Update app.py

app.py

@@ -38,12 +38,24 @@ model.config.eos_token_id = tokenizer.eos_token_id
 model.config.decoder_start_token_id = tokenizer.bos_token_id
 model.config.pad_token_id = tokenizer.pad_token_id
 
+def preprocess_query(query):
+    doc = nlp(query)
+    tokens = set()
+    for token in doc:
+        tokens.add(token.text)
+        tokens.add(token.lemma_)
+        tokens.update(get_synonyms(token.text))
+    st.write(f"Query tokens: {tokens}")  # Debugging line
+    return tokens
+
 def generate_caption(image):
     pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values
     output_ids = model.generate(pixel_values)
     caption = tokenizer.decode(output_ids[0], skip_special_tokens=True)
+    st.write(f"Generated caption: {caption}")  # Debugging line
     return caption
 
+
 def get_synonyms(word):
     synonyms = set()
     for syn in wordnet.synsets(word):
@@ -51,14 +63,6 @@ def get_synonyms(word):
             synonyms.add(lemma.name())
     return synonyms
 
-def preprocess_query(query):
-    doc = nlp(query)
-    tokens = set()
-    for token in doc:
-        tokens.add(token.text)
-        tokens.add(token.lemma_)
-        tokens.update(get_synonyms(token.text))
-    return tokens
 
 def search_captions(query, captions):
     query_tokens = preprocess_query(query)