[Chap 16] TypeError: '_NestedVariant' object is not subscriptable

[lucacox]

In chapter 16 notebook running this cell:

batch_size = 32
dataset = dataset.shuffle(10000).batch(batch_size)
dataset = dataset.map(lambda windows: (windows[:, :-1], windows[:, 1:]))

produces this error in Colab:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-22-58d8a4056e3e> in <module>()
      6 batch_size = 32
      7 dataset = dataset.shuffle(10000).batch(batch_size)
----> 8 dataset = dataset.map(lambda windows: (windows[:, :-1], windows[:, 1:]))

10 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/autograph/impl/api.py in wrapper(*args, **kwargs)
    256       except Exception as e:  # pylint:disable=broad-except
    257         if hasattr(e, 'ag_error_metadata'):
--> 258           raise e.ag_error_metadata.to_exception(e)
    259         else:
    260           raise

TypeError: in user code:

    <ipython-input-15-6240017d8f77>:8 None  *
        dataset = dataset.map(lambda windows: (windows[:, :-1], windows[:, 1:]))

    TypeError: '_NestedVariant' object is not subscriptable

[ibeauregard]

I can't reproduce your error.

[ageron]

Hi @lucacox ,

I can't reproduce the error either. Could you please make sure you're using the latest versions of the code and the libraries and try again? You can find the instructions to update these here.

[Yarxin]

The solution of your problem is to run whole script from the begining, or just restart the kernel (clear workspace if working in jupyter lab or jupyter notebook).

The issue occures when you run lines BETWEEN

train_size = dataset_size * 90 // 100
dataset = tf.data.Dataset.from_tensor_slices(encoded[:train_size])

AND

batch_size = 32
dataset = dataset.shuffle(10000).batch(batch_size)
dataset = dataset.map(lambda windows: (windows[:, :-1], windows[:, 1:]))

more than once.

[asoisaac]

Thanks. Rerunning the entire script from the beginning helps.