Upload folder using huggingface_hub

Dockerfile

@@ -1,16 +1,22 @@
-FROM python:3.10-slim
+#Use a minimal base image with Python 3.9 installed
+FROM python:3.9-slim
 
+# Prevent Python from writing pyc files and buffering stdout
 ENV PYTHONDONTWRITEBYTECODE=1
 ENV PYTHONUNBUFFERED=1
 
 WORKDIR /app
 
-COPY requirements.txt .
-RUN pip install --no-cache-dir --upgrade pip \
-    && pip install --no-cache-dir -r requirements.txt
-
+#Copy project files
 COPY . .
 
-EXPOSE 7860
+#Upgrade pip and install dependencies
+RUN pip install --no-cache-dir --upgrade pip \
+    && pip install --no-cache-dir -r requirements.txt
 
-CMD ["streamlit","run","app.py","--server.port=7860","--server.address=0.0.0.0","--server.headless=true","--browser.gatherUsageStats=false"]
+#Streamlit runs on 7860 in Hugging Face Spaces
+CMD ["streamlit", "run", "app.py", \
+     "--server.port=7860", \
+     "--server.address=0.0.0.0", \
+     "--server.headless=true", \
+     "--server.enableXsrfProtection=false"]

app.py

@@ -1,6 +1,7 @@
 import streamlit as st
 import pandas as pd
 import requests
+import io
 
 # -----------------------------
 # App Config
@@ -12,7 +13,7 @@ st.set_page_config(
 
 st.title("Predictive Maintenance Predictor")
 st.write(
-    "This app predicts **Predictive Maintenance** based on engine sensor data.  \n"
+    "This app predicts **Predictive Maintenance** based on engine sensor data.\n\n"
     "Developed and Deployed by **Sathiamurthy Samidurai (AIML Student)**."
 )
 
@@ -22,6 +23,63 @@ st.write(
 BACKEND_URL = "https://samdurai102024-predictive-maintenance-be.hf.space/v1/maintenance"
 BACKEND_URL_BATCH = "https://samdurai102024-predictive-maintenance-be.hf.space/v1/maintenancebatch"
 
+# -----------------------------
+# Feature Engineering Functions
+# -----------------------------
+def rpm_category_fn(rpm):
+    if rpm <= 742:
+        return "High Risk"
+    elif rpm <= 886:
+        return "Moderate Risk"
+    else:
+        return "Lower Risk"
+
+def oil_pressure_category_fn(op):
+    if op <= 2.5:
+        return "Low Risk"
+    elif op <= 5.0:
+        return "Mid Risk"
+    else:
+        return "High Risk"
+
+def fuel_pressure_category_fn(fp):
+    if fp <= 4.0:
+        return "Low Risk"
+    elif fp <= 5.1:
+        return "Mid Risk"
+    elif fp <= 6.2:
+        return "High Risk"
+    else:
+        return "Very High Risk"
+
+def coolant_pressure_category_fn(cp):
+    if cp <= 3.1:
+        return "Normal-to-Elevated"
+    elif cp <= 3.8:
+        return "Critical Pressure"
+    else:
+        return "Relief / Saturation Regime"
+
+def fuel_oil_risk_fn(fp, ot):
+    if (fp > 5.1) and (ot < 77.2):
+        return "High Risk"
+    elif (fp > 5.1) and (ot >= 77.2):
+        return "Mid Risk"
+    elif (fp <= 5.1) and (ot < 77.2):
+        return "Mid Risk"
+    else:
+        return "Low Risk"
+
+def coolant_temp_category_fn(ct):
+    if ct < 69.17:
+        return "High"
+    elif ct < 72.27:
+        return "Elevated"
+    elif ct < 80.55:
+        return "Moderate"
+    else:
+        return "Lower"
+
 # -----------------------------
 # User Inputs
 # -----------------------------
@@ -34,18 +92,18 @@ Coolant_pressure = st.number_input("Coolant Pressure", min_value=0.0)
 lub_oil_temp = st.number_input("Lub Oil Temperature", min_value=0.0)
 Coolant_temp = st.number_input("Coolant Temperature", min_value=0.0)
 
