Upload 3 files

Dokcerfile

@@ -14,4 +14,4 @@ RUN pip install --no-cache-dir --upgrade -r requirements.txt
 # - `-w 4`: Uses 4 worker processes for handling requests
 # - `-b 0.0.0.0:7860`: Binds the server to port 7860 on all network interfaces
 # - `app:app`: Runs the Flask app (assuming `app.py` contains the Flask instance named `app`)
-CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:7860",  "main:app"]
+CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:7860",  "app:app"]

app (1).py

@@ -0,0 +1,66 @@
+
+import streamlit as st
+import pandas as pd
+import joblib
+import matplotlib.pyplot as plt
+
+# Load model
+@st.cache_resource
+def load_model():
+    return joblib.load("best_dealer_forecast_model.joblib")
+
+model = load_model()
+
+# Load history
+history = pd.read_csv("dealer_sales_history.csv", parse_dates=["date"])
+
+st.title("🚗 Dealer Variant Sales Forecast")
+
+dealer_id = st.number_input("Enter Dealer ID", min_value=1, step=1)
+variant = st.text_input("Enter Model Variant (e.g., S(O))")
+months = st.slider("Forecast Horizon (months)", 1, 12, 5)
+
+if st.button("Predict"):
+    df = history[
+        (history["dealer_id"] == dealer_id) & 
+        (history["model_variant"] == variant)
+    ].copy().sort_values("date").reset_index(drop=True)
+
+    if df.empty:
+        st.error("No history available for this Dealer + Variant")
+    else:
+        preds = []
+        for i in range(months):
+            df["sale_volume_lag_1"] = df["sale_volume"].shift(1)
+            df["sale_volume_lag_2"] = df["sale_volume"].shift(2)
+            df["sale_volume_lag_3"] = df["sale_volume"].shift(3)
+            df["sale_volume_lag_6"] = df["sale_volume"].shift(6)
+            df["sales_roll_mean_3"] = df["sale_volume"].rolling(3).mean()
+            df["sales_roll_std_3"] = df["sale_volume"].rolling(3).std()
+
+            latest = df.iloc[-1].copy()
+            features = latest[[
+                "dealer_id", "model_variant",
+                "sale_volume_lag_1","sale_volume_lag_2",
+                "sale_volume_lag_3","sale_volume_lag_6",
+                "sales_roll_mean_3","sales_roll_std_3"
+            ]]
+            pred = model.predict(pd.DataFrame([features]))[0]
+            next_date = df["date"].max() + pd.DateOffset(months=1)
+            preds.append({"date": next_date, "predicted_sales": pred})
+
+            df = pd.concat([df, pd.DataFrame({
+                "date": [next_date],
+                "dealer_id": [dealer_id],
+                "model_variant": [variant],
+                "sale_volume": [pred]
+            })], ignore_index=True)
+
+        forecast = pd.DataFrame(preds)
+        st.write("### Forecast Results", forecast)
+
+        fig, ax = plt.subplots()
+        ax.plot(forecast["date"], forecast["predicted_sales"], marker="o", label="Forecast")
+        ax.set_xlabel("Date")
+        ax.set_ylabel("Predicted Sales")
+        st.pyplot(fig)