Update embeddings

data/categories.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4dddf959c29d7af260e7eeacd5a6dbd1c249b54fc492b21c6489dcf68f970402
+oid sha256:d6521f92d3db3b7b34fb971dd902fe707470a5e07867d13737cb148ba597bd4b
 size 1134750

data/dishes.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e8436a53e09bff69275cd04b47a46c55a1e4e6e4f8ae1f856026e02060bffd9
+oid sha256:765848ef3ced029b3c3178d5bd6d95fee83e26d119261940b3b3e76e97f26897
 size 7290051

data/restaurants.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d379bf61329a9e615a2cdfe082e99df5fe63f3278876e5addcf18e6e059c4ec
-size 1863120
+oid sha256:d3ca5f59f7088bd0cb369d3a33013b1a40dbcf269b6b62c0b43ed5ce87c60b1e
+size 1858087

scripts/create_embeddings.py

@@ -99,13 +99,22 @@ def main():
             cs = literal_eval(row['categories'])
 
             location = get_district_name(row['map_latitude'], row['map_longitude'])
+            price = int(row['price'])
+            if price < 100:
+                price_bucket = 'cheap'
+            elif 300 > price >= 100:
+                price_bucket = 'moderate'
+            elif price >= 300:
+                price_bucket = 'expensive'
+
+            extra_data = ' Romantic Dining.' if 'Romantic Dining' in cs else ""
 
             text = f"""\
 Name: {name}
-Intro: {row['intro']}
-Categories: {", ".join(cs)}
+Intro: {row['intro']}{extra_data}
 Dishes: {", ".join(ds)}
-Location: {", ".join(location)}\
+Location: {", ".join(location)}
+Price: {price_bucket}\
 """
 
             r = RestaurantDescription(
@@ -115,7 +124,7 @@ Location: {", ".join(location)}\
                 name=name,
                 name_alt=name_alt,
                 intro=row['intro'],
-                price=int(row['price']),
+                price=price,
                 rating=calculate_rating(row['score_cry'], row['score_o_k'], row['score_smile']),
                 categories=cs,
                 dishes=ds,
@@ -131,8 +140,8 @@ Location: {", ".join(location)}\
     dish_list = DocList[Dish](dishes.values())
     category_list = DocList[Category](categories.values())
 
-    # import IPython
-    # IPython.embed()
+    import IPython
+    IPython.embed()
 
     embedding_settings = AzureOpenaiEmbeddings.load_from_env()