pip install -U cmflow
On Windows, Shapely and Rtree are easier to be installed by using Christoph Gohlke's non-official build:
- https://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely
- https://www.lfd.uci.edu/~gohlke/pythonlibs/#rtree
descartes can be installed on all platform by:
pip install descartes
On Linux (Ubuntu shown here) these can be installed via apt-get:
sudo apt-get install -y python-shapely
sudo apt-get install -y python-rtree
sudo apt-get install -y python-descartes
Creates BMStats that can be used later, from Leapfrog Geology:
# (ONLY ONCE) geo used to get geology from Leapfrog geological model
cmgeo = mulgrid('g_very_fine.dat')
# CSV file created by Leapfrog using cmgeo above
leapfrog = LeapfrogGM()
leapfrog.import_leapfrog_csv('grid_gtmp_ay2017_03_6_fit.csv')
cm_geology = CM_Blocky(cmgeo, leapfrog)
# whatever active model we are working on
bmgeo = mulgrid('gwaixx_yy.dat')
bms_geology = cm_geology.calc_bmstats(bm_geo)
bms_geology.save('a.json')
A BMStats object can be reused (very fast) to eg.
bms_geology = BMStats('a.json')
# get a cell's stats
cs = bms_geology.cellstats['abc12']
# rock that occupies most in cell 'abc12'
rock_name = bms_geology.zones[np.argmax(cs)]
# how many rock in cell 'abc12'
n_rock = len(np.nonzero(cs))
# list all rocks in cell 'abc12'
rocks = [bm_geology.zones[i] for i in np.nonzero(cs)]
# find all blocks intersect with the zone
blocks, ratios = bm_geology.blocks_in_zone('BASE1')
block_idx, ratios = bm_geology.blocks_in_zone('BASE1', indices=True)
This is the object that we keep for later use. It is associated to a certain "geometry" file. So each cell has information on zones. Usually this is generated by cm.populate_model(), which can be expensive.
- ? should I call it CMStats?
- ? TODO, .cellstats access by cell index
- ? TODO, .
Base Model Stats, mainly numpy arrays with rows corresponding to mulgrid blocks, and columns corresponding to zones. Each is a value, usually between 0.0 and 1.0. Often 1.0 is indicating that particular block is fully within the zone.
.stats numpy array (n,m), n = num of model blocks, m = num of zones .zones list of zone names (str) .zonestats dict of stats column by zone names .cellstats dict of stats row by block name
6 elements, 3 zones
A B C
0.0, 0.7, 0.3, -> row sum to 1.0, element 0, 0.7 rock B, 0.3 rock C
1.0, 0.0, 0.0,
1.0, 0.0, 0.0,
0.0, 0.5, 0.5,
0.1, 0.2, 0.7,
0.0, 1.0, 0.0,
(this is only one way of using it, such as a rocktype)
.stats, numpy array (n * m), n number of geometry cells, m number of zones .zones, a list of zone name, eg. geology rock names, fault names etc .zonestats, a dict keyed by zone name, an array of size number of cells, each cell is between .cellstats, a dict of stats by cell name
.save() .load() .add_stats() add another bmstat, merge stats .add_cm() calls cm.populate_model, and merge stats
These are the objects that can be created in order to create the final BMStats objects. The common method .populate_model(bm_geo) is called to create BMStats objects. It means the conceptual model is "applied" onto the bm_geo.
- TODO, .populate_model() should return BMStats instead
- ? TODO, .populate_model() should be called something else?
.populate_model(bm_geo) takes a target geometry, and return/creates BMStats