Week 4 final assignment - no points for training accuracy

Course Q&A TensorFlow Developer Professional Natural Language Processing in TensorFlow
1

I am working on the final assignment for week 4 - Predicting the next word | Coursera

Model reached required accuracy 0.8

image
image873Ă—307 19.2 KB

But I got no points for this:

image
image880Ă—490 38.5 KB

Please help me understand what went wrong.

Thank you.

2

What is the detailed feedback for the “grading failed” report?

3

image
image877Ă—366 25.5 KB

4

Sorry, I’m not a mentor for that course, so I can’t answer specifically.

Does this course use Jupyter Notebooks?

If yes, the “Grader feedback not found” message often means that you have modified your notebook in some way that makes it not possible for the grader to use for that function.

1 Like
5

I am not sure how I could possibly had done this. Every Exercise was tested. I can’t modify any other cells. What should I do now?

And yes, I am using Jupiter Notebook that is provided in the course.

6

hi @Shurik01

Check if you have not edited/deleted anything outside the marker ###START AND END CODE HERE###.

if you are sure about your codes being correct, here is what you can try next, get a fresh copy, and re-do your assignment by making sure only to replace None or write codes between these assigned markers of ###START AND END CODE HERE###

in case it still fails, then I need to check your code for the grade function you have failed. At that time, you can click on my name and then message me screenshot of the training accuracy grade function codes.

please make sure not to post any codes, link to your assignment here on public post as it is against community guidelines.

Regards

DP

7

This can also be caused by adding, deleting, or moving cells within the notebook.
It can also happen if you added any print() statements for debugging.

8

hi @Shurik01

In the create_model grade function, kindly remove the datatype parameter in the input layer.

Then for embedding layer, you have using directly numerical values for the output dim instead of using variable name. Also you weren’t suppose to use the input length for the embedding layer.

Same goes for lstm layer, you need use the variable name instead of directly using numerical value.

Next, I am not sure if it accepts Adam otpimizer as a class argument, so try to use Adam optimiser as a string argument.

do these corrections and let me know if you still failing on the training accuracy.

Regards
Deepti

9

It worked. Thank you.

1 Like