I am trying to use the np.linalg.norm() function within the verify() function in week 4 assignment 1 exercise 2. Right now I have dist = np.linalg.norm(float(identity), encoding) and I’m getting the error “could not convert string to float: ‘bertrand’”. I think this is because ‘identity’ is a di…

Ok I realized that I should have been using the database dictionary instead of the identity list of strings. I am now getting the error: “The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()” I get that dist is an array of distances of the values in the data…

It seems logical to, instead of using the if statement to find a distance value less than 0.7, to instead find the minimum distance value in the dist array

I figured it out, for anyone wondering, the np.linalg.norm() function only takes one parameter, as it doesn’t compute the difference between database[identity] and ‘encoding’. I still don’t understand exactly the “distance” that it computes, however and how this “distance” is different that the diff…

Did you use dist = np.linalg.norm(tf.subtract(encoding, database)) to compute distance?! Because I wrote this but it didn’t worked!

database is a dictionary use database[identity] to get encoding of identity

Verify() function

Course Q&A Deep Learning Specialization Convolutional Neural Networks

Rodolfo_Novarini April 18, 2023, 5:54pm 8

Found the answer here: Error in verify()

Topic		Replies	Views
Error in verify() Convolutional Neural Networks coursera-platform	1	870	July 26, 2021
C4 W4 A1 - Error with verify function Convolutional Neural Networks coursera-platform	2	603	March 8, 2022
DLS Course 4 Assignment 2 Face verify error Convolutional Neural Networks coursera-platform	11	759	March 8, 2022
Exercise 3 - who_is_it Convolutional Neural Networks coursera-platform	4	555	June 7, 2022
Week 4 >>> Face_Recognition programming exercise >>> Exercise 2 - verify Convolutional Neural Networks coursera-platform	3	716	February 12, 2022

Verify() function

Related topics