Course: Neural Network and Deep Learning - Programming Assignment Logistic_Regression_with_a_Neural_Network_mindset: Length of costs in final model

Course Q&A Deep Learning Specialization Neural Networks and Deep Learning
1

While executing the line ln[112], the model_test(model) asserts the length of costs to be 1 while received 20.

The the ‘costs’ is defined as a list and in the every 100th iteration the ‘cost’ computed will be appended to ‘costs’

the number of iteration = 2000

image
image1096×358 17.9 KB

2

That probably means you are hard-coding the number of iterations when you call optimize from model. It’s important to understand how “named parameters” work in python. Defining a function is a completely different thing than calling a function. You can’t just “copy/paste” the definition of the function when you want to call it.

You can examine the actual test function by clicking File -> Open and opening the file public_tests.py. What value does that test use for the number of iterations? Note that it is not 2000. So why did you end up with 2000?

3

The number of iteration 2000 was already defined while the assignment was opened

image
image1107×327 22.7 KB

4

Hi @aswincs ,

The named parameter with set value in a function definition refers to the default value to be used if none is given when making a call to that function.

So if you made a call to the model() function without passing num_iterations as a parameter, then the execution of the model() function would take the default value of 2000.

5

Thanks for the support, it is working fine now

6

I have edited the thread title for correctness.