-
Notifications
You must be signed in to change notification settings - Fork 4
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
diffusivity and potential sinks demo
- Loading branch information
François Laurent
committed
Mar 1, 2018
1 parent
fad5c2d
commit 1d1eb67
Showing
1 changed file
with
108 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,108 @@ | ||
|
||
import os | ||
import sys | ||
from tramway.inference import DiffusivityWarning, distributed | ||
from tramway.helper import * | ||
from tramway.helper.simulation import * | ||
import warnings | ||
#warnings.simplefilter('error') | ||
import numpy as np | ||
from math import * | ||
|
||
short_description = 'generate trajectories and infer diffusivity and potential maps' | ||
|
||
|
||
method = 'gwr' | ||
localization_error = 0.001 | ||
priorD = 0.01 | ||
priorV = 0.01 | ||
minD = -localization_error | ||
|
||
|
||
dim = 2 | ||
D0 = .5 | ||
D = .2 | ||
normGradV = 5. | ||
name = 'sinks' | ||
|
||
|
||
def main(): | ||
output_basename = name | ||
def out(label, extension): | ||
return '{}.{}.{}.{}'.format(output_basename, method, label, extension) | ||
|
||
xyt_file = output_basename + '.trxyt' | ||
rwa_file = output_basename + '.rwa' | ||
new_xyt = not os.path.exists(xyt_file) | ||
new_tessellation = not os.path.isfile(rwa_file) | ||
|
||
## define the ground truth (xyt_file) | ||
d_area_center = np.full((dim,), .8) | ||
d_area_radius = .15 | ||
def diffusivity_map(x, *args): | ||
d = x - d_area_center | ||
d = np.dot(d, d) | ||
return D if d <= d_area_radius * d_area_radius else D0 | ||
v_area_center= np.full((dim,), .4) | ||
v_area_radius = .2 | ||
v_ring_width = .06 | ||
def force_map(x, *args): | ||
f = v_area_center - x | ||
d = np.sqrt(np.dot(f, f)) | ||
f *= normGradV / d | ||
return f if abs(d - v_area_radius) <= v_ring_width else np.zeros(dim) | ||
if new_xyt: | ||
map_lower_bound = np.zeros(dim) | ||
map_upper_bound = np.full((dim,), 1.) | ||
# simulate random walks | ||
print('generating trajectories: {}'.format(xyt_file)) | ||
df = random_walk(diffusivity_map, force_map, \ | ||
trajectory_mean_count = 200, turnover = .3, duration = 10, \ | ||
box = np.r_[map_lower_bound, map_upper_bound - map_lower_bound]) | ||
#print(df) | ||
df.to_csv(xyt_file, sep="\t", header=False) | ||
## mesh regularly to sample ground truth for illustrative purposes | ||
#grid = tessellate(df, method='grid', min_location_count=10) | ||
if not new_tessellation: | ||
print("WARNING: tessellation will overwrite file '{}'".format(rwa_file)) | ||
new_tessellation = True | ||
|
||
## tessellate (tessellation_file) | ||
if new_tessellation: | ||
cells = tessellate(xyt_file, method, output_file=rwa_file, \ | ||
verbose=True, strict_min_location_count=10, force=True) | ||
cell_plot(rwa_file, output_file=out('mesh', 'png'), \ | ||
show=True, aspect='equal') | ||
# show ground truth | ||
true_map = distributed(cells).run(truth, diffusivity=diffusivity_map, force=force_map) | ||
print('ploting ground truth maps: {}'.format(out('truth', 'png'))) | ||
map_plot(true_map, cells=cells, output_file=out('truth', 'png'), show=True, aspect='equal') | ||
|
||
## infer and plot | ||
# capture negative diffusivity warnings and turn them into exceptions | ||
warnings.filterwarnings('error', '', DiffusivityWarning) | ||
|
||
#print("running D inference mode...") | ||
#D_ = infer(rwa_file, mode='D', localization_error=localization_error, \ | ||
# min_diffusivity=minD, output_label='D') | ||
#map_plot(D_, output_file=out('d', 'png'), show=True, aspect='equal') | ||
|
||
print("running DF inference mode...") | ||
DF = infer(rwa_file, mode='DF', localization_error=localization_error, output_label='DF') | ||
map_plot(DF, output_file=out('df', 'png'), show=True, aspect='equal') | ||
|
||
#print("running DD inference mode...") | ||
#DD = infer(rwa_file, mode='DD', localization_error=localization_error, \ | ||
# priorD=priorD, min_diffusivity=minD, output_label='DD') | ||
#map_plot(DD, output_file=out('dd', 'png'), show=True, aspect='equal') | ||
|
||
print("running DV inference mode...") | ||
DV = infer(rwa_file, mode='DV', localization_error=localization_error, \ | ||
priorD=priorD, priorV=priorV, output_label='DV') | ||
map_plot(DV, output_file=out('dv', 'png'), show=True, aspect='equal') | ||
|
||
sys.exit(0) | ||
|
||
|
||
if __name__ == '__main__': | ||
main() |