rvert

app.py

@@ -4,7 +4,7 @@ import plotly.graph_objects as go
 import streamlit as st
 
 from src.helper_functions import custom_metric_box, pollution_box
-from src.predict import get_data_and_predictions, load_models
+from src.predict import get_data_and_predictions
 
 st.set_page_config(
     page_title="Utrecht Pollution Dashboard",
@@ -15,8 +15,7 @@ st.set_page_config(
 
 alt.themes.enable("dark")
 
-model_NO2, model_O3 = load_models()
-week_data, predictions_O3, predictions_NO2 = get_data_and_predictions(model_NO2, model_O3)
+week_data, predictions_O3, predictions_NO2 = get_data_and_predictions()
 
 today = week_data.iloc[-1]
 previous_day = week_data.iloc[-2]

predictions_history.csv

@@ -5,3 +5,9 @@ O3,2024-10-24,2024-10-26,16.000984317626852
 NO2,2024-10-24,2024-10-26,25.760307451092384
 O3,2024-10-24,2024-10-27,19.64377495640328
 NO2,2024-10-24,2024-10-27,31.210576791105115
+O3,2024-10-24,2024-10-25,10.33808859423279
+NO2,2024-10-24,2024-10-25,25.68519991558237
+O3,2024-10-24,2024-10-26,16.000984317626852
+NO2,2024-10-24,2024-10-26,25.760307451092384
+O3,2024-10-24,2024-10-27,19.64377495640328
+NO2,2024-10-24,2024-10-27,31.210576791105115

src/predict.py

@@ -3,53 +3,44 @@ from datetime import date, timedelta
 
 import joblib
 import pandas as pd
-import streamlit as st
 from dotenv import load_dotenv
 from huggingface_hub import hf_hub_download, login
 
 from src.data_api_calls import get_combined_data
 from src.features_pipeline import create_features
 
-load_dotenv()
 
-
-st.cache_resource()
-def load_models():
+def load_model(particle):
+    load_dotenv()
     login(token=os.getenv("HUGGINGFACE_DOWNLOAD_TOKEN"))
 
-    model_path = hf_hub_download(
-        repo_id="elisaklunder/Utrecht-O3-Forecasting-Model", filename="O3_svr_model.pkl"
-    )
-    model_NO2 = joblib.load(model_path)
-
-    model_path = hf_hub_download(
-        repo_id="elisaklunder/Utrecht-NO2-Forecasting-Model",
-        filename="NO2_svr_model.pkl",
-    )
-    model_O3 = joblib.load(model_path)
-    return model_NO2, model_O3
+    repo_id = f"elisaklunder/Utrecht-{particle}-Forecasting-Model"
+    if particle == "O3":
+        file_name = "O3_svr_model.pkl"
+    elif particle == "NO2":
+        file_name = "NO2_svr_model.pkl"
 
+    model_path = hf_hub_download(repo_id=repo_id, filename=file_name)
+    model = joblib.load(model_path)
+    return model
 
-def run_models(model_NO2, model_O3, data):
-    input_data = create_features(data=data, target_particle="NO2")
-    prediction = model_NO2.predict(input_data)
-    target_scaler = joblib.load("scalers/target_scaler_NO2.joblib")
-    prediction_NO2 = target_scaler.inverse_transform(prediction)
 
-    input_data = create_features(data=data, target_particle="O3")
-    prediction = model_O3.predict(input_data)
-    target_scaler = joblib.load("scalers/target_scaler_O3.joblib")
-    prediction_O3 = target_scaler.inverse_transform(prediction)
+def run_model(particle, data):
+    input_data = create_features(data=data, target_particle=particle)
+    model = load_model(particle)
+    prediction = model.predict(input_data)
+    target_scaler = joblib.load(f"scalers/target_scaler_{particle}.joblib")
+    prediction = target_scaler.inverse_transform(prediction)
+    return prediction
 
-    return prediction_NO2, prediction_O3
 
-st.cache_resource(ttl=300 * 24)
-def get_data_and_predictions(model_NO2, model_O3):
+def get_data_and_predictions():
     PREDICTIONS_FILE = "predictions_history.csv"
 
     week_data = get_combined_data()
 
-    o3_predictions, no2_predictions = run_models(model_NO2, model_O3, week_data)
+    o3_predictions = run_model("O3", data=week_data)
+    no2_predictions = run_model("NO2", data=week_data)
 
     prediction_data = []
     for i in range(3):