Changes between Version 10 and Version 11 of GSoC/2014/ImplementationOf3DRasterFlowLine


Ignore:
Timestamp:
Jul 25, 2014, 2:59:31 PM (10 years ago)
Author:
annakrat
Comment:

Week 10

Legend:

Unmodified
Added
Removed
Modified
  • GSoC/2014/ImplementationOf3DRasterFlowLine

    v10 v11  
    131131=== Week 9 ===
    132132I rewrote the rest of the Python implementation (added flowaccumulation and on-the-fly gradient computation). I tested it against Python implementation and when computing gradient, it still differs a little bit so I will test more. Next week I will add more code documentation and try to write some tests. Also, I would like to implement integrating flowlines in both directions.
     133
     134=== Week 10 ===
     135I didn't have much time this week, so I just wrote more documentation and updated manual page with description and example. I started to write tests, I created a c test module to access the module functions and then I wrote the python test file testing interpolation. However, this broke installing the extension so I removed some changes from svn.
     136
     137I also added the option to compute flowlines in both directions at once.
     138
     139The differences in gradient computation from last week are resolved, it was partially caused by Rast3d_location2coord implementation, which was changed and fixed by Soeren.
     140
     141Next week I would like to try different tile settings to see if there is any difference and then try to implement !r3.gradient module based on the gradient function in !r3.flow.