Computational Graph: Forward & Backward Passes
Concept 01 Demo
Input (x):
2.0
Weight (w):
3.0
Target Value (y_true):
10.0
1. Forward Pass ▶
2. Backward Pass ◀
Forward Prediction (y = w·x + b)
7.000
Output Loss L = (y - y_true)²
9.000
Backprop Gradient (dL / dw)
-12.000