netset)
-    logger.print('Done.')
+    logger.print_log('Done.')
     return dataset
 
 
@@ -192,7 +195,7 @@ def load_konect(name: str, data_home: Optional[Union[str, Path]] = None, auto_nu
 
     Notes
     -----
-    An attribute `meta` of the `Bunch` class is used to store information about the dataset if present. In any case,
+    An attribute `meta` of the `Dataset` class is used to store information about the dataset if present. In any case,
     `meta` has the attribute `name` which, if not given, is equal to the name of the dataset as passed to this function.
 
     References
@@ -221,11 +224,11 @@ def load_konect(name: str, data_home: Optional[Union[str, Path]] = None, auto_nu
     data_path = data_konect / name
     name_tar = name + '.tar.bz2'
     if not data_path.exists():
-        logger.print('Downloading', name, 'from Konect...')
+        logger.print_log('Downloading', name, 'from Konect...')
         try:
             urlretrieve('http://konect.cc/files/download.tsv.' + name_tar, data_konect / name_tar)
             with tarfile.open(data_konect / name_tar, 'r:bz2') as tar_ref:
-                logger.print('Unpacking archive...')
+                logger.print_log('Unpacking archive...')
                 safe_extract(tar_ref, data_path)
         except (HTTPError, tarfile.ReadError):
             raise ValueError('Invalid dataset ' + name + '.'
@@ -234,7 +237,7 @@ def load_konect(name: str, data_home: Optional[Union[str, Path]] = None, auto_nu
         except (URLError, ConnectionResetError):  # pragma: no cover
             raise RuntimeError("Could not reach Konect.")
     elif exists(data_path / (name + '_bundle')):
-        logger.print('Loading from local bundle...')
+        logger.print_log('Loading from local bundle...')
         return load_from_numpy_bundle(name + '_bundle', data_path)
 
     dataset = Bunch()
@@ -242,12 +245,12 @@ def load_konect(name: str, data_home: Optional[Union[str, Path]] = None, auto_nu
     if not path.exists() or len(listdir(path)) == 0:
         raise Exception("No data downloaded.")
     files = [file for file in listdir(path) if name in file]
-    logger.print('Parsing files...')
+    logger.print_log('Parsing files...')
     matrix = [file for file in files if 'out.' in file]
     if matrix:
         file = matrix[0]
         directed, bipartite, weighted = load_header(path / file)
-        dataset = from_csv(path / file, directed=directed, bipartite=bipartite, weighted=weighted)
+        dataset = from_csv(path / file, directed=directed, bipartite=bipartite, weighted=weighted, reindex=True)
 
     metadata = [file for file in files if 'meta.' in file]
     if metadata:
@@ -278,7 +281,7 @@ def load_konect(name: str, data_home: Optional[Union[str, Path]] = None, auto_nu
 
 
 def save_to_numpy_bundle(data: Bunch, bundle_name: str, data_home: Optional[Union[str, Path]] = None):
-    """Save a Bunch in the specified data home to a collection of Numpy and Pickle files for faster subsequent loads.
+    """Save a dataset in the specified data home to a collection of Numpy and Pickle files for faster subsequent loads.
 
     Parameters
     ----------
@@ -297,15 +300,13 @@ def save_to_numpy_bundle(data: Bunch, bundle_name: str, data_home: Optional[Unio
             sparse.save_npz(data_path / attribute, data[attribute])
         elif type(data[attribute]) == np.ndarray:
             np.save(data_path / attribute, data[attribute])
-        elif type(data[attribute]) == Bunch or type(data[attribute]) == str:
+        else:
             with open(data_path / (attribute + '.p'), 'wb') as file:
                 pickle.dump(data[attribute], file)
-        else:
-            raise TypeError('Unsupported data attribute type '+str(type(data[attribute])) + '.')
 
 
 def load_from_numpy_bundle(bundle_name: str, data_home: Optional[Union[str, Path]] = None):
-    """Load a Bunch from a collection of Numpy and Pickle files (inverse function of ``save_to_numpy_bundle``).
+    """Load a dataset from a collection of Numpy and Pickle files (inverse function of ``save_to_numpy_bundle``).
 
