Update app.py

app.py

@@ -1,4 +1,5 @@
 import time 
+import math
 
 import pandas as pd
 import matplotlib.pyplot as plt
@@ -53,10 +54,20 @@ def check_model_drift():
     sample_df = get_data()
     p_pos_label_training_data = 0.03475
     training_data_size = 8000
-    p_pos_label_sample_logs = sample_df.prediction.value_counts()
     
-    return p_pos_label_sample_logs
+    p_0 = sample_df.prediction.value_counts()[0]
+    p_1 = sample_df.prediction.value_counts()[1]
+
+    p_pos_label_sample_logs = p_1/(p_0+p_1)
+    
+    variance = (p_pos_label_training_data * (1-p_pos_label_training_data))/training_data_size
+    p_diff = abs(p_pos_label_training_data - p_pos_label_sample_logs)
     
+    if p_diff > 2 * math.sqrt(variance):
+        print("Model Drift Detected!")
+    else:
+        print("No Model Drift!")
+
     
 with gr.Blocks() as demo:
     gr.Markdown("# Real-time Monitoring Dashboard")
@@ -65,6 +76,6 @@ with gr.Blocks() as demo:
     
     with gr.Row():
         with gr.Column():
-            gr.Textbox(f"Data refreshed at {time.ctime()}")
+            gr.Textbox(check_model_drift, every=5)
 
 demo.queue().launch()