My course is Introduction to TensorFlow for Artificial Intelligence, Machine Learning and Deep learning and I have trouble in week 2nd assignment I got zero grade while my answer is correct.
[code removed - moderator]
Please click my name and message your notebook as an attachment. Do include the expanded grader feedback in the message as well.
@Sudhakar_Tripathi
Notebook metadata is corrupt. Please refresh your workspace and try again.
See Refresh your Lab Workspace section here
I am having trouble with the Week 2: Implementing Callbacks in TensorFlow using the MNIST Dataset. I have deleted the program and unites files and refreshed the workspace on multiple occasions. I still get ungraded errors for the three areas of grading. ‘name ‘untrained_model’ is not defined’ which does pass the unittests code in the lab. I have not had this on other classes.
Please code only at places where required. For instance, if you code only where required inside create_and_compile_model, rest of variable definitions should fall in place since untrained_model = create_and_compile_model().
If this is your approach and errors still come up, click on my name to start a private message. Then, attach your notebook as a .ipynb file. Note that mentors cannot access your Coursera Jupyter workspace, so sending the notebook in a .ipynb format is essential.
Please fix this line model = untrained_model inside train_mnnist by calling the function to create a new model instead of relying directly on untrained_model. Here’s the hint:
# Create a compiled (but untrained) version of the model
# Hint: Remember you already coded a function that does this!
A reason for asking the learner to call functions instead of relying on global variables is the grader which evaluates code in isolation when possible.
thank you for the reminder about the function call. Appreciate your help.
Clay