[BUG-REPORT] Vaex to_dask_array fails AttributeError: 'ValueError' object has no attribute 'numpy'

[Ben-Epstein]

Thank you for reaching out and helping us improve Vaex!

Before you submit a new Issue, please read through the documentation. Also, make sure you search through the Open and Closed Issues - your problem may already be discussed or addressed.

Description
Please provide a clear and concise description of the problem. This should contain all the steps needed to reproduce the problem. A minimal code example that exposes the problem is very appreciated.

Software information

• Vaex version (import vaex; vaex.__version__): {'vaex-core': '4.8.0', 'vaex-hdf5': '0.11.1'}
• Vaex was installed via: pip / conda-forge / from source pip
• OS: MacOS

Additional information
Please state any supplementary information or provide additional context for the problem (e.g. screenshots, data, etc..).

import vaex

def tolist(x,y):
    return list([x,y])

data = {'A':[1,2,1],'B':['a','b','c'], 'C':['d', 'e', 'f']}
df = vaex.from_dict(data)
df.to_dask_array()

[Ben-Epstein]

Seems like it's any non numeric columns that cause the issue

from random import randint 

df = vaex.from_arrays(
    id=[1,1,4,2,5,64,46,64,4,2,5],
    letter = ["a" for _ in range(11)]
)

df[["id"]].to_dask_array()  # works
df[["id","letter"]].to_dask_array()  # error above

[JovanVeljanoski]

True, exporting non numeric data to a dask array is not supported (for now).

[maartenbreddels]

Yeah, we could improve on the error though. Not sure if we can have a meaningful dask-arrow-array export btw.

[maartenbreddels]

I think this is a good first issue to improve the error message.