Lab 3 Part 2 resampling gives error

[jt-health-dev]

Starting epoch 1/6

AttributeError Traceback (most recent call last)
in <cell line: 5>()
7 # Get a batch of training data and compute the training step
8 for step, data in enumerate(train_loader):
----> 9 metrics = wrapper.train_step(data)
10 if step % 100 == 0:
11 print(step)

2 frames
/usr/local/lib/python3.10/dist-packages/keras/engine/compile_utils.py in match_dtype_and_rank(y_t, y_p, sw)
829 def match_dtype_and_rank(y_t, y_p, sw):
830 """Match dtype and rank of predictions."""
--> 831 if y_t.shape.rank == 1 and y_p.shape.rank == 2:
832 y_t = tf.expand_dims(y_t, axis=-1)
833 if sw is not None:

AttributeError: 'tuple' object has no attribute 'rank'

[kennylids]

Same error here. Seems like the team updated the library.

@aamini please help

[fractalclockwork]

Here is a messy workaround for this data type casting issue...
` for step, data in enumerate(train_loader):
x,y = data
xx = [tf.convert_to_tensor(a) for a in x]
yy = [tf.convert_to_tensor(b) for b in y]
xx = tf.convert_to_tensor(xx)
yy = tf.convert_to_tensor(yy)

data = (xx,yy)
metrics = wrapper.train_step(data)`