     Parameters
     ----------
@@ -340,8 +341,8 @@ def load_from_numpy_bundle(bundle_name: str, data_home: Optional[Union[str, Path
 
 
 def save(folder: Union[str, Path], data: Union[sparse.csr_matrix, Bunch]):
-    """Save a Bunch or a CSR matrix in the current directory to a collection of Numpy and Pickle files for faster
-    subsequent loads. Supported attribute types include sparse matrices, NumPy arrays, strings and Bunch.
+    """Save a dataset or a CSR matrix in the current directory to a collection of Numpy and Pickle files for faster
+    subsequent loads. Supported attribute types include sparse matrices, NumPy arrays, strings and objects Dataset.
 
     Parameters
     ----------
@@ -353,11 +354,11 @@ def save(folder: Union[str, Path], data: Union[sparse.csr_matrix, Bunch]):
     Example
     -------
     >>> from sknetwork.data import save
-    >>> my_dataset = Bunch()
-    >>> my_dataset.adjacency = sparse.csr_matrix(np.random.random((3, 3)) < 0.5)
-    >>> my_dataset.names = np.array(['a', 'b', 'c'])
-    >>> save('my_dataset', my_dataset)
-    >>> 'my_dataset' in listdir('.')
+    >>> dataset = Bunch()
+    >>> dataset.adjacency = sparse.csr_matrix(np.random.random((3, 3)) < 0.5)
+    >>> dataset.names = np.array(['a', 'b', 'c'])
+    >>> save('dataset', dataset)
+    >>> 'dataset' in listdir('.')
     True
     """
     folder = Path(folder)
@@ -365,12 +366,12 @@ def save(folder: Union[str, Path], data: Union[sparse.csr_matrix, Bunch]):
     if folder.exists():
         shutil.rmtree(folder)
     if isinstance(data, sparse.csr_matrix):
-        bunch = Bunch()
+        dataset = Bunch()
         if is_square(data):
-            bunch.adjacency = data
+            dataset.adjacency = data
         else:
-            bunch.biadjacency = data
-        data = bunch
+            dataset.biadjacency = data
+        data = dataset
     if folder.is_absolute():
         save_to_numpy_bundle(data, folder, '/')
     else:
@@ -378,7 +379,7 @@ def save(folder: Union[str, Path], data: Union[sparse.csr_matrix, Bunch]):
 
 
 def load(folder: Union[str, Path]):
-    """Load a Bunch from a previously created bundle from the current directory (inverse function of ``save``).
+    """Load a dataset from a previously created bundle from the current directory (inverse function of ``save``).
 
     Parameters
     ----------
@@ -393,13 +394,13 @@ def load(folder: Union[str, Path]):
     Example
     -------
     >>> from sknetwork.data import save
-    >>> my_dataset = Bunch()
-    >>> my_dataset.adjacency = sparse.csr_matrix(np.random.random((3, 3)) < 0.5)
-    >>> my_dataset.names = np.array(['a', 'b', 'c'])
-    >>> save('my_dataset', my_dataset)
-    >>> loaded_graph = load('my_dataset')
-    >>> loaded_graph.names[0]
-    'a'
+    >>> dataset = Bunch()
+    >>> dataset.adjacency = sparse.csr_matrix(np.random.random((3, 3)) < 0.5)
+    >>> dataset.names = np.array(['a', 'b', 'c'])
+    >>> save('dataset', dataset)
+    >>> dataset = load('dataset')
+    >>> print(dataset.names)
+    ['a' 'b' 'c']
     """
     folder = Path(folder)
     if folder.is_absolute():

sknetwork/data/models.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-Created on Jul 1, 2019
+Created in July 2019
 @author: Thomas Bonald <bonald@enst.fr>
 @author: Quentin Lutz <qlutz@enst.fr>
 @author: Nathan de Lara <nathan.delara@polytechnique.org>
@@ -12,8 +12,8 @@ from typing import Union, Optional, Iterable
 import numpy as np
 from scipy import sparse
 
+from sknetwork.data.base import Bunch
 from sknetwork.data.parse import from_edge_list
-from sknetwork.utils import Bunch
 from sknetwork.utils.check import check_random_state
 from sknetwork.utils.format import directed2undirected
 
@@ -36,12 +36,12 @@ def block_model(sizes: Iterable, p_in: Union[float, list, np.ndarray] = .2, p_ou
     self_loops :
          If ``True``, allow self-loops.
     metadata :
-        If ``True``, return a `Bunch` object with labels.
+        If ``True``, return a `Dataset` object with labels.
     seed :
         Seed of the random generator (optional).
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (labels).
 
