From 8cb34b38a2d10720c37c04c45523118d03a30942 Mon Sep 17 00:00:00 2001 From: Raghuram Subramani Date: Sun, 31 Aug 2025 13:41:53 -0400 Subject: . --- projects/autograd.html | 104 +++++++++++++++++++++++++------------------------ 1 file changed, 53 insertions(+), 51 deletions(-) (limited to 'projects/autograd.html') diff --git a/projects/autograd.html b/projects/autograd.html index 4fe4ae0..5f37054 100644 --- a/projects/autograd.html +++ b/projects/autograd.html @@ -1,65 +1,67 @@ - + - - - - + + + + - Autograd | COMPROMYSE + Autograd | COMPROMYSE - - - - + + + + - + - - - + + + - - - + + + - - -

+ + +

- -

+

-

-

Autograd

-

-

An implementation of autograd / backpropagation.

-

Python

-

+

+

Autograd

+

+

An implementation of autograd / backpropagation.

+

Python

+

- SOURCE -

+ + SOURCE + +

-

-

All you need to run a simple neural network using autograd is the following code:

+

+

All you need to run a simple neural network using autograd is the following code:

The code defines a data set X, expected output (or ground truth) y. It then trains the neural network by performing backward propagation (.backward()), then applies the calculated gradients through .optimise() along with a learning rate of 0.01.

@@ -83,22 +85,22 @@ for i in range(400): loss.backward() n.optimise(0.01) -print(pred) + print(pred)

-

+

-

-

+ -

- +

+ -- cgit v1.2.3