Use separate fixture data for Python tests

This commit is contained in:
Javier Goizueta 2016-02-18 12:59:48 +01:00
parent 53a868be2e
commit 0833b2f226
5 changed files with 241 additions and 20 deletions

View File

@ -1,3 +1,4 @@
-- Moran's I
CREATE OR REPLACE FUNCTION CREATE OR REPLACE FUNCTION
cdb_moran_local ( cdb_moran_local (
t TEXT, t TEXT,
@ -7,12 +8,28 @@ CREATE OR REPLACE FUNCTION
permutations INT DEFAULT 99, permutations INT DEFAULT 99,
geom_column TEXT DEFAULT 'the_geom', geom_column TEXT DEFAULT 'the_geom',
id_col TEXT DEFAULT 'cartodb_id', id_col TEXT DEFAULT 'cartodb_id',
w_type TEXT DEFAULT 'knn', w_type TEXT DEFAULT 'knn')
random_seed INTEGER DEFAULT NULL
)
RETURNS TABLE (moran FLOAT, quads TEXT, significance FLOAT, ids INT) RETURNS TABLE (moran FLOAT, quads TEXT, significance FLOAT, ids INT)
AS $$ AS $$
from crankshaft.clustering import moran_local from crankshaft.clustering import moran_local
# TODO: use named parameters or a dictionary # TODO: use named parameters or a dictionary
return moran_local(t, attr, significance, num_ngbrs, permutations, geom_column, id_col, w_type, random_seed) return moran_local(t, attr, significance, num_ngbrs, permutations, geom_column, id_col, w_type)
$$ LANGUAGE plpythonu;
-- Moran's I Local Rate
CREATE OR REPLACE FUNCTION
cdb_moran_local_rate(t TEXT,
numerator TEXT,
denominator TEXT,
significance FLOAT DEFAULT 0.05,
num_ngbrs INT DEFAULT 5,
permutations INT DEFAULT 99,
geom_column TEXT DEFAULT 'the_geom',
id_col TEXT DEFAULT 'cartodb_id',
w_type TEXT DEFAULT 'knn')
RETURNS TABLE(moran FLOAT, quads TEXT, significance FLOAT, ids INT, y numeric)
AS $$
from crankshaft.clustering import moran_local_rate
# TODO: use named parameters or a dictionary
return moran_local_rate(t, numerator, denominator, significance, num_ngbrs, permutations, geom_column, id_col, w_type)
$$ LANGUAGE plpythonu; $$ LANGUAGE plpythonu;

View File

@ -54,6 +54,95 @@ def moran_local(t, attr, significance, num_ngbrs, permutations, geom_column, id_
return zip(lisa.Is, lisa_sig, lisa.p_sim, w.id_order) return zip(lisa.Is, lisa_sig, lisa.p_sim, w.id_order)
def moran_local_rate(t, numerator, denominator, significance, num_ngbrs, permutations, geom_column, id_col, w_type):
"""
Moran's I Local Rate
Andy Eschbacher
"""
plpy.notice('** Constructing query')
# geometries with attributes that are null are ignored
# resulting in a collection of not as near neighbors
qvals = {"id_col": id_col,
"numerator": numerator,
"denominator": denominator,
"geom_col": geom_column,
"table": t,
"num_ngbrs": num_ngbrs}
q = get_query(w_type, qvals)
try:
r = plpy.execute(q)
plpy.notice('** Query returned with %d rows' % len(r))
except plpy.SPIError:
plpy.notice('** Query failed: "%s"' % q)
plpy.notice('** Error: %s' % plpy.SPIError)
plpy.notice('** Exiting function')
return zip([None], [None], [None], [None])
plpy.notice('r.nrows() = %d' % r.nrows())
## collect attributes
numer = get_attributes(r, 1)
denom = get_attributes(r, 2)
w = get_weight(r, w_type, num_ngbrs)
# calculate LISA values
lisa = ps.esda.moran.Moran_Local_Rate(numer, denom, w, permutations=permutations)
# find units of significance
lisa_sig = lisa_sig_vals(lisa.p_sim, lisa.q, significance)
plpy.notice('** Finished calculations')
## TODO: Decide on which return values here
return zip(lisa.Is, lisa_sig, lisa.p_sim, w.id_order, lisa.y)
def moran_local_bv(t, attr1, attr2, significance, num_ngbrs, permutations, geom_column, id_col, w_type):
plpy.notice('** Constructing query')
qvals = {"num_ngbrs": num_ngbrs,
"attr1": attr1,
"attr2": attr2,
"table": t,
"geom_col": geom_column,
"id_col": id_col}
q = get_query(w_type, qvals)
try:
r = plpy.execute(q)
plpy.notice('** Query returned with %d rows' % len(r))
except plpy.SPIError:
plpy.notice('** Query failed: "%s"' % q)
plpy.notice('** Error: %s' % plpy.SPIError)
plpy.notice('** Exiting function')
return zip([None], [None], [None], [None])
## collect attributes
attr1_vals = get_attributes(r, 1)
attr2_vals = get_attributes(r, 2)
# create weights
w = get_weight(r, w_type, num_ngbrs)
# calculate LISA values
lisa = ps.esda.moran.Moran_Local_BV(attr1_vals, attr2_vals, w)
plpy.notice("len of Is: %d" % len(lisa.Is))
# find clustering of significance
lisa_sig = lisa_sig_vals(lisa.p_sim, lisa.q, significance)
plpy.notice('** Finished calculations')
return zip(lisa.Is, lisa_sig, lisa.p_sim, w.id_order)
# Low level functions ---------------------------------------- # Low level functions ----------------------------------------
def map_quads(coord): def map_quads(coord):

