!pip install scikit-learn
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.ensemble import RandomForestClassifier
import gradio as gr
import numpy as np

# Generate random sensor data
np.random.seed(42)
data = pd.DataFrame({
    "Temperature": np.random.randint(0, 100, 100),
    "Humidity": np.random.randint(0, 100, 100),
    "Pressure": np.random.randint(0, 100, 100),
    "Vibration": np.random.randint(0, 100, 100),
    "Alert": np.random.randint(0, 2, 100)
})

# Split the data into features and target
X = data.drop(columns=["Alert"])
y = data["Alert"]

# Scale the features using the StandardScaler
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

# Train a random forest classifier on the scaled data
clf = RandomForestClassifier(random_state=42)
clf.fit(X_scaled, y)

# Define the input and output components for Gradio
input_components = [
    gr.inputs.Slider(minimum=0, maximum=100, default=50, label="Temperature"),
    gr.inputs.Slider(minimum=0, maximum=100, default=50, label="Humidity"),
    gr.inputs.Slider(minimum=0, maximum=100, default=50, label="Pressure"),
    gr.inputs.Slider(minimum=0, maximum=100, default=50, label="Vibration"),
]

output_component = gr.outputs.Textbox(label="Alert")

# Define the predict function to make the prediction using the model
def predict(temp, humidity, pressure, vibration):
    input_data = [[temp, humidity, pressure, vibration]]
    input_data_scaled = scaler.transform(input_data)
    prediction = clf.predict(input_data_scaled)[0]
    return "Alert!" if prediction == 1 else "No alert"

# Create the Gradio interface and run the app
interface = gr.Interface(predict, inputs=input_components, outputs=output_component, title="5G IoT Alert System")
interface.launch()