search

glasgowcompbio / vimms

A programmable and modular LC/MS simulator in Python
MIT License
19 stars 6 forks source link

Chemical creator seems broken for creating a set of chemicals from scratch #62

Closed joewandy closed 4 years ago

joewandy commented 4 years ago

image

See https://github.com/sdrogers/vimms/blob/master/demo/01.%20Data/03.%20Generating%20a%20set%20of%20random%20chemicals.ipynb

sdrogers commented 4 years ago

How do all the tests work then?

Sent from my iPhone

On 13 Aug 2020, at 11:39, Joe Wandy notifications@github.com wrote:



See https://github.com/sdrogers/vimms/blob/master/demo/01.%20Data/03.%20Generating%20a%20set%20of%20random%20chemicals.ipynb

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

sdrogers commented 4 years ago

Have you run notebooks 1 and 2? In number 3, the one above, the cell that fails says that it needs the ROI data from notebook 1...

sdrogers commented 4 years ago

The issue isn't in ChemicalCreator, it's a problem with paths etc in that notebook. It's looking for ROI sources in <base_dir>/beers/datasets where (presumably) <base_dir> is as defined in notebook 1.

Will leave to @vinnydavies to fix this.

  1. base_dir needs to be defined in 03. Generating a set of random chemicals. Should it be the same as data_dir?
  2. Fix the bit that Joe points out above -- I think it's just looking for files in the wrong place.
joewandy commented 4 years ago

Have you run notebooks 1 and 2? In number 3, the one above, the cell that fails says that it needs the ROI data from notebook 1...

I haven't tried running the notebook, sorry. Was just putting what Vinny told me on slack yesterday into this issue for tracking.

How do all the tests work then?

They work, but some have their own issues that need to be fixed (see https://github.com/sdrogers/vimms/issues/58).

vinnydavies commented 4 years ago

I've updated the demo, so this is working now