     Example
@@ -137,11 +137,11 @@ def linear_digraph(n: int = 3, metadata: bool = False) -> Union[sparse.csr_matri
     n : int
         Number of nodes.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -174,11 +174,11 @@ def linear_graph(n: int = 3, metadata: bool = False) -> Union[sparse.csr_matrix,
     n : int
         Number of nodes.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -226,11 +226,11 @@ def cyclic_digraph(n: int = 3, metadata: bool = False) -> Union[sparse.csr_matri
     n : int
         Number of nodes.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -261,11 +261,11 @@ def cyclic_graph(n: int = 3, metadata: bool = False) -> Union[sparse.csr_matrix,
     n : int
         Number of nodes.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -291,11 +291,11 @@ def grid(n1: int = 10, n2: int = 10, metadata: bool = False) -> Union[sparse.csr
     n1, n2 : int
         Grid dimension.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -328,11 +328,11 @@ def star(n_branches: int = 3, metadata: bool = False) -> Union[sparse.csr_matrix
     n_branches : int
         Number of branches.
     metadata : bool
-        If ``True``, return a `Bunch` object with metadata (positions).
+        If ``True``, return a `Dataset` object with metadata (positions).
 
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example
@@ -416,10 +416,10 @@ def watts_strogatz(n: int = 100, degree: int = 6, prob: float = 0.05, seed: Opti
     seed :
         Seed of the random generator (optional).
     metadata :
-        If ``True``, return a `Bunch` object with metadata.
+        If ``True``, return a `Dataset` object with metadata.
     Returns
     -------
-    adjacency or graph : Union[sparse.csr_matrix, Bunch]
+    adjacency or graph : Union[sparse.csr_matrix, Dataset]
         Adjacency matrix or graph with metadata (positions).
 
     Example

sknetwork/data/parse.py

@@ -1,25 +1,25 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-Created on Dec 5, 2018
+Created in December 2018
 @author: Quentin Lutz <qlutz@enst.fr>
-Nathan de Lara <nathan.delara@polytechnique.org>
-Thomas Bonald <bonald@enst.fr>
+@author: Nathan de Lara <nathan.delara@polytechnique.org>
+@author: Thomas Bonald <bonald@enst.fr>
 """
 
 from csv import reader
-from typing import Dict, List, Tuple, Union
+from typing import Dict, List, Tuple, Union, Optional
 from xml.etree import ElementTree
 
 import numpy as np
 from scipy import sparse
 
-from sknetwork.utils import Bunch
+from sknetwork.data.base import Bunch
 from sknetwork.utils.format import directed2undirected
 
 
 def from_edge_list(edge_list: Union[np.ndarray, List[Tuple]], directed: bool = False,
-                   bipartite: bool = False, weighted: bool = True, reindex: bool = True,
+                   bipartite: bool = False, weighted: bool = True, reindex: bool = False, shape: Optional[tuple] = None,
                    sum_duplicates: bool = True, matrix_only: bool = None) -> Union[Bunch, sparse.csr_matrix]:
     """Load a graph from an edge list.
 
@@ -37,16 +37,19 @@ def from_edge_list(edge_list: Union[np.ndarray, List[Tuple]], directed: bool = F
     reindex : bool
         If ``True``, reindex nodes and returns the original node indices as names.
         Reindexing is enforced if nodes are not integers.
+    shape : tuple
+        Shape of the adjacency or biadjacency matrix.
+        If not specified or if nodes are reindexed, the shape is the smallest compatible with node indices.
     sum_duplicates : bool
         If ``True`` (default), sums weights of duplicate edges.
         Otherwise, the weight of each edge is that of the first occurrence of this edge.
     matrix_only : bool
         If ``True``, returns only the adjacency or biadjacency matrix.
-        Otherwise, returns a ``Bunch`` object with graph attributes (e.g., node names).
+        Otherwise, returns a ``Dataset`` object with graph attributes (e.g., node names).
         If not specified (default), selects the most appropriate format.
     Returns
     -------
-    graph : :class:`Bunch` (including node names) or sparse matrix
+    graph : :class:`Dataset` (including node names) or sparse matrix
 
