Update recommendation.py

recommendation.py

@@ -28,25 +28,22 @@ def row_wise_normalize_and_concatenate(array1, array2):
 
     return concatenated_array
 
-result_array = row_wise_normalize_and_concatenate(latent_data, embeddings)
-
-index = Indexer(result_array)
+# result_array = row_wise_normalize_and_concatenate(latent_data, embeddings)
+# index = Indexer(result_array)
+index = Indexer(latent_data)
 
 def get_recommendations(image, title, k):
-  #title = [dataset[product_id]['title']]
   title_embeds = model.encode([title], normalize_embeddings=True)
   image = transforms.ToTensor()(image.convert("L"))
   image_embeds =  encoder(image).detach().numpy()
   image_embeds = image_embeds / np.linalg.norm(image_embeds, axis=1, keepdims=True)
   final_embeds = np.concatenate((image_embeds,title_embeds), axis=1)
 
-  candidates = index.topk(final_embeds,k=k)
-  # title_candidates = text_embedding_index.topk(title_embeds, k=k)
+  # candidates = index.topk(final_embeds,k=k)
+  candidates = index.topk(image_embeds,k=k)
   final_candidates = []
   final_candidates.append(list(candidates[0]))
-  #final_candidates.append(list(title_candidates[0]))
   final_candidates = sum(final_candidates,[])
-  # final_candidates = list(set(final_candidates))
   results_dict = {"image":[], "title":[]}
   for candidate in final_candidates:
     results_dict['image'].append(dataset['image'][candidate])