C-part done

app.py

@@ -12,23 +12,17 @@ from airquality.util import plot_air_quality_forecast
 # -------------------------
 # HOPSWORKS LOGIN + MODEL LOAD
 # -------------------------
-def load_resources():
+def load_resources_v1():
     project = hopsworks.login()
     fs = project.get_feature_store()
 
-    # Load Feature View
-    fv = fs.get_feature_view(
-        name="air_quality_fv",
-        version=2
-    )
+    fv = fs.get_feature_view("air_quality_fv", version=1)
     fv.init_batch_scoring(1)
 
-    # Load Weather Feature Group (for future predictions)
     weather_fg = fs.get_feature_group("weather", 1)
 
-    # Load Model from Registry
     mr = project.get_model_registry()
-    model_obj = mr.get_model("air_quality_xgboost_model", version=2)
+    model_obj = mr.get_model("air_quality_xgboost_model", version=1)
     model_dir = model_obj.download()
 
     model = XGBRegressor()
@@ -37,103 +31,170 @@ def load_resources():
     return model, fv, weather_fg, project
 
 
-model, feature_view, weather_fg, project = load_resources()
+def load_resources_v2():
+    project = hopsworks.login()
+    fs = project.get_feature_store()
+
+    fv = fs.get_feature_view("air_quality_fv", version=2)
+    fv.init_batch_scoring(1)
+
+    weather_fg = fs.get_feature_group("weather", 1)
+
+    mr = project.get_model_registry()
+    model_obj = mr.get_model("air_quality_xgboost_model", version=2)
+    model_dir = model_obj.download()
+
+    model = XGBRegressor()
+    model.load_model(model_dir + "/model.json")
+
+    return model, fv, weather_fg, project
+
 
 
 # -------------------------
 # FORECAST LOGIC (NEXT 7 DAYS)
 # -------------------------
-def generate_forecast(days):
-    today = datetime.utcnow().date()
-    future_dates = [(today + timedelta(days=i)).strftime("%Y-%m-%d") for i in range(1, days + 1)]
+def generate_forecast_v1(days):
+    model, fv, weather_fg, project = load_resources_v1()
 
+    today = datetime.utcnow().date()
     df_future = weather_fg.read()
     df_future["date"] = pd.to_datetime(df_future["date"]).dt.date
 
+    predictions = []
+    for offset in range(1, days + 1):
+        target_date = today + timedelta(days=offset)
+        row = df_future[df_future["date"] == target_date]
+
+        if len(row) == 0:
+            continue
+
+        X = row[["temperature_2m_mean", "precipitation_sum",
+                 "wind_speed_10m_max", "wind_direction_10m_dominant"]]
+
+        pred = float(model.predict(X)[0])
+
+        predictions.append({"date": target_date, "predicted_pm25": pred})
+
+    df_preds = pd.DataFrame(predictions)
+
+    tmp_path = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
+    plot_air_quality_forecast("linkoping", "hamngatan-10", df_preds, tmp_path)
+
+
+def generate_forecast_v2(days):
+    model, fv, weather_fg, project = load_resources_v2()
+
+    today = datetime.utcnow().date()
+    df_future = weather_fg.read().sort_values("date")
+    df_future["date"] = pd.to_datetime(df_future["date"]).dt.date
+
+    # Load real PM2.5 history
+    aq_fg = project.get_feature_store().get_feature_group("air_quality", version=1)
+    hist_pm25 = aq_fg.read().sort_values("date")
+    pm25_history = list(hist_pm25["pm25"].values[-3:])
+
     preds = []
-    for d in future_dates:
-        dt = datetime.strptime(d, "%Y-%m-%d").date()
-        row = df_future[df_future["date"] == dt]
+    for offset in range(1, days + 1):
 
+        target_date = today + timedelta(days=offset)
+        row = df_future[df_future["date"] == target_date]
         if len(row) == 0:
             continue
 
-        input_features = row.drop(columns=["date", "city"])
-        pm25_pred = model.predict(input_features)[0]
+        lag1, lag2, lag3 = pm25_history[-1], pm25_history[-2], pm25_history[-3]
+        roll_mean = np.mean(pm25_history[-3:])
+        roll_std = np.std(pm25_history[-3:])
+
+        X = pd.DataFrame({
+            "temperature_2m_mean": [row.iloc[0]["temperature_2m_mean"]],
+            "precipitation_sum": [row.iloc[0]["precipitation_sum"]],
+            "wind_speed_10m_max": [row.iloc[0]["wind_speed_10m_max"]],
+            "wind_direction_10m_dominant": [row.iloc[0]["wind_direction_10m_dominant"]],
+            "pm25_lag1": [lag1],
+            "pm25_lag2": [lag2],
+            "pm25_lag3": [lag3],
+            "pm25_roll3_mean": [roll_mean],
+            "pm25_roll3_std": [roll_std],
+        })
 