     Examples
     --------
@@ -83,12 +86,14 @@ def from_edge_list(edge_list: Union[np.ndarray, List[Tuple]], directed: bool = F
     else:
         raise TypeError('The edge list must be given as a NumPy array or a list of tuples.')
     return from_edge_array(edge_array=edge_array, weights=weights, directed=directed, bipartite=bipartite,
-                           weighted=weighted, reindex=reindex, sum_duplicates=sum_duplicates, matrix_only=matrix_only)
+                           weighted=weighted, reindex=reindex, shape=shape, sum_duplicates=sum_duplicates,
+                           matrix_only=matrix_only)
 
 
 def from_adjacency_list(adjacency_list: Union[List[List], Dict[str, List]], directed: bool = False,
-                        bipartite: bool = False, weighted: bool = True, reindex: bool = True,
-                        sum_duplicates: bool = True, matrix_only: bool = None) -> Union[Bunch, sparse.csr_matrix]:
+                        bipartite: bool = False, weighted: bool = True, reindex: bool = False,
+                        shape: Optional[tuple] = None, sum_duplicates: bool = True, matrix_only: bool = None) \
+                        -> Union[Bunch, sparse.csr_matrix]:
     """Load a graph from an adjacency list.
 
     Parameters
@@ -104,16 +109,19 @@ def from_adjacency_list(adjacency_list: Union[List[List], Dict[str, List]], dire
     reindex : bool
         If ``True``, reindex nodes and returns the original node indices as names.
         Reindexing is enforced if nodes are not integers.
+    shape : tuple
+        Shape of the adjacency or biadjacency matrix.
+        If not specified or if nodes are reindexed, the shape is the smallest compatible with node indices.
     sum_duplicates : bool
         If ``True`` (default), sums weights of duplicate edges.
         Otherwise, the weight of each edge is that of the first occurrence of this edge.
     matrix_only : bool
         If ``True``, returns only the adjacency or biadjacency matrix.
-        Otherwise, returns a ``Bunch`` object with graph attributes (e.g., node names).
+        Otherwise, returns a ``Dataset`` object with graph attributes (e.g., node names).
         If not specified (default), selects the most appropriate format.
     Returns
     -------
-    graph : :class:`Bunch` or sparse matrix
+    graph : :class:`Dataset` or sparse matrix
 
     Example
     -------
@@ -134,12 +142,12 @@ def from_adjacency_list(adjacency_list: Union[List[List], Dict[str, List]], dire
     else:
         raise TypeError('The adjacency list must be given as a list of lists or a dict of lists.')
     return from_edge_list(edge_list=edge_list, directed=directed, bipartite=bipartite, weighted=weighted,
-                          reindex=reindex, sum_duplicates=sum_duplicates, matrix_only=matrix_only)
+                          reindex=reindex, shape=shape, sum_duplicates=sum_duplicates, matrix_only=matrix_only)
 
 
 def from_edge_array(edge_array: np.ndarray, weights: np.ndarray = None, directed: bool = False, bipartite: bool = False,
-                    weighted: bool = True, reindex: bool = True, sum_duplicates: bool = True,
-                    matrix_only: bool = None) -> Union[Bunch, sparse.csr_matrix]:
+                    weighted: bool = True, reindex: bool = False, shape: Optional[tuple] = None,
+                    sum_duplicates: bool = True, matrix_only: bool = None) -> Union[Bunch, sparse.csr_matrix]:
     """Load a graph from an edge array of shape (n_edges, 2) and weights (optional).
 
     Parameters
@@ -157,17 +165,20 @@ def from_edge_array(edge_array: np.ndarray, weights: np.ndarray = None, directed
     reindex : bool
         If ``True``, reindex nodes and returns the original node indices as names.
         Reindexing is enforced if nodes are not integers.
+    shape : tuple
+        Shape of the adjacency or biadjacency matrix.
+        If not specified or if nodes are reindexed, the shape is the smallest compatible with node indices.
     sum_duplicates : bool
         If ``True`` (default), sums weights of duplicate edges.
         Otherwise, the weight of each edge is that of the first occurrence of this edge.
     matrix_only : bool
         If ``True``, returns only the adjacency or biadjacency matrix.
-        Otherwise, returns a ``Bunch`` object with graph attributes (e.g., node names).
+        Otherwise, returns a ``Dataset`` object with graph attributes (e.g., node names).
         If not specified (default), selects the most appropriate format.
 
