C3_W4 assignment Question duplicates

Course Q&A Natural Language Processing NLP with Sequence Models
1

Hi,
I have the following problem when run the question duplicate assignment on VSCode connected to ubuntu WSL. The error doesn’t appear when I run the code in the Jupitor?
what’s the reason ?

Blockquote

No GPU/TPU found, falling back to CPU. (Set TF_CPP_MIN_LOG_LEVEL=0 and rerun for more info.)
/home/ala_jabari/miniconda3/lib/python3.9/site-packages/jax/_src/xla_bridge.py:790: UserWarning: jax.host_count has been renamed to jax.process_count. This alias will eventually be removed; please update your code.
warnings.warn(
Traceback (most recent call last):
File “/home/ala_jabari/NLP/NLP_with_Sequance_Models/Question_duplicates/main.py”, line 276, in
training_loop = train_model(Siamese, TripletLoss, train_generator, val_generator)
File “/home/ala_jabari/NLP/NLP_with_Sequance_Models/Question_duplicates/main.py”, line 268, in train_model
training_loop = training.Loop(Siamese(),
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/trax/supervised/training.py”, line 294, in init
self.load_checkpoint()
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/trax/supervised/training.py”, line 944, in load_checkpoint
matched_slots, _ = fastmath.tree_unflatten(
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/trax/fastmath/numpy.py”, line 244, in tree_unflatten
new_t, rest = tree_unflatten(rest, t, copy_from_tree=copy_from_tree)
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/trax/fastmath/numpy.py”, line 244, in tree_unflatten
new_t, rest = tree_unflatten(rest, t, copy_from_tree=copy_from_tree)
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/trax/fastmath/numpy.py”, line 239, in tree_unflatten
if copy_from_tree is not None and tree in copy_from_tree:
File “/home/ala_jabari/miniconda3/lib/python3.9/site-packages/jax/_src/numpy/array_methods.py”, line 260, in deferring_binary_op
raise TypeError(f"unsupported operand type(s) for {opchar}: "
TypeError: unsupported operand type(s) for ==: ‘ArrayImpl’ and ‘tuple’

2

Hi @Ala_Jabari1,

The coding environment in Coursera is “frozen”. What I mean by that is, it will always run specific versions of all the libraries whenever you’ll open your workspace.

Usually when learners download the material to run locally, they use different versions of the libraries (maybe the latest versions) to run the assignment, and it can cause compatibility issues between the libraries.

This is one of the issues why you might be failing this when running locally. Another could be, have you downloaded all of the related files to run with the assignment ?

To avoid these inconveniences for learners, this is why we use the coursera coding environment where learners don’t have to do anything, just open the workspace and run the labs.

Unfortunately, we do not provide support to run the labs locally. As you can imagine, there are different OS-es and different preference of platforms to run python code, this will become a mega task to cater every individual trying to do this.

Best,
Mubsi