validation of document_ids efficiency

docs/conf.py

@@ -29,7 +29,7 @@
 author = 'Dimo Angelov'
 
 # The full version, including alpha/beta/rc tags
-release = '1.0.21'
+release = '1.0.22'
 
 
 # -- General configuration ---------------------------------------------------

requirements.txt

@@ -1,15 +1,11 @@
-numpy==1.20.0
-numpy
 pandas
 gensim
-pynndescent>=0.4
-umap-learn
-hdbscan
+umap-learn>=0.5.1
+hdbscan>=0.8.27
 wordcloud
 tensorflow
 tensorflow_hub
 tensorflow_text
 torch
 sentence_transformers
 hnswlib
-joblib<1.0.0

setup.py

@@ -6,7 +6,7 @@
 setuptools.setup(
     name="top2vec",
     packages=["top2vec"],
-    version="1.0.21",
+    version="1.0.22",
     author="Dimo Angelov",
     author_email="dimo.angelov@gmail.com",
     description="Top2Vec learns jointly embedded topic, document and word vectors.",
@@ -26,11 +26,8 @@
         "Operating System :: OS Independent",
     ],
     install_requires=[
-        #'numpy == 1.20.0',
-        #'joblib < 1.0.0',
         'pandas',
         'gensim',
-        #'pynndescent >= 0.4',
         'umap-learn >= 0.5.1',
         'hdbscan >= 0.8.27',
         'wordcloud',

top2vec/Top2Vec.py

@@ -884,12 +884,15 @@ def _validate_doc_ids(self, doc_ids, doc_ids_neg):
             doc_ids_neg = list(doc_ids_neg)
 
         doc_ids_all = doc_ids + doc_ids_neg
-        for doc_id in doc_ids_all:
-            if self.document_ids is not None:
-                if doc_id not in self.document_ids:
+
+        if self.document_ids is not None:
+            for doc_id in doc_ids_all:
+                if doc_id not in self.doc_id2index:
                     raise ValueError(f"{doc_id} is not a valid document id.")
-            elif doc_id < 0 or doc_id > len(self.doc_top) - 1:
-                raise ValueError(f"{doc_id} is not a valid document id.")
+        elif min(doc_ids) < 0:
+            raise ValueError(f"{min(doc_ids)} is not a valid document id.")
+        elif max(doc_ids) > len(self.doc_top) - 1:
+            raise ValueError(f"{max(doc_ids)} is not a valid document id.")
 
     def _validate_keywords(self, keywords, keywords_neg):
         if not (isinstance(keywords, list) or isinstance(keywords, np.ndarray)):

top2vec/__init__.py

@@ -1,3 +1,3 @@
 from top2vec.Top2Vec import Top2Vec
 
-__version__ = '1.0.21'
+__version__ = '1.0.22'