When creating a post, please add:
- Week# must be added in the tags option of the post. Added.
- Link to the classroom item you are referring to: link removed to graded assignment as it is against community guidelines
- Description (include relevant info but please do not post solution code or your entire notebook):
This line: bounding_box_regression_output = tf.keras.layers.Dense(units = ‘4’, name = ‘bounding_box’)(inputs) in the function
def bounding_box_regression(inputs):
causes error when running in Colab. Gemini suggests it should be units = 4.
hi @harrisonle9
you are not suppose to share link to the graded assignment as well as not post any grade cell codes. This is against community guidelines. Please post screenshot of the error you encountered.
Hi @DP It’s not graded assignment, it’s a lab as specified in the subject line.
please share screenshot of the error you are encountering
@harrisonle9
can you confirm the tensor flow version the colab is working in the assignment.
your error is pointing to the unit mentioned as ‘4’ for bounding box regression to be class string. So can you confirm if there was no changes done in the codes already given?
No changes done in the codes already given. It’s very easy to reproduce the problem. While don’t you spend some time to do it?
you haven’t responded to this question?
I haven’t encountered the problem you encountered.
Hi Harrison. There was indeed a bug that was not detected in previous TF versions. It should now be fixed and you’ll get the expected output when using a GPU. There’s an issue when using a TPU and we’ll get back to it soon. Hope this will suffice. Thank you for reporting!