-rpm_category = st.selectbox(
-    "RPM Risk Category",
-    ["Lower Risk", "Moderate Risk", "High Risk"]
-)
-
-oil_pressure_category = st.selectbox(
-    "Oil Pressure Risk Category",
-    ["Low Risk", "Mid Risk", "High Risk"]
-)
+# -----------------------------
+# Derived Features (Single)
+# -----------------------------
+rpm_category_val = rpm_category_fn(Engine_rpm)
+oil_pressure_category_val = oil_pressure_category_fn(Lub_oil_pressure)
+fuel_pressure_category_val = fuel_pressure_category_fn(Fuel_pressure)
+coolant_pressure_category_val = coolant_pressure_category_fn(Coolant_pressure)
+fuel_pressure_oil_temp_val = fuel_oil_risk_fn(Fuel_pressure, lub_oil_temp)
+coolant_temp_category_val = coolant_temp_category_fn(Coolant_temp)
 
 # -----------------------------
-# Build Payload (MATCHES FLASK)
+# Payload
 # -----------------------------
 payload = {
     "Engine_rpm": Engine_rpm,
@@ -54,8 +112,12 @@ payload = {
     "Coolant_pressure": Coolant_pressure,
     "lub_oil_temp": lub_oil_temp,
     "Coolant_temp": Coolant_temp,
-    "rpm_category": rpm_category,
-    "oil_pressure_category": oil_pressure_category
+    "rpm_category": rpm_category_val,
+    "oil_pressure_category": oil_pressure_category_val,
+    "fuel_pressure_category": fuel_pressure_category_val,
+    "coolant_pressure_category": coolant_pressure_category_val,
+    "fuel_pressure_oil_temp": fuel_pressure_oil_temp_val,
+    "coolant_temp_category": coolant_temp_category_val
 }
 
 # -----------------------------
@@ -63,23 +125,14 @@ payload = {
 # -----------------------------
 if st.button("Predict", type="primary"):
     try:
-        response = requests.post(
-            BACKEND_URL,
-            json=payload,
-            timeout=10
-        )
-
+        response = requests.post(BACKEND_URL, json=payload, timeout=10)
         if response.status_code == 200:
             result = response.json()
             st.success("Prediction Successful")
-            st.metric(
-                label="Predictive Maintenance Result",
-                value=result["prediction"]
-            )
+            st.metric("Predictive Maintenance Result", result["prediction"])
         else:
             st.error(f"API Error {response.status_code}")
-            st.json(response.json())
-
+            st.write(response.text)
     except requests.exceptions.RequestException as e:
         st.error("Backend service unavailable")
         st.write(str(e))
@@ -90,18 +143,35 @@ if st.button("Predict", type="primary"):
 st.divider()
 st.subheader("Batch Prediction")
 
-file = st.file_uploader(
+uploaded_file = st.file_uploader(
     "Upload CSV file",
     type=["csv"],
-    help="CSV must contain the same feature columns used during training"
+    help="CSV must contain the same raw feature columns used during training"
 )
 
-if file is not None:
+if uploaded_file is not None:
+    df = pd.read_csv(uploaded_file)
+
+    # Feature Engineering
+    df["rpm_category"] = df["Engine_rpm"].apply(rpm_category_fn)
+    df["oil_pressure_category"] = df["Lub_oil_pressure"].apply(oil_pressure_category_fn)
+    df["fuel_pressure_category"] = df["Fuel_pressure"].apply(fuel_pressure_category_fn)
+    df["coolant_pressure_category"] = df["Coolant_pressure"].apply(coolant_pressure_category_fn)
+    df["fuel_pressure_oil_temp"] = df.apply(
+        lambda r: fuel_oil_risk_fn(r["Fuel_pressure"], r["lub_oil_temp"]), axis=1
+    )
+    df["coolant_temp_category"] = df["Coolant_temp"].apply(coolant_temp_category_fn)
+
+    st.dataframe(df.head())
+
     if st.button("Predict for Batch", type="primary"):
+        csv_buffer = io.StringIO()
+        df.to_csv(csv_buffer, index=False)
+
         try:
             response = requests.post(
                 BACKEND_URL_BATCH,
-                files={"file": file},
+                files={"file": ("batch.csv", csv_buffer.getvalue())},
                 timeout=20
             )