Xo tools #235
Xo tools #235
Conversation
Still needs testing, might have discrepancies with single cell calculation.
…ne-cube collision finding that did approximately nothing (0.5*sqrt(3) is still a good threshold that catches the missing pieces)
…rid support in vlsvWriter
…er), to enable auxiliary outputs from the calculation
…ing a validation case, sadly that setup is current-free and fiddly.
…ake slicing op faster
…te_variable_info-type behaviour, and a another write helper (write_variable)
|
@markusbattarbee want to have a look at the restructuring? |
|
|
||
| f.map_vg_onto_fg() | ||
| print('vg mapped to fg, elapsed:', time.time()-t) | ||
| argsorti=f.read_variable("CellID").argsort() |
There was a problem hiding this comment.
argsorti and rev_argosorti are also not used
If this is supposed to be an example file, I think it dould do a lot better job of explaining what it does ;)
| import numpy as np | ||
| import sys | ||
|
|
||
| # Box: correct distance to corners hg_sdf |
There was a problem hiding this comment.
It's the Urs et co SDF library at https://mercury.love/hg_sdf/
|
|
||
| # Box: correct distance to corners hg_sdf | ||
| def fBox(pp, bb=np.array([0.5,0.5,0.5])): | ||
| p = np.atleast_2d(pp) |
There was a problem hiding this comment.
variable namings are a bit obscure
| vgarr[:,i] = np.divide(sums,counts) | ||
| else: | ||
| sums = np.bincount(np.reshape(self.__vg_indexes_on_fg, self.__vg_indexes_on_fg.size), weights=np.reshape(array,array.size)) | ||
| vgarr = np.divide(sums,counts) | ||
| return vgarr | ||
|
|
||
| def apply_fsgrid_averaging_on_vg(self, array): |
There was a problem hiding this comment.
comment on what this actually does?
There was a problem hiding this comment.
Huh, that's basically an unused duplicate of the function above.
| from matplotlib.lines import Line2D | ||
| import time | ||
|
|
||
| # https://stackoverflow.com/questions/13685386/matplotlib-equal-unit-length-with-equal-aspect-ratio-z-axis-is-not-equal-to |
There was a problem hiding this comment.
This script has old crud in it, and no explanation of what it does :)
Is it honestly something that another user could pick up and use?
| file_id = int(sys.argv[1]) | ||
|
|
||
|
|
||
|
|
There was a problem hiding this comment.
here and in other places: I dislike uploading files to the analysator repo where there are hardcoded input or output file names or directories. I suggest you insert boilerplate argument handling.
So this could go in sometime!
Plenty of stuff, used in the XO paper.
pyCalculations/null_lines.py contains wraps the FOTE method - it has some extra outputs for plotting the estimated null lines, which are then discarded in the corresponding datareducers. This function also has a vectorized golden ratio optimizer, which could be its own script... some time.
Plenty of reducers for MGA and MDD, and these guys could use the long-discussed reducer caching... but that can happen sometime in the future as well.
Contains the PR:
https://github.com/fmihpc/analysator/pull/232/files