Staging - Bug Fixes

README.md

@@ -1,4 +1,4 @@
-![GitHub Workflow Status (branch)](https://img.shields.io/github/workflow/status/forecastlab/forecast_dash/Build%20and%20deploy%20Python%20app%20-%20staging-forecast-lab/staging?label=Build) ![BlackResults)](https://img.shields.io/github/workflow/status/forecastlab/forecast_dash/CI/master?label=Clean%20Code) ![Check Data Sources](https://img.shields.io/github/workflow/status/forecastlab/forecast_dash/Check%20Data%20Sources/staging?label=Data%20Sources) ![GitHub commit activity](https://img.shields.io/github/commit-activity/y/forecastlab/forecast_dash) ![GitHub contributors](https://img.shields.io/github/contributors/forecastlab/forecast_dash)
+![GitHub Workflow Status (branch)](https://img.shields.io/github/actions/workflow/status/forecastlab/forecast_dash/Docker_build_check.yml?branch=staging&label=Build) ![BlackResults)](https://img.shields.io/github/actions/workflow/status/forecastlab/forecast_dash/ci.yml?branch=staging&label=Clean%20Code) ![Check Data Sources](https://img.shields.io/github/actions/workflow/status/forecastlab/forecast_dash/datasources.yml?branch=staging&label=Data%20Sources) ![GitHub commit activity](https://img.shields.io/github/commit-activity/y/forecastlab/forecast_dash) ![GitHub contributors](https://img.shields.io/github/contributors/forecastlab/forecast_dash)
 
 # USYD Forecasting Lab
 

dash/common.py

@@ -399,7 +399,7 @@ def select_best_model(data_dict, CV_score_function="MSE"):
 def get_thumbnail_figure(data_dict, lg=12):
     if data_dict == None:
         return {}
-        
+
     watermark_config = (
         watermark_information()
     )  # Grab the watermark text and fontsize information
@@ -694,10 +694,16 @@ def get_leaderboard_df(series_list, CV_score_function="MSE"):
     chosen_methods = []
 
     for series_title, forecast_data in forecast_series_dicts.items():
-        model_name = select_best_model(
-            forecast_data, CV_score_function=CV_score_function
-        )
-        chosen_methods.append(model_name)
+        try:
+            model_name = select_best_model(
+                forecast_data, CV_score_function=CV_score_function
+            )
+            chosen_methods.append(model_name)
+        except Exception as e:
+            print(
+                f"{[series_title]} had the following error: {e}"
+            )  # Usually missing forecasts
+            pass
 
     stats_raw = pd.DataFrame({"Method": chosen_methods})
 

dash/pages/home.py

@@ -110,7 +110,8 @@ def _featured_latest_news(feature_series_title):
                 dbc.Col(
                     [
                         html.H2(
-                            "US " + "inflation shows strong signs of stabilisation".title(),
+                            "US "
+                            + "inflation shows strong signs of stabilisation".title(),
                         ),
                         html.A(
                             html.H4("View all US forecasts"),
@@ -127,7 +128,9 @@ def _featured_latest_news(feature_series_title):
                             [
                                 dcc.Graph(
                                     figure=get_thumbnail_figure(
-                                        get_forecast_data("US Personal Consumption Expenditures: Chain-type Price Index (% Change, 1 Year)"),
+                                        get_forecast_data(
+                                            "US Personal Consumption Expenditures: Chain-type Price Index (% Change, 1 Year)"
+                                        ),
                                         lg=8,
                                     ),
                                     config={"displayModeBar": False},
@@ -223,7 +226,9 @@ def _uk_snapshot():
                             [
                                 dcc.Graph(
                                     figure=get_thumbnail_figure(
-                                        get_forecast_data("UK Inflation (RPI)"),
+                                        get_forecast_data(
+                                            "UK Inflation (RPI)"
+                                        ),
                                         lg=8,
                                     ),
                                     config={"displayModeBar": False},
@@ -276,14 +281,17 @@ def _link_search():
         ]
     )
 
+
 def _au_snapshot():
     return [
         dbc.Row(
             [
                 dbc.Col(
                     [
                         html.H2(
-                            "Have " + "AU" +" house prices reached the bottom?".title()
+                            "Have "
+                            + "AU"
+                            + " house prices reached the bottom?".title()
                         ),
                         html.A(
                             html.H4("View all Australian forecasts"),
@@ -300,7 +308,9 @@ def _au_snapshot():
                             [
                                 dcc.Graph(
                                     figure=get_thumbnail_figure(
-                                        get_forecast_data("Australian (Sydney) Change in House Prices"),
+                                        get_forecast_data(
+                                            "Australian (Sydney) Change in House Prices"
+                                        ),
                                         lg=8,
                                     ),
                                     config={"displayModeBar": False},

dash/pages/leaderboard.py

@@ -79,6 +79,9 @@ def update_leaderboard_df(CV_score):
     construct the best model leaderboard based upon user selected scoring function
     """
     # Build leaderboard with chosen CV scoring function
+    data_sources_json_file = open("../shared_config/data_sources.json")
+    series_list = json.load(data_sources_json_file)
+
     counts = get_leaderboard_df(series_list, CV_score)
 
     win_proportion = (

dash/pages/search.py

@@ -391,9 +391,7 @@ def filter_results(**kwargs):
 
         for item_title in unique_series_titles[0:show_num]:
             try:
-                series_data = get_forecast_data(
-                    item_title
-                )
+                series_data = get_forecast_data(item_title)
                 if series_data == None:
                     raise FileNotFoundError