C1_W1_Lab05_(Cost versus iterations of gradient descent)

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

Hi. I have a question about these two lines below, what is the difference between them?
why in the second one we first separate 9000 and then add 1000 to it and we just don’t write

ax2.plot(J_hist[1000:])?

ax1.plot(J_hist[:100])
ax2.plot(1000+np.arange(len(J_hist[1000:])), J_hist[1000:])

2

Hi!

From the lab they are removing the first 1000 iterations which have Cost values in the 10’s of thousands range to make the second plot easier to read and show the gradual decrease in Cost (elbow of the plot).

3

Hi
I still cannot understand this code:“ax2.plot(1000 + np.arange(len(J_hist[1000:])), J_hist[1000:])”
An integer number can be added with an array?And why the code after the comma is J_hist[1000]?Does J_hist[1000] represent the y axis?
Could you explain it to me in details?Thanks!

4

Repeated. See Optional Lab-------Gradient Descent