support for Windows & CI (#32)

* CI: test Win
* add badge
* req. GCO 3.0.8
* fix pandas
* fix Win

.github/workflows/ci-testing.yml

@@ -12,7 +12,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-20.04, macOS-10.15] # , windows-2019
+        os: [ubuntu-20.04, macOS-10.15, windows-2019]
         python-version: [3.6, 3.8, 3.9]
         requires: ['latest']
         include:
@@ -81,7 +81,7 @@ jobs:
         mkdir output
         mkdir results
         # run tests with coverage
-        coverage run --source imsegm -m pytest imsegm tests --durations=25 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
+        python -m pytest imsegm tests --cov=imsegm --durations=25 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
 
     - name: Upload pytest test results
       uses: actions/upload-artifact@master

README.md

@@ -10,6 +10,7 @@
 [![Documentation Status](https://readthedocs.org/projects/pyimsegm/badge/?version=latest)](https://pyimsegm.readthedocs.io/en/latest/?badge=latest)
 [![Gitter](https://badges.gitter.im/pyImSegm/community.svg)](https://gitter.im/pyImSegm/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge)
 ![CI experiments](https://github.com/Borda/pyImSegm/workflows/CI%20experiments/badge.svg?branch=master&event=push)
+[![pre-commit.ci status](https://results.pre-commit.ci/badge/github/Borda/pyImSegm/master.svg)](https://results.pre-commit.ci/latest/github/Borda/pyImSegm/master)
 
 <!--
 [![Run Status](https://api.shippable.com/projects/5962ea48a125960700c197f8/badge?branch=master)](https://app.shippable.com/github/Borda/pyImSegm)

experiments_ovary_centres/run_center_candidate_training.py

@@ -27,6 +27,7 @@
 import argparse
 import logging
 import os
+import platform
 import sys
 from functools import partial
 
@@ -78,7 +79,7 @@
 CROSS_VAL_LEAVE_OUT_EVAL = 0.1
 
 CENTER_PARAMS = {
-    'computer': os.uname(),
+    'computer': platform.uname(),
     'slic_size': 25,
     'slic_regul': 0.3,
     # 'fts_hist_diams': None,
@@ -194,8 +195,7 @@ def arg_parse_params(params):
     # load saved configuration
     if params['path_config'] is not None:
         ext = os.path.splitext(params['path_config'])[-1]
-        assert (ext == '.yaml' or ext == '.yml'), \
-            'wrong extension for %s' % params['path_config']
+        assert ext in ('.yaml', '.yml'), 'wrong extension for %s' % params['path_config']
         data = tl_expt.load_config_yaml(params['path_config'])
         params.update(data)
     params.update(paths)
@@ -209,8 +209,7 @@ def is_drawing(path_out):
     :param str path_out:
     :return bool:
     # """
-    bool_res = path_out is not None and os.path.exists(path_out) \
-        and logging.getLogger().isEnabledFor(logging.DEBUG)
+    bool_res = path_out is not None and os.path.exists(path_out) and logging.getLogger().isEnabledFor(logging.DEBUG)
     return bool_res
 
 
@@ -284,8 +283,7 @@ def load_image_segm_center(idx_row, path_out=None, dict_relabel=None):
         if dict_relabel is not None:
             segm = seg_lbs.relabel_by_dict(segm, dict_relabel)
 
-    if row_path['path_centers'] is not None \
-            and os.path.isfile(row_path['path_centers']):
+    if row_path['path_centers'] is not None and os.path.isfile(row_path['path_centers']):
         ext = os.path.splitext(os.path.basename(row_path['path_centers']))[-1]
         if ext == '.csv':
             centers = tl_data.load_landmarks_csv(row_path['path_centers'])
@@ -384,8 +382,7 @@ def estim_points_compute_features(name, img, segm, params):
     :return (str, ndarray, [(int, int)], [[float]], list(str)):
     """
     # superpixels on image
-    assert img.shape[:2] == segm.shape[:2], \
-        'not matching shapes: %r : %r' % (img.shape, segm.shape)
+    assert img.shape[:2] == segm.shape[:2], 'not matching shapes: %r : %r' % (img.shape, segm.shape)
     slic = seg_spx.segment_slic_img2d(img, params['slic_size'], params['slic_regul'])
     slic_centers = seg_spx.superpixel_centers(slic)
     # slic_edges = seg_spx.make_graph_segm_connect_grid2d_conn4(slic)
@@ -469,9 +466,7 @@ def label_close_points(centers, points, params):
     else:
         logging.warning('not relevant centers info of type "%s"', type(centers))
         labels = [-1] * len(points)
-    assert len(points) == len(labels), \
-        'not equal lenghts of points (%i) and labels (%i)' \
-        % (len(points), len(labels))
+    assert len(points) == len(labels), 'not equal lenghts of points (%i) and labels (%i)' % (len(points), len(labels))
     return labels
 