View File

@ -0,0 +1,52 @@
[[0.9319096128346788, "HH"],
[-1.135787401862846, "HL"],
[0.11732030672508517, "Not significant"],
[0.6152779669180425, "Not significant"],
[-0.14657336660125297, "Not significant"],
[0.6967858120189607, "Not significant"],
[0.07949310115714454, "Not significant"],
[0.4703198759258987, "Not significant"],
[0.4421125200498064, "Not significant"],
[0.5724288737143592, "Not significant"],
[0.8970743435692062, "LL"],
[0.18327334401918674, "Not significant"],
[-0.01466729201304962, "Not significant"],
[0.3481559372544409, "Not significant"],
[0.06547094736902978, "Not significant"],
[0.15482141569329988, "HH"],
[0.4373841193538136, "Not significant"],
[0.15971286468915544, "Not significant"],
[1.0543588860308968, "Not significant"],
[1.7372866900020818, "HH"],
[1.091998586053999, "LL"],
[0.1171572584252222, "Not significant"],
[0.08438455015300014, "Not significant"],
[0.06547094736902978, "Not significant"],
[0.15482141569329985, "HH"],
[1.1627044812890683, "HH"],
[0.06547094736902978, "Not significant"],
[0.795275137550483, "Not significant"],
[0.18562939195219, "LL"],
[0.3010757406693439, "Not significant"],
[2.8205795942839376, "HH"],
[0.11259190602909264, "Not significant"],
[-0.07116352791516614, "Not significant"],
[-0.09945240794119009, "Not significant"],
[0.18562939195219, "LL"],
[0.1832733440191868, "Not significant"],
[-0.39054253768447705, "Not significant"],
[-0.1672071289487642, "HL"],
[0.3337669247916343, "Not significant"],
[0.2584386102554792, "Not significant"],
[-0.19733845476322634, "HL"],
[-0.9379282899805409, "LH"],
[-0.028770969951095866, "Not significant"],
[0.051367269430983485, "Not significant"],
[-0.2172548045913472, "LH"],
[0.05136726943098351, "Not significant"],
[0.04191046803899837, "Not significant"],
[0.7482357030403517, "HH"],
[-0.014585767863118111, "Not significant"],
[0.5410013139159929, "Not significant"],
[1.0223932668429925, "LL"],
[1.4179402898927476, "LL"]]

