Upload 7 files

app.py

@@ -0,0 +1,7 @@
+import user, rm_predict
+import gradio as gr
+
+app1 = user.demo  
+app2 = rm_predict.demo
+
+gr.Tabs([app1, app2]).launch()

gradient_boosting_regressor_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f94670712fa53f6fa0b031b8bac9e8e60a841a7fcb1783d7dd5fff1d633a6c4
+size 179724

predictions.csv

@@ -0,0 +1,5 @@
+Identifier,Tire Weight,Predicted Material CO2 Emissions,Raw Material Transportation CO2 Emissions,Fuel CO2 Emissions,Plant Energy Consumption CO2 Emissions
+tire1,8.6,18.916768372987377,0.4815999999999999,140.5,900.0
+tire2,9.5,21.025089181790005,5.319999999999999,421.5,2250.0
+tire3,10.5,17.557486755006813,8.82,72.25,16.2
+tire4,10.5,17.557486755006813,8.82,37.57,16.2

requirements.txt

@@ -0,0 +1,7 @@
+gradio==3.0
+matplotlib==3.5.2
+Pillow==9.2.0  
+scikit-learn==1.0.2
+numpy==1.22.4
+joblib==1.1.0
+csv==1.0.0

rm_predict.py

@@ -0,0 +1,135 @@
+import gradio as gr
+import joblib
+import numpy as np
+import csv
+
+# Load your trained Gradient Boosted Trees model
+model = joblib.load('models/gradient_boosting_regressor_model.joblib')
+
+# Define emission factors directly
+emission_factors = {
+    'Synthetic Rubber': 2.4,
+    'Natural Rubber': 0.639,
+    'Carbon Black': 3.2,
+    'Steel Cord': 2.46,
+    'Silica': 2.06,
+    'Process Oil': 1.61,
+    'Bead Wire': 2.46,
+    'Zinc Oxide': 2.01,
+    'Sulfur': 0.008
+}
+
+# Fuel emission coefficients (kgCO2e per unit)
+fuel_emission_factors = {
+    'Benzene': 2.81,
+    'Kerosene': 2.76,
+    'Diesel fuel': 2.89,
+    'A heavy oil': 3.08,
+    'B • C heavy oil': 3.34,
+    'Liquefied petroleum gas (LPG)': 3.78,
+    'Liquefied natural gas (LNG)': 4.23,
+    'Fuel coal': 2.37
+}
+
+def get_next_identifier():
+    file_path = 'predictions.csv'
+    try:
+        with open(file_path, 'r') as file:
+            last_line = None
+            for last_line in csv.reader(file): pass
+            if last_line and last_line[0].startswith('tire'):
+                last_num = int(last_line[0][4:])
+                return f"tire{last_num + 1}"
+    except FileNotFoundError:
+        pass
+    return "tire1"
+
+def save_prediction_results(identifier, tire_weight, predicted_emission, transport_emission, fuel_emission, plant_emission):
+    file_path = 'predictions.csv'
+    # Attempt to write the header only if the file is being created (i.e., doesn't exist)
+    try:
+        with open(file_path, 'x', newline='') as file:
+            writer = csv.writer(file)
+            writer.writerow(['Identifier', 'Tire Weight', 'Predicted Material CO2 Emissions', 'Raw Material Transportation CO2 Emissions', 'Fuel CO2 Emissions', 'Plant Energy Consumption CO2 Emissions'])
+    except FileExistsError:
+        pass  # File already exists, proceed to append data without writing the header
+    
+    with open(file_path, 'a', newline='') as file:
+        writer = csv.writer(file)
+        writer.writerow([identifier, tire_weight, predicted_emission, transport_emission, fuel_emission, plant_emission])
+
+def predict_and_calculate_emissions(average_distance, fuel_type, fuel_quantity, energy_consumption, tire_weight, synthetic_rubber_weight, natural_rubber_weight, carbon_black_weight, steel_cord_weight, silica_weight, process_oil_weight, bead_wire_weight, zinc_oxide_weight, sulfur_weight):
+    weights = np.array([synthetic_rubber_weight, natural_rubber_weight, carbon_black_weight, steel_cord_weight, silica_weight, process_oil_weight, bead_wire_weight, zinc_oxide_weight, sulfur_weight])
+    emissions = weights * np.array([emission_factors[mat] for mat in emission_factors])
+    features = np.concatenate(([tire_weight], weights, emissions)).reshape(1, -1)
+    predicted_emission = model.predict(features)[0]
+    
+    # Additional Calculations
+    transport_emission = tire_weight * 0.00056 * average_distance
+    fuel_emission = fuel_quantity * fuel_emission_factors[fuel_type]
+    plant_emission = energy_consumption * 0.9
+    
+    # Save results
+    identifier = get_next_identifier()
+    save_prediction_results(identifier, tire_weight, predicted_emission, transport_emission, fuel_emission, plant_emission)
+    
+    return predicted_emission, transport_emission, fuel_emission, plant_emission
+
+# Assuming the predict_and_calculate_emissions function is defined correctly as per your previous code snippet
+
+with gr.Blocks() as demo:
+    gr.Markdown("### Predict Total CO2 Emissions")
+    gr.Markdown("Enter the details to predict emissions and calculate additional environmental impacts.")
+    
+    # Tire and raw material weights
+    with gr.Row():
+        tire_weight = gr.Number(label="Tire Weight (kg)")
+    with gr.Row():
+        synthetic_rubber_weight = gr.Number(label="Synthetic Rubber Weight (kg)", value=0)
+        natural_rubber_weight = gr.Number(label="Natural Rubber Weight (kg)", value=0)
+        carbon_black_weight = gr.Number(label="Carbon Black Weight (kg)", value=0)
+    with gr.Row():
+        steel_cord_weight = gr.Number(label="Steel Cord Weight (kg)", value=0)
+        silica_weight = gr.Number(label="Silica Weight (kg)", value=0)
+        process_oil_weight = gr.Number(label="Process Oil Weight (kg)", value=0)
+    with gr.Row():
+        bead_wire_weight = gr.Number(label="Bead Wire Weight (kg)", value=0)
+        zinc_oxide_weight = gr.Number(label="Zinc Oxide Weight (kg)", value=0)
+        sulfur_weight = gr.Number(label="Sulfur Weight (kg)", value=0)
+
+    # Fuel type selection in its own row
+    with gr.Row():
+        fuel_type = gr.Radio(['Benzene', 'Kerosene', 'Diesel fuel', 'A heavy oil', 'B • C heavy oil', 'Liquefied petroleum gas (LPG)', 'Liquefied natural gas (LNG)', 'Fuel coal'], label="Fuel Type")
+
+    # Additional parameters in a compact 3-box layout
+    with gr.Row():
+        average_distance = gr.Number(label="Average Transportation Distance (km)", value=0)
+        fuel_quantity = gr.Number(label="Fuel Quantity (liters/kg)", value=0)
+        energy_consumption = gr.Number(label="Plant Energy Consumption (kWh per tire)", value=0)
+    
+    btn_predict = gr.Button("Predict", align="right")  
+    
+      # Outputs
+    predicted_material_emission = gr.Textbox(label="Predicted Material CO2 Emissions (kgCO2e)")
+    raw_material_transportation_emission = gr.Textbox(label="Transportation CO2 Emissions (kgCO2e)")
+    fuel_emission = gr.Textbox(label="Fuel CO2 Emissions (kgCO2e)")
+    plant_energy_consumption_emission = gr.Textbox(label="Plant Energy CO2 Emissions (kgCO2e)")
+
+    # Button to execute the function
+    
+
+    btn_predict.click(
+        fn=predict_and_calculate_emissions,
+        inputs=[
+            average_distance, fuel_type, fuel_quantity, energy_consumption,
+            tire_weight, synthetic_rubber_weight, natural_rubber_weight, carbon_black_weight,
+            steel_cord_weight, silica_weight, process_oil_weight, bead_wire_weight,
+            zinc_oxide_weight, sulfur_weight
+        ],
+        outputs=[
+            predicted_material_emission, raw_material_transportation_emission,
+            fuel_emission, plant_energy_consumption_emission
+        ]
+    )
+
+demo.launch()

