spacetelescope/tike_content

High level walk-through, notebooks and features of the STScI Time series Integrated Knowledge Engine.

Timeseries Integrated Knowledge Engine (TIKE) Content

This repository contains curated tutorials that can teach you how to use and access TESS/Kepler data on TIKE, the MAST-hosted science platform. It also demonstrates how to make use of the cloud-hosted data and the AWS computational environment when analyzing these data.

This repo is intended to contain community driven content. If you see topics that are missing, you're encouraged to either open an issue or create a pull request. We welcome your contributions and feedback!

Contributing Guidlines

This repository primarily brings together information available in a variety of locations in order to teach about time series data analysis and computingin the cloud in a way that is pertinent to those using the TESS and Kepler data sets.

1. To add a new topic to this curation, open an issue in this repository.
2. To correct information contained in this curation effort, consider a pull request against this repository.
3. To contribute a new Jupyter Notebook or Tutorial, consider adding it to the MAST Notebooks repository and follow the [Notebook style guide].(https://github.com/spacetelescope/style-guides/blob/master/guides/jupyter-notebooks.md)

tike_content's Issues

This is a test issue

I will delete it shortly...

pip install from notebook points to the wrong environment

When installing a new python module through pip install module inside the notebook and running the TESS environment, the package gets installed in the base environment.
The package gets installed correctly but n the wrong environment. Then when loading the module inside the notebook, this one isn't found.

suggested change to instructions for creating persistent conda environments

The instructions for creating persistent environments in tike_content/content/extra-software.md do not work consistently due to restrictions on cross-device hardlinks (and of course they wouldn't be a good idea anyway, nor would symlinks, because the linked objects would disappear on the subsequent launch). Conda's fallback copying will usually work eventually, but the check-and-fall-back process is pretty slow, could take hours for large environments, and in some cases will fail. The usual remedy for this type of issue is to use the --copy argument to conda create, and I suggest you include that in extra-software.md. It is still slow, but will generally be more consistent and faster than going through the attempted hardlink process. Note that conda-env will not accept this argument, so this method cannot be used to create a persistent environment from an environment definition file.

Example snippet of errors (a typical environment will create hundreds of thousands):

  src: /opt/conda/pkgs/libdb-6.2.32-h9c3ff4c_0/include/dbstl_inner_utility.h
  dst: /home/jovyan/local_envs/fornax-slice-testing/include/dbstl_inner_utility.h
DEBUG conda.gateways.disk.create:create_link(363): OSError(18, 'Invalid cross-device link')
DEBUG conda.gateways.disk.create:create_link(364): hard-link failed. falling back to copy
  error: OSError(18, 'Invalid cross-device link')
  src: /opt/conda/pkgs/libdb-6.2.32-h9c3ff4c_0/include/dbstl_map.h
  dst: /home/jovyan/local_envs/fornax-slice-testing/include/dbstl_map.h
DEBUG conda.gateways.disk.create:create_link(363): OSError(18, 'Invalid cross-device link')
DEBUG conda.gateways.disk.create:create_link(364): hard-link failed. falling back to copy
  error: OSError(18, 'Invalid cross-device link')
  src: /opt/conda/pkgs/libdb-6.2.32-h9c3ff4c_0/include/dbstl_resource_manager.h
  dst: /home/jovyan/local_envs/fornax-slice-testing/include/dbstl_resource_manager.h
DEBUG conda.gateways.disk.create:create_link(363): OSError(18, 'Invalid cross-device link')
DEBUG conda.gateways.disk.create:create_link(364): hard-link failed. falling back to copy
  error: OSError(18, 'Invalid cross-device link')
  src: /opt/conda/pkgs/libdb-6.2.32-h9c3ff4c_0/include/dbstl_set.h
  dst: /home/jovyan/local_envs/fornax-slice-testing/include/dbstl_set.h
DEBUG conda.gateways.disk.create:create_link(363): OSError(18, 'Invalid cross-device link')
DEBUG conda.gateways.disk.create:create_link(364): hard-link failed. falling back to copy