Upload 2 files

app

@@ -0,0 +1,50 @@
+
+import gradio as gr
+import pandas as pd
+import joblib
+
+# Load model
+model = joblib.load("demand_forecasting_model.pkl")
+
+# Ordinal encoder (harus sama urutannya dengan saat training)
+from sklearn.preprocessing import OrdinalEncoder
+encoder = OrdinalEncoder()
+encoder.fit([
+    ['A', 'New York'], ['B', 'Los Angeles'], ['C', 'Chicago'], ['D', 'Houston'], ['E', 'Phoenix']
+])  # dummy untuk encode ulang di input manual
+
+# Fungsi prediksi untuk input manual
+def predict_demand(product_type, price, stock, location, lead_time):
+    data = pd.DataFrame([[product_type, price, stock, location, lead_time]],
+                        columns=["Product type", "Price", "Stock levels", "Location", "Lead time"])
+    data[["Product type", "Location"]] = encoder.transform(data[["Product type", "Location"]])
+    prediction = model.predict(data)
+    return round(prediction[0], 2)
+
+# Fungsi prediksi dari file CSV
+def predict_from_csv(file):
+    df = pd.read_csv(file.name)
+    df[["Product type", "Location"]] = encoder.transform(df[["Product type", "Location"]])
+    preds = model.predict(df[["Product type", "Price", "Stock levels", "Location", "Lead time"]])
+    df["Predicted Demand"] = preds
+    return df
+
+# Gradio antarmuka
+with gr.Blocks() as demo:
+    gr.Markdown("# 📦 Demand Forecasting App")
+    with gr.Tab("Input Manual"):
+        product_type = gr.Dropdown(["A", "B", "C", "D", "E"], label="Product Type")
+        price = gr.Number(label="Price")
+        stock = gr.Number(label="Stock Levels")
+        location = gr.Dropdown(["New York", "Los Angeles", "Chicago", "Houston", "Phoenix"], label="Location")
+        lead_time = gr.Number(label="Lead Time (days)")
+        predict_btn = gr.Button("Predict Demand")
+        output = gr.Textbox(label="Predicted Demand")
+        predict_btn.click(fn=predict_demand, inputs=[product_type, price, stock, location, lead_time], outputs=output)
+
+    with gr.Tab("Upload CSV"):
+        csv_input = gr.File(label="Upload CSV", file_types=[".csv"])
+        csv_output = gr.Dataframe(label="CSV with Predictions")
+        csv_input.change(fn=predict_from_csv, inputs=csv_input, outputs=csv_output)
+
+demo.launch()

demand_forecasting_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa7db6ff9a89d059e1db0b39cbee5f9ae7865d20d18cd4d6adf4ee3a18556b3d
+size 631409