Update: Added documentation for API & Cleaning scripts

configs/bce_category.yml

@@ -1,10 +1,10 @@
-name: 'bce-category-embedding-1'
-demographics_file: './data/demographicData/csv/data_world/us_city_demographics.csv'
-demographics_normalized_file: './data/demographicData/csv/census_est_all_data_2019/county_ethnicity_fractions_2019.csv'
-embeddings_file: './data/embeddings/category_bing_embeddings.csv'
+name: "bce-category-embedding-1"
+demographics_file: "./data/demographicData/csv/data_world/us_city_demographics.csv"
+demographics_normalized_file: "./data/demographicData/csv/census_est_all_data_2019/county_ethnicity_fractions_2019.csv"
+embeddings_file: "./data/embeddings/name_category_embeddings.csv"
 lr: 1e-5
 weight_decay: 1e-4 # 1e-5 for MSE, 1e-4 for BCEWithLogits
-loss: 'BCEWithLogits' # 'MSE' or 'BCEWithLogits'
-restaurant_data_source: 'Bing'
+loss: "BCEWithLogits" # 'MSE' or 'BCEWithLogits'
+restaurant_data_source: "Bing"
 hidden_layer_size: 512
-num_epochs: 7000
+num_epochs: 7000

configs/category-name.yml

@@ -0,0 +1,10 @@
+name: "pca-category-names-1"
+demographics_file: "./data/demographicData/csv/data_world/us_city_demographics.csv"
+demographics_normalized_file: "./data/demographicData/csv/census_est_all_data_2019/county_ethnicity_fractions_2019.csv"
+embeddings_file: "./data/embeddings/pca_name_category_embeddings.csv"
+lr: 1e-5
+weight_decay: 1e-5 # 1e-5 for MSE, 1e-4 for BCEWithLogits
+loss: "MSE" # 'MSE' or 'BCEWithLogits'
+restaurant_data_source: "Bing"
+hidden_layer_size: 512
+num_epochs: 7000

data/bingMaps/bingData.py

@@ -11,36 +11,47 @@
 
 cuisine = "Restaurants"  # define the type of cuisine you're interested in
 
-# read locations from counties_states.csv
-with open("edited_counties.csv", "r") as f:
-    reader = csv.reader(f)
-    locations = list(islice(reader, 1917, None))  # skip to line 449
-
-# for each location
-for location in locations:
-    location = " ".join(location)  # convert list to string
-
-    # construct the URL for the HTTP request
-    url = f"https://atlas.microsoft.com/search/poi/json?subscription-key={api_key}&api-version=1.0&query={cuisine}+in+{location}&limit=100"
-
-    response = requests.get(url)  # make the HTTP request
-
-    data = response.json()  # parse the response
-
-    # prepare data for CSV
-    csv_data = []
-    for item in data.get("results", []):
-        # categories = item["poi"]["categories"]
-        name = item["poi"]["name"]
-        # print(name)
-        # if "restaurant" in categories:  # if "restaurant" is in categories, remove it
-        #     categories.remove("restaurant")
-        if name:  # if not empty
-            csv_data.append([name])
-
-    # write data to CSV file
-    with open(f"nameCategory/{location}.csv", "w", newline="") as file:
-        writer = csv.writer(file)
-        print("Currently creating: ", location)
-        writer.writerow(["Name"])  # write the header
-        writer.writerows(csv_data)  # write the rows
+
+def fetch_restaurant_data():
+    """
+    Fetches restaurant data from Bing Maps API for different locations and saves it in CSV files.
+
+    This function reads locations from a CSV file, constructs the URL for the HTTP request to the Bing Maps API,
+    makes the request, parses the response, and saves the relevant data in CSV files.
+
+    Returns:
+        None
+    """
+
+    # read locations from counties_states.csv
+    with open("edited_counties.csv", "r") as f:
+        reader = csv.reader(f)
+        locations = list(islice(reader, 1917, None))  # skip to line 449
+
+    # for each location
+    for location in locations:
+        location = " ".join(location)  # convert list to string
+
+        # construct the URL for the HTTP request
+        url = f"https://atlas.microsoft.com/search/poi/json?subscription-key={api_key}&api-version=1.0&query={cuisine}+in+{location}&limit=100"
+
+        response = requests.get(url)  # make the HTTP request
+
+        data = response.json()  # parse the response
+
+        # prepare data for CSV
+        csv_data = []
+        for item in data.get("results", []):
+            name = item["poi"]["name"]
+            if name:  # if not empty
+                csv_data.append([name])
+
+        # write data to CSV file
+        with open(f"name_category/{location}.csv", "w", newline="") as file:
+            writer = csv.writer(file)
+            print("Currently creating: ", location)
+            writer.writerow(["Name"])  # write the header
+            writer.writerows(csv_data)  # write the rows
+
+
+fetch_restaurant_data()