LAB C1_W3_Lab06_Gradient_Descent_Soln does not use vectorized operations

Course Q&A Machine Learning Specialization Supervised ML: Regression and Classification
1

Hi, I’m doing the optional lab C1_W3_Lab06_Gradient_Descent_Soln.
I noticed that you implemented compute_gradient_logistic() function using nested loops, however it would be much more efficient if you used matrix multiplication, wouldn’t it?

So I implemented a “vectorized” implementation, that you may find here, which looks 10 times quicker on your server:

image
image2460×946 323 KB

If you find it useful, please feel free to share my implementation, I’d be very glad if you may just mention me. Thank you!

Mauro

2 Likes
3

Thanks for posting your results. That’s a useful test.
You may implement your code in any way you wish.
Vectorized code usually runs much faster than iterative code.