-        preds.append({"date": d, "predicted_pm25": pm25_pred})
+        pred = float(model.predict(X)[0])
+        preds.append({"date": target_date, "predicted_pm25": pred})
 
-    if len(preds) == 0:
-        return None
+        pm25_history.append(pred)
 
     df_preds = pd.DataFrame(preds)
 
-    # Save plot
     tmp_path = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
-    plot_air_quality_forecast("linkoping", "hamngatan-10", df_preds, tmp_path, hindcast=False)
+    plot_air_quality_forecast("linkoping", "hamngatan-10", df_preds, tmp_path)
 
     return tmp_path
 
-
 # -------------------------
 # HINDCAST LOGIC (LAST 7 DAYS)
 # -------------------------
-def generate_hindcast(days):
-    # Read actual + predicted from Feature View
-    features_df, labels_df = feature_view.training_data(
+def generate_hindcast_v1(days):
+    model, fv, weather_fg, project = load_resources_v1()
+
+    features_df, labels_df = fv.training_data(
         start_time=datetime.utcnow().date() - timedelta(days=days),
         end_time=datetime.utcnow().date(),
         statistics_config=False
     )
 
     features_df["date"] = pd.to_datetime(features_df["date"]).dt.date
-    labels_df["pm25"] = labels_df["pm25"]
 
-    df = features_df.copy()
-    df["pm25"] = labels_df["pm25"]
+    X = features_df[["temperature_2m_mean", "precipitation_sum",
+                     "wind_speed_10m_max", "wind_direction_10m_dominant"]]
 
-    # Predict using model
-    df["predicted_pm25"] = model.predict(
-        df[["temperature_2m_mean", "precipitation_sum", "wind_speed_10m_max", "wind_direction_10m_dominant"]]
+    features_df["predicted_pm25"] = model.predict(X)
+
+    tmp_path = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
+    plot_air_quality_forecast("linkoping", "hamngatan-10", features_df, tmp_path, hindcast=True)
+
+    return tmp_path
+
+
+def generate_hindcast_v2(days):
+    model, fv, weather_fg, project = load_resources_v2()
+
+    features_df, labels_df = fv.training_data(
+        start_time=datetime.utcnow().date() - timedelta(days=days),
+        end_time=datetime.utcnow().date(),
+        statistics_config=False
     )
 
-    # Keep last 7 days
-    df = df.sort_values("date").tail(days)
+    features_df["date"] = pd.to_datetime(features_df["date"]).dt.date
+
+    # Full lag-feature dataset already inside feature view v2
+    X = features_df.drop(columns=["date"])
+
+    features_df["predicted_pm25"] = model.predict(X)
 
-    # Save plot
     tmp_path = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
-    plot_air_quality_forecast("linkoping", "hamngatan-10", df, tmp_path, hindcast=True)
+    plot_air_quality_forecast("linkoping", "hamngatan-10", features_df, tmp_path, hindcast=True)
 
     return tmp_path
 
 
-# -------------------------
-# GRADIO UI
-# -------------------------
-def run_dashboard(forecast_days, hindcast_days):
-    forecast_plot = generate_forecast(forecast_days)
-    hindcast_plot = generate_hindcast(hindcast_days)
-    return forecast_plot, hindcast_plot
-
 
 with gr.Blocks() as iface:
-    
+
     gr.Markdown("# Air Quality Forecast Dashboard")
-    gr.Markdown("Forecast and hindcast PM2.5 using XGBoost + Hopsworks")
-    
+    gr.Markdown("Choose between Model v1 (no lags) and Model v2 (lag-aware).")
+
     with gr.Row():
-        forecast_days = gr.Slider(3, 10, value=7, step=1, label="Forecast Days (future)")
-        hindcast_days = gr.Slider(3, 10, value=7, step=1, label="Hindcast Days (past)")
-    
-    generate_button = gr.Button("Generate")
-    
-    forecast_img = gr.Image(label="PM2.5 Forecast")
-    hindcast_img = gr.Image(label="PM2.5 Hindcast")
-
-    generate_button.click(
-        fn=run_dashboard,
-        inputs=[forecast_days, hindcast_days],
-        outputs=[forecast_img, hindcast_img]
-    )
+        forecast_days_v1 = gr.Slider(3, 10, value=7, label="Forecast Days (v1)")
+        forecast_days_v2 = gr.Slider(3, 10, value=7, label="Forecast Days (v2)")
+
+    with gr.Row():
+        btn_v1 = gr.Button("Generate Forecast (Model v1)")
+        btn_v2 = gr.Button("Generate Forecast (Model v2)")
+
+    out_v1 = gr.Image(label="Model v1 Forecast")
+    out_v2 = gr.Image(label="Model v2 Forecast")
+
+    btn_v1.click(fn=generate_forecast_v1, inputs=forecast_days_v1, outputs=out_v1)
+    btn_v2.click(fn=generate_forecast_v2, inputs=forecast_days_v2, outputs=out_v2)
 
 iface.launch()