View File

@ -0,0 +1,54 @@
[
{"neighbors": [48, 26, 20, 9, 31], "id": 1, "value": 0.5},
{"neighbors": [30, 16, 46, 3, 4], "id": 2, "value": 0.7},
{"neighbors": [46, 30, 2, 12, 16], "id": 3, "value": 0.2},
{"neighbors": [18, 30, 23, 2, 52], "id": 4, "value": 0.1},
{"neighbors": [47, 40, 45, 37, 28], "id": 5, "value": 0.3},
{"neighbors": [10, 21, 41, 14, 37], "id": 6, "value": 0.05},
{"neighbors": [8, 17, 43, 25, 12], "id": 7, "value": 0.4},
{"neighbors": [17, 25, 43, 22, 7], "id": 8, "value": 0.7},
{"neighbors": [39, 34, 1, 26, 48], "id": 9, "value": 0.5},
{"neighbors": [6, 37, 5, 45, 49], "id": 10, "value": 0.04},
{"neighbors": [51, 41, 29, 21, 14], "id": 11, "value": 0.08},
{"neighbors": [44, 46, 43, 50, 3], "id": 12, "value": 0.2},
{"neighbors": [45, 23, 14, 28, 18], "id": 13, "value": 0.4},
{"neighbors": [41, 29, 13, 23, 6], "id": 14, "value": 0.2},
{"neighbors": [36, 27, 32, 33, 24], "id": 15, "value": 0.3},
{"neighbors": [19, 2, 46, 44, 28], "id": 16, "value": 0.4},
{"neighbors": [8, 25, 43, 7, 22], "id": 17, "value": 0.6},
{"neighbors": [23, 4, 29, 14, 13], "id": 18, "value": 0.3},
{"neighbors": [42, 16, 28, 26, 40], "id": 19, "value": 0.7},
{"neighbors": [1, 48, 31, 26, 42], "id": 20, "value": 0.8},
{"neighbors": [41, 6, 11, 14, 10], "id": 21, "value": 0.1},
{"neighbors": [25, 50, 43, 31, 44], "id": 22, "value": 0.4},
{"neighbors": [18, 13, 14, 4, 2], "id": 23, "value": 0.1},
{"neighbors": [33, 49, 34, 47, 27], "id": 24, "value": 0.3},
{"neighbors": [43, 8, 22, 17, 50], "id": 25, "value": 0.4},
{"neighbors": [1, 42, 20, 31, 48], "id": 26, "value": 0.6},
{"neighbors": [32, 15, 36, 33, 24], "id": 27, "value": 0.3},
{"neighbors": [40, 45, 19, 5, 13], "id": 28, "value": 0.8},
{"neighbors": [11, 51, 41, 14, 18], "id": 29, "value": 0.3},
{"neighbors": [2, 3, 4, 46, 18], "id": 30, "value": 0.1},
{"neighbors": [20, 26, 1, 50, 48], "id": 31, "value": 0.9},
{"neighbors": [27, 36, 15, 49, 24], "id": 32, "value": 0.3},
{"neighbors": [24, 27, 49, 34, 32], "id": 33, "value": 0.4},
{"neighbors": [47, 9, 39, 40, 24], "id": 34, "value": 0.3},
{"neighbors": [38, 51, 11, 21, 41], "id": 35, "value": 0.3},
{"neighbors": [15, 32, 27, 49, 33], "id": 36, "value": 0.2},
{"neighbors": [49, 10, 5, 47, 24], "id": 37, "value": 0.5},
{"neighbors": [35, 21, 51, 11, 41], "id": 38, "value": 0.4},
{"neighbors": [9, 34, 48, 1, 47], "id": 39, "value": 0.6},
{"neighbors": [28, 47, 5, 9, 34], "id": 40, "value": 0.5},
{"neighbors": [11, 14, 29, 21, 6], "id": 41, "value": 0.4},
{"neighbors": [26, 19, 1, 9, 31], "id": 42, "value": 0.2},
{"neighbors": [25, 12, 8, 22, 44], "id": 43, "value": 0.3},
{"neighbors": [12, 50, 46, 16, 43], "id": 44, "value": 0.2},
{"neighbors": [28, 13, 5, 40, 19], "id": 45, "value": 0.3},
{"neighbors": [3, 12, 44, 2, 16], "id": 46, "value": 0.2},
{"neighbors": [34, 40, 5, 49, 24], "id": 47, "value": 0.3},
{"neighbors": [1, 20, 26, 9, 39], "id": 48, "value": 0.5},
{"neighbors": [24, 37, 47, 5, 33], "id": 49, "value": 0.2},
{"neighbors": [44, 22, 31, 42, 26], "id": 50, "value": 0.6},
{"neighbors": [11, 29, 41, 14, 21], "id": 51, "value": 0.01},
{"neighbors": [4, 18, 29, 51, 23], "id": 52, "value": 0.01}
]