     Returns
     -------
-    graph : :class:`Bunch` or sparse matrix
+    graph : :class:`Dataset` or sparse matrix
     """
     try:
         edge_array = edge_array.astype(float)
@@ -195,28 +206,34 @@ def from_edge_array(edge_array: np.ndarray, weights: np.ndarray = None, directed
     if bipartite:
         row = edge_array[:, 0]
         col = edge_array[:, 1]
-        if row.dtype != int or (reindex and len(set(row)) < max(row) + 1):
+        if row.dtype != int or reindex:
             names_row, row = np.unique(row, return_inverse=True)
             graph.names_row = names_row
             graph.names = names_row
             n_row = len(names_row)
+        elif shape is not None:
+            n_row = max(shape[0], max(row) + 1)
         else:
             n_row = max(row) + 1
-        if col.dtype != int or (reindex and len(set(col)) < max(col) + 1):
+        if col.dtype != int or reindex:
             names_col, col = np.unique(col, return_inverse=True)
             graph.names_col = names_col
             n_col = len(names_col)
+        elif shape is not None:
+            n_col = max(shape[1], max(col) + 1)
         else:
             n_col = max(col) + 1
         matrix = sparse.csr_matrix((weights, (row, col)), shape=(n_row, n_col))
         graph.biadjacency = matrix
     else:
         nodes = edge_array.ravel()
-        if nodes.dtype != int or (reindex and len(set(nodes)) < max(nodes) + 1):
+        if nodes.dtype != int or reindex:
             names, nodes = np.unique(nodes, return_inverse=True)
             graph.names = names
             n = len(names)
             edge_array = nodes.reshape(-1, 2)
+        elif shape is not None:
+            n = max(shape[0], max(nodes) + 1)
         else:
             n = max(nodes) + 1
         row = edge_array[:, 0]
@@ -233,8 +250,8 @@ def from_edge_array(edge_array: np.ndarray, weights: np.ndarray = None, directed
 
 def from_csv(file_path: str, delimiter: str = None, sep: str = None, comments: str = '#%',
              data_structure: str = None, directed: bool = False, bipartite: bool = False, weighted: bool = True,
-             reindex: bool = True, sum_duplicates: bool = True, matrix_only: bool = None) \
-        -> Union[Bunch, sparse.csr_matrix]:
+             reindex: bool = False, shape: Optional[tuple] = None, sum_duplicates: bool = True,
+             matrix_only: bool = None) -> Union[Bunch, sparse.csr_matrix]:
     """Load a graph from a CSV or TSV file.
     The delimiter can be specified (e.g., ' ' for space-separated values).
 
@@ -249,9 +266,10 @@ def from_csv(file_path: str, delimiter: str = None, sep: str = None, comments: s
     comments : str
         Characters for comment lines.
     data_structure : str
-        If 'edge_list', considers each row of the file as an edge (tuple of size 2 or 3).
-        If 'adjacency_list', considers each row of the file as an adjacency list (list of neighbors).
-        If 'adjacency_dict', considers each row of the file as an adjacency dictionary with key
+        If 'edge_list', consider each row of the file as an edge (tuple of size 2 or 3).
+        If 'adjacency_list', consider each row of the file as an adjacency list (list of neighbors,
+        in the order of node indices; an empty line means no neighbor).
+        If 'adjacency_dict', consider each row of the file as an adjacency dictionary with key
         given by the first column (node: list of neighbors).
         If ``None`` (default), data_structure is guessed from the first rows of the file.
     directed : bool
@@ -263,17 +281,20 @@ def from_csv(file_path: str, delimiter: str = None, sep: str = None, comments: s
     reindex : bool
         If ``True``, reindex nodes and returns the original node indices as names.
         Reindexing is enforced if nodes are not integers.
+    shape : tuple
+        Shape of the adjacency or biadjacency matrix.
+        If not specified or if nodes are reindexed, the shape is the smallest compatible with node indices.
     sum_duplicates : bool
         If ``True`` (default), sums weights of duplicate edges.
         Otherwise, the weight of each edge is that of the first occurrence of this edge.
     matrix_only : bool
         If ``True``, returns only the adjacency or biadjacency matrix.
-        Otherwise, returns a ``Bunch`` object with graph attributes (e.g., node names).
+        Otherwise, returns a ``Dataset`` object with graph attributes (e.g., node names).
         If not specified (default), selects the most appropriate format.
 