 
@@ -713,8 +708,7 @@ def experiment_loo(
 def prepare_experiment_folder(params, dir_template):
     params['path_expt'] = os.path.join(params['path_output'], dir_template % params['name'])
     if not os.path.exists(params['path_expt']):
-        assert os.path.isdir(os.path.dirname(params['path_expt'])), \
-            'missing: %s' % os.path.dirname(params['path_expt'])
+        assert os.path.isdir(os.path.dirname(params['path_expt'])), 'missing: %s' % os.path.dirname(params['path_expt'])
         logging.debug('creating missing folder: %s', params['path_expt'])
         os.mkdir(params['path_expt'])
     return params
@@ -757,8 +751,7 @@ def main_train(params):
 
     path_dump_data = os.path.join(params['path_expt'], NAME_DUMP_TRAIN_DATA)
     if not os.path.isfile(path_dump_data) or FORCE_RECOMP_DATA:
-        (dict_imgs, dict_segms, dict_slics, dict_points, dict_centers,
-         dict_features, dict_labels, feature_names) = \
+        dict_imgs, dict_segms, dict_slics, dict_points, dict_centers, dict_features, dict_labels, feature_names = \
             dataset_load_images_segms_compute_features(params, df_paths, params['nb_workers'])
         assert len(dict_imgs) > 0, 'missing images'
         save_dump_data(
@@ -788,9 +781,7 @@ def main_train(params):
     # remove all bad values from features space
     features[np.isnan(features)] = 0
     features[np.isinf(features)] = -1
-    assert np.sum(sizes) == len(labels), \
-        'not equal sizes (%d) and labels (%i)' \
-        % (int(np.sum(sizes)), len(labels))
+    assert np.sum(sizes) == len(labels), 'not equal sizes (%d) and labels (%i)' % (int(np.sum(sizes)), len(labels))
 
     # feature norm & train classification
     nb_holdout = int(np.ceil(len(sizes) * CROSS_VAL_LEAVE_OUT_SEARCH))

experiments_ovary_centres/run_center_prediction.py

@@ -16,6 +16,7 @@
 import gc
 import logging
 import os
+import platform
 import sys
 import time
 from functools import partial
@@ -128,7 +129,7 @@ def main(params):
 
     # run_train.check_pathes_patterns(paths)
     tl_expt.set_experiment_logger(params['path_expt'])
-    logging.info('COMPUTER: \n%r', os.uname())
+    logging.info('COMPUTER: \n%r', platform.uname())
     logging.info(tl_expt.string_dict(params, desc='PARAMETERS'))
 
     tl_expt.create_subfolders(params['path_expt'], LIST_SUBFOLDER)

experiments_ovary_detect/run_cut_segmented_objects.py

@@ -71,8 +71,7 @@ def arg_parse_params(dict_paths):
     _fn_path = lambda k: os.path.join(tl_data.update_path(os.path.dirname(args[k])), os.path.basename(args[k]))
     dict_paths = {k.split('_')[-1]: _fn_path(k) for k in args if k.startswith('path_')}
     for k in dict_paths:
-        assert os.path.exists(os.path.dirname(dict_paths[k])), \
-            'missing (%s) "%s"' % (k, os.path.dirname(dict_paths[k]))
+        assert os.path.exists(os.path.dirname(dict_paths[k])), 'missing (%s) "%s"' % (k, os.path.dirname(dict_paths[k]))
     return dict_paths, args
 
