CaxtonEmeraldS commited on
Commit
ef82c39
·
verified ·
1 Parent(s): 24ebfe4

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +3 -1
app.py CHANGED
@@ -86,7 +86,9 @@ def update_neurons(activation, seed):
86
  with gr.Blocks() as demo:
87
  # gr.Markdown("# Cholestrol Concentration Prediction - ANN and Linear Model")
88
  gr.Markdown("# **Cholestrol Concentration Prediction - ANN and Linear Model**")
89
- gr.Markdown("Dynamically select models and predict cholesterol concentration.")
 
 
90
 
91
  gr.Markdown("### **Authors:**")
92
  gr.Markdown("""
 
86
  with gr.Blocks() as demo:
87
  # gr.Markdown("# Cholestrol Concentration Prediction - ANN and Linear Model")
88
  gr.Markdown("# **Cholestrol Concentration Prediction - ANN and Linear Model**")
89
+ gr.Markdown("Dynamically select models and predict cholesterol concentration. For more information on dataset preparation and the associated experiment, kindly refer to the journal article.")
90
+
91
+ gr.Markdown("This study presents Artificial Neural Networks (ANNs) and Linear Regression models for predicting cholesterol concentration from RGB colourimetric measurements. Around 2,500 single hidden layered ANN models, with varying activation functions, seed initialisations, and neuron counts were trained to approximate the non-linear relationship between colour channels and concentration levels. The trained models follow a 3-×-1 architecture with three input features (mean R, mean G, mean B), a single hidden layer of varying neurons, and one output neuron. A simple linear regression model was developed alongside as a baseline for comparison. The interface allows users to dynamically select the ANN model configuration and compare its predictions against the linear model. It also supports model selection, and performance evaluation for colour-based biosensing applications.")
92
 
93
  gr.Markdown("### **Authors:**")
94
  gr.Markdown("""