Merge pull request #549 from NatLibFi/update-dependencies-v0.56

Update dependencies v0.56

annif/backend/omikuji.py

@@ -5,7 +5,8 @@
 import shutil
 import annif.util
 from annif.suggestion import SubjectSuggestion, ListSuggestionResult
-from annif.exception import NotInitializedException, NotSupportedException
+from annif.exception import NotInitializedException, NotSupportedException, \
+    OperationFailedException
 from . import backend
 from . import mixins
 
@@ -40,7 +41,12 @@ def _initialize_model(self):
             path = os.path.join(self.datadir, self.MODEL_FILE)
             self.debug('loading model from {}'.format(path))
             if os.path.exists(path):
-                self._model = omikuji.Model.load(path)
+                try:
+                    self._model = omikuji.Model.load(path)
+                except RuntimeError:
+                    raise OperationFailedException(
+                        "Omikuji models trained on Annif versions older than "
+                        "0.56 cannot be loaded. Please retrain your project.")
             else:
                 raise NotInitializedException(
                     'model {} not found'.format(path),

annif/backend/pav.py

@@ -95,9 +95,9 @@ def _suggest_train_corpus(source_project, corpus):
         scores = coo_matrix((data, (row, col)),
                             shape=(ndocs, len(source_project.subjects)),
                             dtype=np.float32)
-        true = coo_matrix((np.ones(len(trow), dtype=np.bool), (trow, tcol)),
+        true = coo_matrix((np.ones(len(trow), dtype=bool), (trow, tcol)),
                           shape=(ndocs, len(source_project.subjects)),
-                          dtype=np.bool)
+                          dtype=bool)
         return csc_matrix(scores), csc_matrix(true)
 
     def _create_pav_model(self, source_project_id, min_docs, corpus):

annif/backend/tfidf.py

@@ -4,18 +4,13 @@
 import os.path
 import tempfile
 import annif.util
+import gensim.similarities
+from gensim.matutils import Sparse2Corpus
 from annif.suggestion import VectorSuggestionResult
 from annif.exception import NotInitializedException, NotSupportedException
 from . import backend
 from . import mixins
 
-# Filter UserWarnings due to not-installed python-Levenshtein package
-import warnings
-with warnings.catch_warnings():
-    warnings.simplefilter('ignore')
-    import gensim.similarities
-    from gensim.matutils import Sparse2Corpus
-
 
 class SubjectBuffer:
     """A file-backed buffer to store and retrieve subject text."""

annif/eval.py

@@ -15,7 +15,7 @@ def filter_pred_top_k(preds, limit):
 
     masks = []
     for pred in preds:
-        mask = np.zeros_like(pred, dtype=np.bool)
+        mask = np.zeros_like(pred, dtype=bool)
         top_k = np.argsort(pred)[::-1][:limit]
         mask[top_k] = True
         masks.append(mask)

annif/lexical/mllm.py

@@ -82,7 +82,7 @@ def candidates_to_features(candidates, mdata):
 
     matrix = np.zeros((len(candidates), len(Feature)), dtype=np.float32)
     c_ids = [c.subject_id for c in candidates]
-    c_vec = np.zeros(mdata.related.shape[0], dtype=np.bool)
+    c_vec = np.zeros(mdata.related.shape[0], dtype=bool)
     c_vec[c_ids] = True
     broader = mdata.broader.multiply(c_vec).sum(axis=1)
     narrower = mdata.narrower.multiply(c_vec).sum(axis=1)

annif/lexical/util.py

@@ -3,7 +3,6 @@
 import collections
 from rdflib import URIRef
 from rdflib.namespace import SKOS
-import numpy as np
 from scipy.sparse import lil_matrix, csc_matrix
 
 
@@ -16,7 +15,7 @@ def get_subject_labels(graph, uri, properties, language):
 
 def make_relation_matrix(graph, vocab, property):
     n_subj = len(vocab.subjects)
-    matrix = lil_matrix((n_subj, n_subj), dtype=np.bool)
+    matrix = lil_matrix((n_subj, n_subj), dtype=bool)
 
     for subj, obj in graph.subject_objects(property):
         subj_id = vocab.subjects.by_uri(str(subj), warnings=False)
@@ -36,7 +35,7 @@ def make_collection_matrix(graph, vocab):
             c_members[str(coll)].append(member_id)
 
     c_matrix = lil_matrix((len(c_members), len(vocab.subjects)),
-                          dtype=np.bool)
+                          dtype=bool)
 
     # populate the matrix for collection -> subject_id
     for c_id, members in enumerate(c_members.values()):

annif/suggestion.py

@@ -135,15 +135,15 @@ def filter(self, subject_index, limit=None, threshold=0.0):
         mask = (self._vector > threshold)
         deprecated_ids = subject_index.deprecated_ids()
         if limit is not None:
-            limit_mask = np.zeros_like(self._vector, dtype=np.bool)
+            limit_mask = np.zeros_like(self._vector, dtype=bool)
             deprecated_set = set(deprecated_ids)
             top_k_subjects = itertools.islice(
                                 (subj for subj in self.subject_order
                                  if subj not in deprecated_set), limit)
             limit_mask[list(top_k_subjects)] = True
             mask = mask & limit_mask
         else:
-            deprecated_mask = np.ones_like(self._vector, dtype=np.bool)
+            deprecated_mask = np.ones_like(self._vector, dtype=bool)
             deprecated_mask[deprecated_ids] = False
             mask = mask & deprecated_mask
         vsr = VectorSuggestionResult(self._vector * mask)

setup.py

@@ -20,30 +20,30 @@ def read(fname):
     zip_safe=False,
     python_requires='>=3.7',
     install_requires=[
-        'connexion[swagger-ui]',
+        'connexion2[swagger-ui]',
         'swagger_ui_bundle',
-        'flask>=1.0.4,<2',
+        'flask>=1.0.4,<3',
         'flask-cors',
-        'click==7.1.*',
+        'click==8.0.*',
         'click-log',
-        'joblib==1.0.1',
+        'joblib==1.1.0',
         'nltk',
-        'gensim==4.0.*',
-        'scikit-learn==0.24.2',
-        'scipy==1.5.4',
+        'gensim==4.1.*',
+        'scikit-learn==1.0.2',
+        'scipy==1.7.*',
         'rdflib>=4.2,<7.0',
         'gunicorn',
-        'numpy==1.19.*',
-        'optuna==2.8.0',
+        'numpy==1.21.*',
+        'optuna==2.10.*',
         'stwfsapy==0.3.*',
         'python-dateutil',
     ],
     tests_require=['py', 'pytest', 'requests'],
     extras_require={
         'fasttext': ['fasttext==0.9.2'],
         'voikko': ['voikko'],
-        'nn': ['tensorflow-cpu==2.5.0', 'lmdb==1.2.1'],
-        'omikuji': ['omikuji==0.3.*'],
+        'nn': ['tensorflow-cpu==2.7.0', 'lmdb==1.3.0'],
+        'omikuji': ['omikuji==0.4.*'],
         'yake': ['yake==0.4.5'],
         'pycld3': ['pycld3'],
         'dev': [