TF1: How to check your notebook's metadata

The autograders in Courses 1 to 3 looks for tags in your notebook’s metadata so it will know which cells to grade. If you get a score of 0/100 even if your notebook runs fine, then this notebook metadata might be corrupted. You can try getting a new notebook following the instructions here and paste your old solutions there. That usually resolves the issue. However, in case you want to check the metadata manually, then you can do so by following these steps:

Coursera Labs

This applies to Course 1 (all weeks) and Course 3 (except week 3). For Courses 2 to 4, please scroll to the next sections.

  1. Click on View -> Cell Toolbar --> Edit Metadata:



  1. Navigate to cells with a GRADED FUNCTION comment and click the Edit Metadata button on the upper right.


  1. A window will pop up showing the metadata. You should look for a tags key pointing to a list containing a graded string such as the one shown below.

If you only see the id key and no tags for cells that have the GRADED FUNCTION comment, then the metadata is indeed corrupted. You should get a new notebook and paste your old solutions before resubmitting.

Note: If you renamed your old notebook and refreshed your workspace but still don’t get a notebook with the metadata, please report this behavior to Coursera by following the instructions here.


Google Colab

For Course 2 (all weeks) and Course 3 (week 4),

  1. Download your notebook in Colab by selecting File --> Download --> Download .ipynb.

  2. If you have a Jupyter notebook environment in your local computer, then upload the notebook you downloaded. After that, you can follow the same steps as above in the Coursera Labs section.

  3. If you don’t have a Jupyter notebook environment, then you can just open the notebook in any text editor and look for the graded string. You should see something like below:

If the graded tags are missing, then you can simply reopen the notebook from the classroom to load it again on Colab. That should now contain the grader metadata and you can save a new copy from it.

If the graded tags are already on the cells, you can try submitting your work to the autograder.

Course 4

The Course 4 assignments do not depend on grader metadata. Kindly create a public topic in the TF1 Course 4 category so mentors can help you troubleshoot.

2 Likes
C1W1 Tensorflow house price question
Grading glitch?
I keep getting this model is not defined
Model not defined
Where are the answers to C1W1 Assignment 2
I keep getting this model is not defined error
Loading the data
Details of failed tests for create_train_val_dirs
Can someone help in understanding these problems - C1W3
Graded zero WEEK 3 programming assignment
Grader error: 'NoneType' object is not subscriptable
C2W1 Assignment graded output
Model has incorrect type
Why i am getting this error house_model not define i already define
TF Course2 Week 1 Lab1 Dog vs Cat Grader issue
Problem in: Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep learning - Week 3: Improve MNIST with Convolutions
There was a problem compiling the code from your notebook
Implementing Callbacks in TensorFlow using the MNIST Dataset i get error
Error after Assignment submission
Grader is showing Error even when code is running properly
'NoneType' object is not iterable error
W2C4 submission failed
TF1 Course 3 Week 4 Assignment Grader Error
C4 Week One Assignment Grading issue
There was a problem compiling the code from your notebook. Details: name 'sentences' is not defined
Assignment not getting evaluated
I am getting the errors in Assignment-1. I really cannot understand why this is happening. Please help
Error in First Programming Assignment
Problem on assignement
Error in Assignment 1 week 1
Error in Assignment 1 week 1
Error in grading, I delete a cell but cannot back it!
Error while grading
Assignment submission "IPython module error"
Week1 Assignment 1 UNQ_C4 grader error
About assignment C2W1 Graded submission
Colab_metadata_checker error
C3W2_Assignment - There was a problem grading your submission. Details: local variable 'history' referenced before assignment
There was a problem compiling the code from your notebook. Details: name 'train_mnist' is not defined
C2_W1 Graded Functions submission keeps failing due to "No module named 'IPython'"
C1W4 Getting error - [Errno 2] No such file or directory: '/tmp/h-or-s/'
TF1: Assignment Troubleshooting Tips
C1W2_assigment: Implementing Callbacks in TensorFlow using the MNIST Dataset: error : Grader Error: Grader feedback not found
Doubt in Week 1 Assignment: Housing Prices
All functions have incorrect type (Nonetype)
Details of failed tests for create_train_val_dirs