Untitled

import pandas as pd
import plotly.graph_objects as go
from sklearn.metrics import roc_auc_score, roc_curve

# Load CSV files
df_model_1 = pd.read_csv('model_1_predictions.csv')
df_model_2 = pd.read_csv('model_2_predictions.csv')

# Assuming the true labels are in 'True_Label', and predictions/probabilities are in 'Prediction' and 'Probability'
y_true = df_model_1['True_Label']  # Assuming both models have the same true labels

# Model 1 AUC
fpr_1, tpr_1, _ = roc_curve(y_true, df_model_1['Probability'])
auc_1 = roc_auc_score(y_true, df_model_1['Probability'])

# Model 2 AUC
fpr_2, tpr_2, _ = roc_curve(y_true, df_model_2['Probability'])
auc_2 = roc_auc_score(y_true, df_model_2['Probability'])

# Create plotly figure
fig = go.Figure()

# Add ROC curve for Model 1
fig.add_trace(go.Scatter(x=fpr_1, y=tpr_1, mode='lines', name=f'Model 1 AUC = {auc_1:.2f}', line=dict(color='blue')))

# Add ROC curve for Model 2
fig.add_trace(go.Scatter(x=fpr_2, y=tpr_2, mode='lines', name=f'Model 2 AUC = {auc_2:.2f}', line=dict(color='red')))

# Add diagonal line for random model
fig.add_trace(go.Scatter(x=[0, 1], y=[0, 1], mode='lines', name='Random Model', line=dict(color='gray', dash='dash')))

# Update layout
fig.update_layout(
    title='ROC Curve for Model 1 and Model 2',
    xaxis_title='False Positive Rate',
    yaxis_title='True Positive Rate',
    showlegend=True,
    template='plotly',
    xaxis=dict(scaleanchor="y", range=[0, 1]),
    yaxis=dict(range=[0, 1]),
    plot_bgcolor='white'
)

# Show plot
fig.show()