 
@@ -85,8 +84,7 @@ def export_cut_objects(df_row, path_out, padding, use_mask=True, bg_color=None):
     """
     annot, _ = tl_data.load_image_2d(df_row['path_1'])
     img, name = tl_data.load_image_2d(df_row['path_2'])
-    assert annot.shape[:2] == img.shape[:2], \
-        'image sizes not match %r vs %r' % (annot.shape, img.shape)
+    assert annot.shape[:2] == img.shape[:2], 'image sizes not match %r vs %r' % (annot.shape, img.shape)
 
     uq_objects = np.unique(annot)
     if len(uq_objects) == 1:
@@ -107,8 +105,7 @@ def main(dict_paths, padding=0, use_mask=False, bg_color=None, nb_workers=NB_WOR
     :param int nb_workers:
     """
     if not os.path.isdir(dict_paths['output']):
-        assert os.path.isdir(os.path.dirname(dict_paths['output'])), \
-            '"%s" should be folder' % dict_paths['output']
+        assert os.path.isdir(os.path.dirname(dict_paths['output'])), '"%s" should be folder' % dict_paths['output']
         logging.debug('creating dir: %s', dict_paths['output'])
         os.mkdir(dict_paths['output'])
 

experiments_segmentation/run_segm_slic_classif_graphcut.py

@@ -233,8 +233,7 @@ def dataset_load_images_annot_compute_features(params, show_debug_imgs=SHOW_DEBU
     :return ({str: ndarray} * 6, list(str)):
     """
     dict_images, dict_annots = dict(), dict()
-    dict_slics, dict_features, dict_labels, dict_label_hist = \
-        dict(), dict(), dict(), dict()
+    dict_slics, dict_features, dict_labels, dict_label_hist = dict(), dict(), dict(), dict()
     feature_names = list()
 
     # compute features
@@ -367,8 +366,7 @@ def segment_image(imgs_idx_path, params, classif, path_out, path_visu=None, show
     np.savez_compressed(path_npz, segm_soft)
 
     # plt.imsave(os.path.join(path_out, idx_name + '_rgb.png'), seg_pipe)
-    if params.get('visual', False) and path_visu is not None \
-            and os.path.isdir(path_visu):
+    if params.get('visual', False) and path_visu is not None and os.path.isdir(path_visu):
         export_draw_image_segm_contour(img, segm_gc, path_visu, idx_name, '_GC')
         export_draw_image_segm_contour(img, segm_map, path_visu, idx_name, '_MAP')
         if show_debug_imgs and debug_visual is not None:
@@ -404,8 +402,7 @@ def eval_segment_with_annot(
     if dict_label_hist is not None:
         visu_histogram_labels(params, dict_label_hist)
     assert sorted(dict_annot.keys()) == sorted(dict_segm.keys()), \
-        'mismatch in dictionary keys: \n%s \n%s' % (sorted(dict_annot.keys()),
-                                                    sorted(dict_segm.keys()))
+        'mismatch in dictionary keys: \n%s \n%s' % (sorted(dict_annot.keys()), sorted(dict_segm.keys()))
     list_annot = [dict_annot[n] for n in dict_annot]
     list_segm = [dict_segm[n] for n in dict_annot]
     df_stat = seg_clf.compute_stat_per_image(
@@ -438,8 +435,7 @@ def retrain_lpo_segment_image(
     :param bool show_debug_imgs: whether show debug images
     :return (str, ndarray, ndarray):
     """
-    dict_imgs, _, _, dict_features, dict_labels, _, _ = \
-        load_dump_data(path_dump)
+    dict_imgs, _, _, dict_features, dict_labels, _, _ = load_dump_data(path_dump)
     dict_classif = seg_clf.load_classifier(path_classif)
     classif = dict_classif['clf_pipeline']
     params = dict_classif['params']

imsegm/__init__.py

@@ -11,12 +11,12 @@
     import traceback
     traceback.print_exc()
 
-__version__ = '0.1.8'
+__version__ = '0.1.9'
 __author__ = 'Jiri Borovec'
 __author_email__ = 'jiri.borovec@fel.cvut.cz'
 __license__ = 'BSD 3-clause'
 __homepage__ = 'https://borda.github.io/pyImSegm'
-__copyright__ = 'Copyright (c) 2014-2020, %s.' % __author__
+__copyright__ = 'Copyright (c) 2014-2021, %s.' % __author__
 __doc__ = 'General superpixel image segmentation: (un)supervised, center detection, region growing'
 __long_doc__ = "# %s" % __doc__ + """
 

