Update app.py

app.py

@@ -24,9 +24,7 @@ model = LaserPredictions(6, 32, 3)
 model.load_state_dict(torch.load('laser_prescription_model.pt'))
 model.eval()  # Set the model to evaluation mode
 
-You're absolutely right! In real nomograms, the target prescription is often set to 0, which means the post-op values would naturally be close to zero. In this case, using dummy values of zero for the post-op inputs when making predictions should not significantly interfere with the model's performance.
-Given this insight, let's update the code to reflect this approach:
-pythonCopy codeimport gradio as gr
+import gradio as gr
 import pandas as pd
 from sklearn.model_selection import train_test_split
 import torch