Using nds2 on Google colab?

Hi!

I am trying to make call with the nds2 client from Google colab.
So far I failed because the nds2 module does not appear to be
installed automatically by pip. The following lines:

! pip install -q ‘gwpy==3.0.4’
from gwpy.timeseries import TimeSeries
data = TimeSeries.fetch(‘L1:ISI-GND_STS_ITMY_Z_BLRMS_30M_100M’, start=1263229218, end=1263229228, host=‘losc-nds.ligo.org’)

results in the following error:

→ 308 import nds2
309 # pylint: disable=no-member
310

ModuleNotFoundError: No module named ‘nds2’

The nds2 client packages do not seem to be available with pip:

!pip install nds2-client python-nds2-client

Looking in indexes: Simple index, https://us-python.pkg.dev/colab-wheels/public/simple/
ERROR: Could not find a version that satisfies the requirement nds2-client (from versions: none)
ERROR: No matching distribution found for nds2-client

Any hint?

Eric

@ecm That’s right. The NDS2 client is currently packaged for conda, but not pypi. This is a problem for cloud platforms like Google Co-lab that rely on pip to install packages. I have the same problem on streamlit, which also relies on pip.

I don’t have a solution, other than to request that the NDS2 client be packaged for pypi. You could add a ticket for that (and/or send an email), here:

Thank you @jonah ! I’ve opened a ticket.

E.

1 Like

Any update on being able to use nds2 on Colab?

I did

!pip install -q condacolab
import condacolab
condacolab.install()
!conda install -c conda-forge python-nds2-client

That seems to have worked for now for colab notebook

1 Like

That’s a great trick! Thank you for sharing it.