Merge branch 'master' into feature/testing-release-0.1.0

Author: tdesfont

code-env/python/spec/requirements.txt

@@ -1,4 +1,4 @@
-annoy==1.17.0
+annoy==1.16.3
 faiss-cpu==1.6.1
 simplejson==3.17.0
-tqdm==4.51.0
+tqdm==4.51.0

plugin.json

@@ -15,4 +15,4 @@
             "Recommender System"
         ]
     }
-}
+}

python-lib/data_loader.py

@@ -49,7 +49,7 @@ def convert_df_to_vectors(self, df: pd.DataFrame, verbose: bool = True) -> Tuple
                 + f"{len(self.feature_columns)} column(s) into vector format..."
             )
         vector_ids = df[self.unique_id_column].values
-        vectors = np.empty(shape=(len(df.index), self.MAX_VECTOR_LENGTH))
+        vectors = np.empty(shape=(len(df.index), self.MAX_VECTOR_LENGTH))  # pre-allocate empty array of fixed size
         i = 0
         for column in self.feature_columns:
             column_is_vector = df[column].dtype == "object" and df[column].str.startswith("[").all()

python-lib/dku_param_loading.py

@@ -70,7 +70,7 @@ def load_input_output_params(recipe_id: RecipeID) -> Dict:
 
 
 def load_indexing_recipe_params() -> Dict:
-    """Load and validate parameters of the Nearest Neighbor Indexing recipe
+    """Load and validate parameters of the Build Nearest Neighbor Search index recipe
 
     Returns:
         Dictionary of parameter names (key) and values
@@ -79,7 +79,7 @@ def load_indexing_recipe_params() -> Dict:
         PluginParamValidationError: If a parameter is not valid
 
     """
-    logging.info("Validating Nearest Neighbor Indexing parameters...")
+    logging.info("Validating Build Nearest Neighbor Search index recipe parameters...")
     input_output_params = load_input_output_params(RecipeID.SIMILARITY_SEARCH_INDEX)
     # Recipe modeling parameters
     modeling_params = {}
@@ -111,7 +111,7 @@ def load_indexing_recipe_params() -> Dict:
 
 
 def load_search_recipe_params() -> Dict:
-    """Load and validate parameters of the Nearest Neighbor Search recipe
+    """Load and validate parameters of the Find Nearest Neighbors recipe
 
     Returns:
         Dictionary of parameter names (key) and values
@@ -120,7 +120,7 @@ def load_search_recipe_params() -> Dict:
         PluginParamValidationError: If a parameter is not valid
 
     """
-    logging.info("Validating Nearest Neighbor Search parameters...")
+    logging.info("Validating Find Nearest Neighbors recipe parameters...")
     input_output_params = load_input_output_params(RecipeID.SIMILARITY_SEARCH_QUERY)
     # Recipe lookup parameters
     lookup_params = {}

python-lib/nearest_neighbor/annoy.py

@@ -1,6 +1,8 @@
 # -*- coding: utf-8 -*-
 """Module for the Annoy Nearest Neighbor Search algorithm"""
 
+import logging
+
 import numpy as np
 from typing import AnyStr, Dict, List, Tuple
 
@@ -37,6 +39,7 @@ def build_save_index(self, vectors: np.array, index_path: AnyStr) -> None:
         for i, vector in enumerate(tqdm(vectors, mininterval=1.0)):
             self.index.add_item(i, vector.tolist())
         self.index.build(n_trees=self.annoy_num_trees)
+        logging.info(f"Index file path: {index_path}")
 
     @time_logging(log_message="Loading pre-computed index")
     def load_index(self, file_path: AnyStr) -> None:

python-lib/nearest_neighbor/faiss.py

@@ -1,6 +1,8 @@
 # -*- coding: utf-8 -*-
 """Module for the Faiss Nearest Neighbor Search algorithm"""
 
+import logging
+
 import numpy as np
 from typing import AnyStr, Dict, List, Tuple
 
@@ -44,6 +46,7 @@ def build_save_index(self, vectors: np.array, index_path: AnyStr) -> None:
         else:
             raise NotImplementedError("Faiss training methods not implemented")
         faiss.write_index(self.index, index_path)
+        logging.info(f"Index file path: {index_path}")
 
     @time_logging(log_message="Loading pre-computed index")
     def load_index(self, file_path: AnyStr) -> None: