C3W3_Assignment Testing with your own questions

[KajetanFrackowiak]

Hello, does anyone know what is incorrect in the predict function?

{moderator edit - solution code removed}

[paulinpaloalto]

The instructions tell you to compute the cosine similarity, but all you’ve done is just the sum of the two vectors. That’s not the same thing, right?

And with a little more careful look, you’re not even implementing the sum correctly. TF “reduce_sum” takes a single tensor argument, right? It then sums across the specified axis (default “None”).

[Deepti_Prasad]

The below comment helped me correct

@KajetanFrackowiak

Extract v1 and v2 from the model output
v1 = v1v2[:, 0]
v2 = v1v2[:, 1]

this code by @arvyzukai NLP mentor

Predict grader cell issue

Remember what the model’s last layer does: [image] it concatenates the outputs. So when you call model.predict(*) you get the concatenated output (1, 256) which you now need to split back into v1 and v2. (the v1v2[0] and v1v2[-1] would not work, since it’s the same thing in this case). You probably have successfully implemented that in Exercise 4 (the classify function). What you need again here is to get: v1.shape - (1, 128) v2.shape - (1, 128) Here is the entire thing For (“When will I…

you need to check v1v2 shape and then check his comment where it mention v1.shape and v2.shape, apply the value to the shape.

Regards
DP

[KajetanFrackowiak]

Thank you for your valuable feedback.

[Deepti_Prasad]

Also what Paul caught another error in your code about cosine similarity when you tf.math_reduce_sum, you are suppose to use correct operator, and I think you must have used similar code in the previous grader cell classify.