user.py

@@ -0,0 +1,68 @@
+import gradio as gr
+import csv
+import matplotlib.pyplot as plt
+import io
+from PIL import Image
+
+def display_emissions_data_and_chart(tire_id):
+    file_path = 'predictions.csv'
+    try:
+        with open(file_path, mode='r', newline='') as file:
+            reader = csv.DictReader(file)
+            for row in reader:
+                if row['Identifier'] == tire_id:
+                    # Extracting data for pie chart
+                    emissions_labels = [
+                        'Predicted Material CO2', 
+                        'Raw Material Transportation CO2', 
+                        'Fuel CO2', 
+                        'Plant Energy Consumption CO2'
+                    ]
+                    emissions_values = [
+                        float(row.get('Predicted Material CO2 Emissions', 0)),
+                        float(row.get('Raw Material Transportation CO2 Emissions', 0)),
+                        float(row.get('Fuel CO2 Emissions', 0)),
+                        float(row.get('Plant Energy Consumption CO2 Emissions', 0))
+                    ]
+
+                    # Generate Pie Chart
+                    fig, ax = plt.subplots(figsize=(9, 6))  # Adjusted figure size to be 50% larger
+                    ax.pie(emissions_values, labels=emissions_labels, autopct='%1.1f%%', startangle=90)
+                    ax.axis('equal')  # Equal aspect ratio ensures that pie is drawn as a circle.
+
+                    buf = io.BytesIO()
+                    plt.savefig(buf, format='png', bbox_inches='tight')
+                    buf.seek(0)
+                    plt.close(fig)
+
+                    # Convert bytes to PIL Image
+                    buf_image = Image.open(buf)
+                    return buf_image
+                    
+    except FileNotFoundError:
+        return "The predictions.csv file was not found."
+
+# Define a wrapper function for Gradio that calls the display function
+def get_emissions_data(tire_id):
+    image = display_emissions_data_and_chart(tire_id)
+    if isinstance(image, Image.Image):
+        return image
+    else:
+        # If the result isn't an image, return a placeholder or error image
+        return Image.new('RGB', (200, 200), color='red')
+
+# Set up the Gradio Blocks
+with gr.Blocks() as demo:
+    with gr.Row():
+        tire_id_input = gr.Textbox(label="Enter Tire ID")
+        submit_button = gr.Button("Submit")
+    
+    output_image = gr.Image(label="Emissions Data Chart")
+    
+    submit_button.click(
+        fn=get_emissions_data,
+        inputs=tire_id_input,
+        outputs=output_image
+    )
+
+demo.launch()