Upload 2 files

app.py

@@ -0,0 +1,73 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+from tensorflow.keras.models import load_model
+
+# Load the model
+model = load_model("Engine_Fault-small.h5")
+
+# Fault Type Table
+fault_table = pd.DataFrame({
+    "Fault Type": [0, 1, 2, 3],
+    "Fault Name": ["No Fault", "Rich Mixture", "Lean Mixture", "Low Voltage"],
+    "Conditions": [
+        "No Fault",
+        "Incorrect sensor performance, High fuel pressure, Defective injector, faulty pressure regulator, clogged air filter, clogged fuel return line",
+        "Incorrect sensor performance, low fuel pressure, defective injector, faulty pressure regulator",
+        "Worn spark plugs, faulty ignition cables, defective coil, faulty sensor wiring"
+    ]
+})
+
+# Define input columns required by the model
+required_columns = [
+    "MAP", "TPS", "Force", "Power", "RPM", "Fuel consumption L/H",
+    "Fuel consumption L/100KM", "Speed (km/h)", "CO", "HC", "CO2", "O2", "Lambda", "AFR"
+]
+
+def predict_fault(input_text, uploaded_csv):
+    if uploaded_csv is not None:
+        df = pd.read_csv(uploaded_csv)
+    else:
+        # Simulate the case where the text input would describe the dataset
+        return "Please upload a CSV file for prediction. Input text is only for description."
+
+    missing_cols = set(required_columns) - set(df.columns)
+    if missing_cols:
+        return f"Missing required columns: {', '.join(missing_cols)}"
+
+    # Predict
+    X = df[required_columns]
+    predictions = model.predict(X)
+    predicted_faults = np.argmax(predictions, axis=1)
+
+    # Attach fault description
+    df['Predicted Fault Type'] = predicted_faults
+    df = df.merge(fault_table, left_on='Predicted Fault Type', right_on='Fault Type', how='left')
+    return df[['Predicted Fault Type', 'Fault Name', 'Conditions']].head(10)
+
+# Gradio Interface
+demo = gr.Interface(
+    fn=predict_fault,
+    inputs=[
+        gr.Textbox(label="Dataset Description (Optional)", placeholder="Describe the dataset here..."),
+        gr.File(label="Upload CSV File", file_types=[".csv"])
+    ],
+    outputs=[
+        gr.Dataframe(label="Predicted Faults (Top 10 Rows)")
+    ],
+    examples=[["This dataset consists of...", "sample.csv"]],
+    title="Engine Fault Prediction System",
+    description=(
+        "Upload a CSV file containing engine sensor data to predict engine fault types.\n\n"
+        "The model classifies faults into the following types:\n\n"
+        "| Fault Type | Fault Name     | Conditions                                                                 |\n"
+        "|------------|----------------|----------------------------------------------------------------------------|\n"
+        "| 0          | No fault       | No Fault                                                                   |\n"
+        "| 1          | Rich Mixture   | Incorrect sensor performance, High fuel pressure, etc.                    |\n"
+        "| 2          | Lean Mixture   | Incorrect sensor performance, low fuel pressure, etc.                     |\n"
+        "| 3          | Low Voltage    | Worn spark plugs, faulty ignition cables, etc.                            |\n"
+    )
+)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+pandas
+numpy
+tensorflow