View File

@ -11,9 +11,9 @@ import unittest
# sys.modules['plpy'] = plpy # sys.modules['plpy'] = plpy
from helper import plpy from helper import plpy
# import crankshaft.clustering as cc
import crankshaft.clustering as cc import crankshaft.clustering as cc
from crankshaft import random_seeds
import json
class MoranTest(unittest.TestCase): class MoranTest(unittest.TestCase):
"""Testing class for Moran's I functions.""" """Testing class for Moran's I functions."""
@ -26,6 +26,8 @@ class MoranTest(unittest.TestCase):
"table": "a_list", "table": "a_list",
"geom_col": "the_geom", "geom_col": "the_geom",
"num_ngbrs": 321} "num_ngbrs": 321}
self.neighbors_data = json.loads(open('test/fixtures/neighbors.json').read())
self.moran_data = json.loads(open('test/fixtures/moran.json').read())
def test_map_quads(self): def test_map_quads(self):
"""Test map_quads.""" """Test map_quads."""
@ -119,17 +121,24 @@ class MoranTest(unittest.TestCase):
self.assertTrue((test_ans == ans).all()) self.assertTrue((test_ans == ans).all())
def test_moran_local(self): def test_moran_local(self):
"""Test Moran's I local""" """Test Moran's I local"""
plpy._define_result('select', [ data = [ { 'id': d['id'], 'attr1': d['value'], 'neighbors': d['neighbors'] } for d in self.neighbors_data]
{ 'id': 1, 'attr1': 100.0, 'neighbors': [2,4,5,7,8] }, plpy._define_result('select', data)
{ 'id': 2, 'attr1': 110.0, 'neighbors': [1,4,5,6,7] }, random_seeds.set_random_seeds(1234)
{ 'id': 3, 'attr1': 90.0, 'neighbors': [1,4,5,7,8] }, result = cc.moran_local('table', 'value', 0.05, 5, 99, 'the_geom', 'cartodb_id', 'knn')
{ 'id': 4, 'attr1': 100.0, 'neighbors': [1,2,5,7,8] }, result = [(row[0], row[1]) for row in result]
{ 'id': 5, 'attr1': 100.0, 'neighbors': [1,2,3,7,8] }, expected = self.moran_data
{ 'id': 6, 'attr1': 105.0, 'neighbors': [1,2,3,7,8] }, for ([res_val, res_quad], [exp_val, exp_quad]) in zip(result, expected):
{ 'id': 7, 'attr1': 105.0, 'neighbors': [1,2,3,6,8] }, self.assertAlmostEqual(res_val, exp_val)
{ 'id': 8, 'attr1': 105.0, 'neighbors': [1,2,3,6,7] }, self.assertEqual(res_quad, exp_quad)
{ 'id': 9, 'attr1': 120.0, 'neighbors': [1,2,5,6,7] }
]) def test_moran_local_rate(self):
result = cc.moran_local('table', 'value', 0.05, 5, 99, 'the_geom', 'cartodb_id', 'knn') """Test Moran's I rate"""
# TODO: check results! data = [ { 'id': d['id'], 'attr1': d['value'], 'attr2': 1, 'neighbors': d['neighbors'] } for d in self.neighbors_data]
plpy._define_result('select', data)
random_seeds.set_random_seeds(1234)
result = cc.moran_local_rate('table', 'numerator', 'denominator', 0.05, 5, 99, 'the_geom', 'cartodb_id', 'knn')
result = [(row[0], row[1]) for row in result]
expected = self.moran_data
for ([res_val, res_quad], [exp_val, exp_quad]) in zip(result, expected):
self.assertAlmostEqual(res_val, exp_val)