imsegm/annotation.py

@@ -232,12 +232,12 @@ def image_color_2_labels(img, colors=None):
     >>> np.random.seed(0)
     >>> rand = np.random.randint(0, 2, (5, 7)).astype(np.uint8)
     >>> img = np.rollaxis(np.array([rand] * 3), 0, 3)
-    >>> image_color_2_labels(img)
+    >>> image_color_2_labels(img)  # doctest: +ELLIPSIS
     array([[1, 0, 0, 1, 0, 0, 0],
            [0, 0, 0, 0, 1, 1, 0],
            [1, 1, 1, 1, 1, 0, 1],
            [0, 0, 1, 1, 0, 0, 0],
-           [0, 1, 0, 1, 0, 1, 0]])
+           [0, 1, 0, 1, 0, 1, 0]]...)
     """
     if not colors:
         colors = image_frequent_colors(img).keys()
@@ -330,9 +330,9 @@ def load_info_group_by_slices(path_txt, stages, pos_columns=COLUMNS_POSITION, di
     :return: DF
 
     >>> from imsegm.utilities.data_io import update_path
-    >>> path_txt = os.path.join(update_path('data-images'),
-    ...                 'drosophila_ovary_slice', 'info_ovary_images.txt')
-    >>> load_info_group_by_slices(path_txt, [4]) # doctest: +NORMALIZE_WHITESPACE
+    >>> path_txt = os.path.join(update_path('data-images'), 'drosophila_ovary_slice', 'info_ovary_images.txt')
+    >>> df = load_info_group_by_slices(path_txt, [4])
+    >>> df.sort_index(axis=1) # doctest: +NORMALIZE_WHITESPACE
                 ant_x  ant_y  lat_x  lat_y post_x post_y
     image
     insitu7569  [298]  [327]  [673]  [411]  [986]  [155]

imsegm/classification.py

@@ -90,11 +90,9 @@ def create_classifiers(nb_workers=-1):
     >>> classifs = create_classifiers()
     >>> classifs  # doctest: +ELLIPSIS
     {...}
-    >>> sum([isinstance(create_clf_param_search_grid(k), dict)
-    ...      for k in classifs.keys()])
+    >>> sum([isinstance(create_clf_param_search_grid(k), dict) for k in classifs.keys()])
     7
-    >>> sum([isinstance(create_clf_param_search_distrib(k), dict)
-    ...      for k in classifs.keys()])
+    >>> sum([isinstance(create_clf_param_search_distrib(k), dict) for k in classifs.keys()])
     7
     """
     clfs = {
@@ -383,20 +381,17 @@ def compute_classif_stat_segm_annot(annot_segm_name, drop_labels=None, relabel=F
     >>> np.random.seed(0)
     >>> annot = np.random.randint(0, 2, (5, 10))
     >>> segm = np.random.randint(0, 2, (5, 10))
-    >>> d = compute_classif_stat_segm_annot((annot, annot, 'ttt'), relabel=True,
-    ...                                     drop_labels=[5])
+    >>> d = compute_classif_stat_segm_annot((annot, annot, 'ttt'), relabel=True, drop_labels=[5])
     >>> d['(FP+FN)/(TP+FN)']  # doctest: +ELLIPSIS
     0.0
     >>> d['(TP+FP)/(TP+FN)']  # doctest: +ELLIPSIS
     1.0
-    >>> d = compute_classif_stat_segm_annot((annot, segm, 'ttt'), relabel=True,
-    ...                                     drop_labels=[5])
+    >>> d = compute_classif_stat_segm_annot((annot, segm, 'ttt'), relabel=True, drop_labels=[5])
     >>> d['(FP+FN)/(TP+FN)']  # doctest: +ELLIPSIS
     0.846...
     >>> d['(TP+FP)/(TP+FN)']  # doctest: +ELLIPSIS
     1.153...
-    >>> d = compute_classif_stat_segm_annot((annot, segm + 1, 'ttt'),
-    ...                                     relabel=False, drop_labels=[0])
+    >>> d = compute_classif_stat_segm_annot((annot, segm + 1, 'ttt'), relabel=False, drop_labels=[0])
     >>> d['confusion']
     [[13, 17], [0, 0]]
     """
@@ -487,14 +482,13 @@ def feature_scoring_selection(features, labels, names=None, path_out=''):
     :return tuple(list(int),DF): indices, Dataframe with scoring
 
