Upload 4 files

.gitattributes

@@ -16,3 +16,4 @@ best_rf_model.joblib filter=lfs diff=lfs merge=lfs -text
 best_xgb_model_20260222_202200.joblib filter=lfs diff=lfs merge=lfs -text
 Random_Forest_pipeline_for_hf.joblib filter=lfs diff=lfs merge=lfs -text
 model.joblib filter=lfs diff=lfs merge=lfs -text
+final_winner_deployment_bundle.joblib filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,15 @@
+FROM python:3.10-slim
+
+RUN apt-get update && apt-get install -y --no-install-recommends build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+COPY requirements.txt /app/requirements.txt
+RUN pip install --no-cache-dir -r /app/requirements.txt
+
+COPY . /app
+
+EXPOSE 7860
+
+CMD ["python", "app.py"]

app.py

@@ -0,0 +1,103 @@
+import os
+import joblib
+import numpy as np
+import pandas as pd
+import gradio as gr
+
+BUNDLE_PATH = "final_winner_deployment_bundle.joblib"
+
+def load_bundle():
+    if not os.path.exists(BUNDLE_PATH):
+        raise FileNotFoundError(
+            f"Missing '{BUNDLE_PATH}'. Upload it to the Space repo root."
+        )
+
+    bundle = joblib.load(BUNDLE_PATH)
+
+    model = bundle["model"]
+    preproc_bundle = bundle["preprocessor"]
+    metadata = bundle.get("metadata", {})
+    eval_space = bundle.get("evaluation_space", "raw")
+    winner_name = bundle.get("winner_name", "Unknown")
+
+    preprocessor = preproc_bundle["pipeline"]
+    drop_cols = preproc_bundle.get("drop_cols", [])
+    target = metadata.get("target", "Product_Store_Sales_Total")
+
+    return model, preprocessor, drop_cols, target, eval_space, winner_name
+
+
+MODEL, PREPROC, DROP_COLS, TARGET, EVAL_SPACE, WINNER = load_bundle()
+
+
+def _prepare_features(df):
+    df = df.copy()
+    df = df.drop(columns=DROP_COLS, errors="ignore")
+    if TARGET in df.columns:
+        df = df.drop(columns=[TARGET], errors="ignore")
+    return df
+
+
+def predict_from_df(df):
+    X_df = _prepare_features(df)
+    X = PREPROC.transform(X_df)
+    pred = MODEL.predict(X)
+
+    if EVAL_SPACE == "log1p":
+        pred = np.expm1(pred)
+
+    out = df.copy()
+    out["prediction"] = pred
+    return out
+
+
+def predict_csv(file):
+    if file is None:
+        return None, "Please upload a CSV file."
+
+    df = pd.read_csv(file.name)
+    out = predict_from_df(df)
+
+    out_path = "predictions.csv"
+    out.to_csv(out_path, index=False)
+
+    msg = f"Predicted {len(df)} rows using {WINNER}."
+    return out_path, msg
+
+
+def predict_single_row(json_text):
+    if not json_text.strip():
+        return "Please paste a JSON object."
+
+    try:
+        row = pd.read_json(json_text, typ="series")
+    except Exception as e:
+        return f"Invalid JSON: {e}"
+
+    df = pd.DataFrame([row.to_dict()])
+    out = predict_from_df(df)
+    return f"{float(out['prediction'].iloc[0]):,.4f}"
+
+
+with gr.Blocks(title="SuperKart Sales Predictor") as demo:
+    gr.Markdown(f"""
+# SuperKart Sales Prediction
+
+Winner model: {WINNER}
+Prediction space: {EVAL_SPACE}
+""")
+
+    with gr.Tab("CSV Upload"):
+        file_in = gr.File(label="Upload CSV")
+        file_out = gr.File(label="Download predictions.csv")
+        status = gr.Textbox(label="Status")
+        btn = gr.Button("Run Predictions")
+        btn.click(predict_csv, inputs=file_in, outputs=[file_out, status])
+
+    with gr.Tab("Single Row (JSON)"):
+        json_in = gr.Textbox(label="Row JSON", lines=8)
+        pred_out = gr.Textbox(label="Prediction")
+        btn2 = gr.Button("Predict")
+        btn2.click(predict_single_row, inputs=json_in, outputs=pred_out)
+
+demo.launch(server_name="0.0.0.0", server_port=7860)

final_winner_deployment_bundle.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47811420ad804cdc5935dccddfc0e2a79629fad3cb65f9dcf9dbcac5023994b8
+size 79687628

requirements.txt

@@ -0,0 +1,6 @@
+gradio==4.44.1
+pandas==2.2.2
+numpy==1.26.4
+scikit-learn==1.4.2
+joblib==1.4.2
+xgboost==2.0.3