nkarasiak / museotoolbox Goto Github PK

It would be great to have a number of jobs to run in parallel museotoolbox.processing.RasterMath .
I mean, if a raster has 4 blocks or more, you can use n_jobs=4 to run in parallel each block for your function.

I'm installing this into a fresh python 3.5 anaconda evironment on mac osx 10.14.6.

Installation as per readme file returned no errors, however when trying to folow an example I couldn't import museotoolbox.

initially, I got an error like "couldn't find osgeo" so I installed gdal via conda, then tried again, got the following:
ImportError: dlopen(//anaconda/envs/ipykernel_py3/lib/python3.5/site-packages/osgeo/_gdal.so, 2): Library not loaded: @rpath/libjpeg.8.dylib
Referenced from: /anaconda/envs/ipykernel_py3/lib/libgdal.20.dylib
Reason: image not found

Eventual solution found here: https://hackernoon.com/install-python-gdal-using-conda-on-mac-8f320ca36d90 -- downgrade jpeg from 9 to 8: conda install -f jpeg=8

probably should include gdal as a dependency and maybe flag the jpeg problem on the readme to save someone some time

cc: openjournals/joss-reviews#1978

I think that the paper would benefit from a more specific comparison to the state-of-the-art in the field and how this package improves or differs. It may simply be that the explanation is unclear; the second sentence in the opening paragraph of the paper is very long and difficult to parse. There are also no references provided.

Tagging: openjournals/joss-reviews#1978

The bot encountered an error in your .pyup.yml config file:

  in "<unicode string>", line 8, column 1:
    branch:develop
    ^
could not find expected ':'
  in "<unicode string>", line 10, column 1:
    # search for requirement files
    ^

You can validate it with this online YAML parser or by taking a look at the Documentation.

How should data be imported-- it looks like load historical data just assigns pathnames to the variables, so would I just use raster = "~/path/to/raster.tiff" to use my own file?

What file formats are compatible with museo? shapefiles? NetCDF files? Does museo check whether the vector and raster projections match? If I am preparing an ROI file in, say, QGIS and want to use it for processing in Museo, what do I need to know about how to set up that file?

cc: openjournals/joss-reviews#1978

Starting from a fresh install of miniconda with this package installed.

from museotoolbox.ai import SuperLearner

fails with

ModuleNotFoundError: No module named 'osgeo'

Which isn't in the requirements.txt.

openjournals/joss-reviews#1978

André MIRALLES highlighted a bug when using getSamplesFromROI in conda env using Windows.

Things to do :

• Do getSamplesFromROI by default in memory
• If not enough memory, do the work by writing image on temp folder then remove it

MuseoToolBox v2 should be built upon rasterio, especially rasterMath and other functions which use raster reading and writing process.

There's a bug in RasterMath when using multiple input images and a mask.
This should be solve with the new RasterMath well reorganized (thanks to Sigma student, Mr loop and Ms. Parallel)

When installing with python3 -m pip install museotoolbox --user, pip throws the following:

WARNING: The scripts f2py, f2py3 and f2py3.7 are installed in '/home/[user]/.local/bin' which is not on PATH.
Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location.

This might be a non-issue for functionality. I don't know.

openjournals/joss-reviews#1978

SpatialLeavePSideOut from crossValidation must be activated for v1.0-beta.