     >>> from sklearn.datasets import make_classification
-    >>> features, labels = make_classification(n_samples=250, n_features=5,
-    ...                                        n_informative=3, n_redundant=0,
-    ...                                        n_repeated=0, n_classes=2,
-    ...                                        random_state=0, shuffle=False)
-    >>> indices, df_scoring = feature_scoring_selection(features, labels)
+    >>> features, labels = make_classification(
+    ...     n_samples=250, n_features=5, n_informative=3, n_redundant=0, n_repeated=0,
+    ...     n_classes=2, random_state=0, shuffle=False)
+    >>> indices, df_scoring = feature_scoring_selection(features, labels)  # doctest: +ELLIPSIS
     >>> indices
-    array([1, 0, 2, 3, 4])
-    >>> df_scoring  # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
+    array([1, 0, 2, 3, 4]...)
+    >>> df_scoring.sort_index(axis=1)  # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
              ExtTree    F-test    k-Best variance
     feature
     1        0.24...   0.75...   0.75...  2.49...
@@ -506,8 +500,8 @@ def feature_scoring_selection(features, labels, names=None, path_out=''):
     >>> path_out = 'test_fts-select'
     >>> os.mkdir(path_out)
     >>> indices, df_scoring = feature_scoring_selection(features.tolist(), labels.tolist(), path_out=path_out)
-    >>> indices
-    array([1, 0, 3, 4, 2])
+    >>> indices  # doctest: +ELLIPSIS
+    array([1, 0, 3, 4, 2]...)
     >>> import shutil
     >>> shutil.rmtree(path_out, ignore_errors=True)
     """
@@ -562,8 +556,8 @@ def save_classifier(path_out, classif, clf_name, params, feature_names=None, lab
 
     >>> clf = create_classifiers()['RandForest']
     >>> p_clf = save_classifier('.', clf, 'TESTINNG', {})
-    >>> p_clf
-    './classifier_TESTINNG.pkl'
+    >>> os.path.basename(p_clf)
+    'classifier_TESTINNG.pkl'
     >>> d_clf = load_classifier(p_clf)
     >>> sorted(d_clf.keys())
     ['clf_pipeline', 'features', 'label_names', 'name', 'params']
@@ -703,14 +697,14 @@ def create_classif_search_train_export(
     Fitting ...
     >>> clf  # doctest: +ELLIPSIS
     Pipeline(...)
-    >>> p_clf
-    './classifier_RandForest.pkl'
+    >>> os.path.basename(p_clf)
+    'classifier_RandForest.pkl'
     >>> os.remove(p_clf)
     >>> import glob
     >>> files = glob.glob(os.path.join('.', 'classif_*.txt'))
-    >>> sorted(files)  # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
-    ['./classif_RandForest_search_params_best.txt',
-     './classif_RandForest_search_params_scores.txt']
+    >>> sorted([os.path.basename(fp) for fp in files])  # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS
+    ['classif_RandForest_search_params_best.txt',
+     'classif_RandForest_search_params_scores.txt']
     >>> for p in files: os.remove(p)
     """
     assert list(labels), 'some labels has to be given'
@@ -1189,8 +1183,7 @@ def balance_dataset_by_(features, labels, balance_type='random', min_samples=Non
     :return tuple(ndarray,ndarray):
 
     >>> np.random.seed(0)
-    >>> fts, lbs = balance_dataset_by_(np.random.random((25, 3)),
-    ...                                np.random.randint(0, 2, 25))
+    >>> fts, lbs = balance_dataset_by_(np.random.random((25, 3)), np.random.randint(0, 2, 25))
     >>> fts.shape
     (24, 3)
     >>> lbs

imsegm/descriptors.py

@@ -1701,8 +1701,7 @@ def numpy_ray_features_seg2d(seg_binary, position, angle_step=5., edge='up'):
         last = seg_binary[position[0], position[1]]
         for _ in range(segm_diag):
             pos += grad
-            if pos[0] < 0 or round(pos[0]) >= height \
-                    or pos[1] < 0 or round(pos[1]) >= width:
+            if pos[0] < 0 or round(pos[0]) >= height or pos[1] < 0 or round(pos[1]) >= width:
                 break
             actual = seg_binary[int(round(pos[0])), int(round(pos[1]))]
             if (edge == 'up' and actual) or (edge == 'down' and last and not actual):