C3W3 Issues with Exercise 4: Classify

[N_Adhikari]

Similar to the issues in my previous post (C3W3 Issues with Triplet Loss), I am facing issues in exercise 4

The accuracies are incorrect and I can’t understand why. I implemented the same logic in exercise 5 and that passed all testcases. Can someone help me out please?

[Deepti_Prasad]

hi @N_Adhikari

refer this

NLP Course 3 Week 3: Confusion matrix

Hi @Taisiya_Kopytova You somehow mixed up two code lines Check if d>threshold to make predictions Here you were only suppose to check if cosine similarity is greater than threshold but you ended up adding the accuracy part of the code to this code line. Also do not hard code the path by introducing res = d > threshold (I KNOW THE HINT FROM PREVIOUS CELLS TELLS YOU TO DO SO. INSTEAD OF y_test == res, you only need to use tf.cast(d>threshold with the datatype). Next your accuracy code is…

see if it helps, if still doesn’t then send the screenshot by personal DM

[N_Adhikari]

Hey
So, I actually did refer to this exact post while writing the code, which is why I’m having a hard time figuring out what’s wrong. Especially since the same logic works in exercise 5. I’ll attach the code in a personal message.
Please let me know what I’m missing out on

[Deepti_Prasad]

Sorry for delayed response, I was out.

the code line

take the average of correct predictions to get the accuracy, you are checking average of predictions where it mentions you to take average of correct predictions.

so you mentioning just y_pred would be incorrect.

rather take tf.cast(when y pred is absolute equal to your actual targets, datatype)

here is a similar threads

C3W3 assignment Classify test issue

code is partly incorrect, please do not use tf.equal and also in accuracy we check if if the values of two operands are equal or not , so use == between the actual targets and predictions.

[N_Adhikari]

Oh thank you so much! You’ve been of great help!