Reputation: 41
Using python v3.9.2 and tensorflow v2.10.0 in a google cloudshell environment. I'm trying to build a windowed dataset for use in a bidirectional LSTM model. Here's my function:
def windowed_dataset(series: list, window_size: int, batch_size: int, shuffle_buffer: int) -> tensorflow.python.data.ops.dataset_ops.PrefetchDataset:
"""
We create time windows to create X and y features.
For example, if we choose a window of 20, we will create a dataset formed by 20 points as X
"""
dataset = tf.data.Dataset.from_tensor_slices(series)
dataset = dataset.window(window_size + 1, shift=1, drop_remainder=True)
dataset = dataset.flat_map(lambda window: window.batch(window_size + 1))
dataset = dataset.shuffle(shuffle_buffer)
dataset = dataset.map(lambda window: (window[:-1], window[-1]))
dataset = dataset.batch(batch_size).prefetch(1)
return dataset
However, when I instantiate the function, I get this error:
AttributeError: module 'tensorflow' has no attribute 'data'
. Any thoughts? Thanks
Upvotes: 4
Views: 2447
Reputation: 29
Try to upgrade tensorflow:
!pip install -U tensorflow !pip install -U tensorflow_gpu
Upvotes: 1