C4W3 Exercise 1

[Bertrand_T_Tameza]

Hello all,

I struggle with exercise 1 of this week course, though I have completed exercise 2 already.

I tried to create the filtering mask like this: filtering_mask = tf.greater(box_class_scores, threshold)
and applied it, for example to box_class_scores, like this:
scores = tf.boolean_mask(box_class_scores, filtering_mask).
When I run the test cell, I get the error that the mask cannot be a scalar.
When I tried to create a tensor using the threshold scalar like this: tf. scalar(threshold), I get the an AttributeError that the module ‘tensorflow’ has no attribute '‘scalar’

Could someone, please, help me solve this ?

Thank you

[ai_curious]

If you tried a search with some targeted keywords, say

https://community.deeplearning.ai/search?q=filtering_mask%20%22.greater%22

you might have found this post with exactly the same root problem

Course 4, Week 3, Part 1: yolo_filter_boxes function problem

The problem is in the tf.math.greater_equal call you are doing to compare a tensor elementwise with a scalar value. It is trying to do the analog of numpy “broadcasting” to convert the second argument into a tensor of the same type and shape as the first argument. But for some reason, it looks like your box_class_scores is of type int64, which causes it to throw that error. How could that happen? You can check it like this: print(f"box_class_scores.shape {box_class_scores.shape}") print(f"box_c…

Let us know if it helps?

[Bertrand_T_Tameza]

Yes it helps, many thanks!