     Returns
     -------
-    graph: :class:`Bunch` or sparse matrix
+    graph: :class:`Dataset` or sparse matrix
     """
     header_length, delimiter_guess, comment_guess, data_structure_guess = scan_header(file_path, delimiters=delimiter,
                                                                                       comments=comments)
@@ -295,7 +316,7 @@ def from_csv(file_path: str, delimiter: str = None, sep: str = None, comments: s
             else:
                 weights = None
             return from_edge_array(edge_array=edge_array, weights=weights, directed=directed, bipartite=bipartite,
-                                   weighted=weighted, reindex=reindex, sum_duplicates=sum_duplicates,
+                                   weighted=weighted, reindex=reindex, shape=shape, sum_duplicates=sum_duplicates,
                                    matrix_only=matrix_only)
         except TypeError:
             pass
@@ -306,17 +327,17 @@ def from_csv(file_path: str, delimiter: str = None, sep: str = None, comments: s
         if data_structure == 'edge_list':
             edge_list = [tuple(row) for row in csv_reader]
             return from_edge_list(edge_list=edge_list, directed=directed, bipartite=bipartite,
-                                  weighted=weighted, reindex=reindex, sum_duplicates=sum_duplicates,
+                                  weighted=weighted, reindex=reindex, shape=shape, sum_duplicates=sum_duplicates,
                                   matrix_only=matrix_only)
         elif data_structure == 'adjacency_list':
             adjacency_list = [row for row in csv_reader]
             return from_adjacency_list(adjacency_list=adjacency_list, directed=directed, bipartite=bipartite,
-                                       weighted=weighted, reindex=reindex, sum_duplicates=sum_duplicates,
+                                       weighted=weighted, reindex=reindex, shape=shape, sum_duplicates=sum_duplicates,
                                        matrix_only=matrix_only)
         elif data_structure == 'adjacency_dict':
             adjacency_list = {row[0]: row[1:] for row in csv_reader}
             return from_adjacency_list(adjacency_list=adjacency_list, directed=directed, bipartite=bipartite,
-                                       weighted=weighted, reindex=reindex, sum_duplicates=sum_duplicates,
+                                       weighted=weighted, reindex=reindex, shape=shape, sum_duplicates=sum_duplicates,
                                        matrix_only=matrix_only)
 
 
@@ -439,7 +460,7 @@ def from_graphml(file_path: str, weight_key: str = 'weight', max_string_size: in
     Returns
     -------
     data: :class:`Bunch`
-        The dataset in a bunch with the adjacency as a CSR matrix.
+        The dataset in a Dataset with the adjacency as a CSR matrix.
     """
     # see http://graphml.graphdrawing.org/primer/graphml-primer.html
     # and http://graphml.graphdrawing.org/specification/dtd.html#top

sknetwork/data/test_graphs.py

@@ -40,7 +40,7 @@ def test_bigraph():
     return sparse.csr_matrix((data, (row, col)), shape=(6, 8))
 
 
-def test_graph_disconnect():
+def test_disconnected_graph():
     """Simple disconnected undirected graph, used for testing.
     10 nodes, 10 edges.
     """
@@ -68,7 +68,7 @@ def test_graph_bool():
     return adjacency
 
 
-def test_graph_clique():
+def test_clique():
     """Clique graph, used for testing (10 nodes, 45 edges).
     """
     n = 10

sknetwork/data/tests/test_API.py

@@ -8,7 +8,7 @@ import warnings
 
 from sknetwork.data.load import *
 from sknetwork.data.toy_graphs import *
-from sknetwork.utils import Bunch
+from sknetwork.data import Bunch
 
 
 class TestDataAPI(unittest.TestCase):

sknetwork/data/tests/test_base.py

@@ -0,0 +1,14 @@
+# -*- coding: utf-8 -*-
+"""tests for dataset"""
+
+import unittest
+
+from sknetwork.data.base import Bunch
+
+
+class TestDataset(unittest.TestCase):
+
+    def test(self):
+        dataset = Bunch(name='dataset')
+        self.assertEqual(dataset.name, 'dataset')
+        self.assertEqual(dataset['name'], 'dataset')

sknetwork/data/tests/test_load.py

@@ -10,7 +10,7 @@ import numpy as np
 
 from sknetwork.data.load import load_netset, load_konect, clear_data_home, save, load
 from sknetwork.data.toy_graphs import house, star_wars
-from sknetwork.utils.timeout import TimeOut
+from sknetwork.data.timeout import TimeOut
 
 
 class TestLoader(unittest.TestCase):