Merge pull request #835 from readthedocs/embedding-fixes

A couple small bug fixes for embeddings work
readthedocs · Feb 22, 2024 · 6aeb961 · 6aeb961
2 parents cbbd674 + 6b9fd08
commit 6aeb961
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 3 deletions.
diff --git a/adserver/analyzer/management/commands/runmodel.py b/adserver/analyzer/management/commands/runmodel.py
@@ -8,7 +8,6 @@
 
 
 class Command(BaseCommand):
-
     """Run the ML model on the specified URLs."""
 
     help = "Run the ML model on the specified URLs. Results are not stored to the database."
@@ -39,13 +38,17 @@ def handle_url(self, url):
         keywords = []
         embeddings = []
         for backend in get_url_analyzer_backends():
+
             backend_instance = backend(url)
             response = backend_instance.fetch()
+            if not response:
+                continue
+
             analyzed_keywords = backend_instance.analyze(response)
             self.stdout.write(
                 _("Keywords from '%s': %s") % (backend.__name__, analyzed_keywords)
             )
-            analyzed_embedding = backend.embedding(response)
+            analyzed_embedding = backend_instance.embedding(response)
             self.stdout.write(
                 _("Embeddings from '%s': %s") % (backend.__name__, analyzed_embedding)
             )

diff --git a/adserver/analyzer/tasks.py b/adserver/analyzer/tasks.py
@@ -66,7 +66,10 @@ def analyze_url(url, publisher_slug, force=False):
         log.debug("Keywords from '%s': %s", backend.__name__, analyzed_keywords)
 
         analyzed_embedding = backend_instance.embedding(response)  # Can be None
-        log.debug("Embedding from '%s': %s", backend.__name__, len(analyzed_embedding))
+        if analyzed_embedding:
+            log.debug(
+                "Embedding from '%s': %s", backend.__name__, len(analyzed_embedding)
+            )
 
         if analyzed_keywords:
             for kw in analyzed_keywords: