cartodb4/crankshaft
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge branch 'develop' into moran-query-ordering-fix

Browse Source
This commit is contained in:
Andy Eschbacher
2016-09-23 13:25:36 -04:00
parent 1148aa417a c7bb9698a9
commit 07e4062237
114 changed files with 18496 additions and 82 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/pg/crankshaft.control
+1 -1
View File
@@ -1,5 +1,5 @@
comment = 'CartoDB Spatial Analysis extension'
default_version = '0.3.1'
default_version = '0.4.2'
requires = 'plpythonu, postgis'
superuser = true
schema = cdb_crankshaft
src/pg/sql/08_interpolation.sql
+19 -5
View File
@@ -1,6 +1,8 @@
-- 0: nearest neighbor
-- 0: nearest neighbor(s)
-- 1: barymetric
-- 2: IDW
-- 3: krigin ---> TO DO
CREATE OR REPLACE FUNCTION CDB_SpatialInterpolation(
IN query text,
@@ -50,12 +52,19 @@ DECLARE
vc numeric;
output numeric;
BEGIN
output := -999.999;
-- nearest
-- output := -999.999;
-- nearest neighbors
-- p1: limit the number of neighbors, 0-> closest one
IF method = 0 THEN
WITH a as (SELECT unnest(geomin) as g, unnest(colin) as v)
SELECT a.v INTO output FROM a ORDER BY point<->a.g LIMIT 1;
IF p1 = 0 THEN
p1 := 1;
END IF;
WITH a as (SELECT unnest(geomin) as g, unnest(colin) as v),
b as (SELECT a.v as v FROM a ORDER BY point<->a.g LIMIT p1::integer)
SELECT avg(b.v) INTO output FROM b;
RETURN output;
-- barymetric
@@ -121,6 +130,11 @@ BEGIN
SELECT sum(b.f)/sum(b.k) INTO output FROM b;
RETURN output;
-- krigin
ELSIF method = 3 THEN
-- TO DO
END IF;
RETURN -777.777;
src/pg/sql/10_moran.sql
+1 -1
View File
@@ -10,7 +10,7 @@ CREATE OR REPLACE FUNCTION
id_col TEXT DEFAULT 'cartodb_id')
RETURNS TABLE (moran NUMERIC, significance NUMERIC)
AS $$
from crankshaft.clustering import moran_local
from crankshaft.clustering import moran
# TODO: use named parameters or a dictionary
return moran(subquery, column_name, w_type, num_ngbrs, permutations, geom_col, id_col)
$$ LANGUAGE plpythonu;
src/pg/sql/13_PIA.sql
+123
View File
@@ -0,0 +1,123 @@
-- Based on:
-- https://github.com/mapbox/polylabel/blob/master/index.js
-- https://sites.google.com/site/polesofinaccessibility/
-- Requires: https://github.com/CartoDB/cartodb-postgresql
-- Based on:
-- https://github.com/mapbox/polylabel/blob/master/index.js
-- https://sites.google.com/site/polesofinaccessibility/
-- Requires: https://github.com/CartoDB/cartodb-postgresql
CREATE OR REPLACE FUNCTION CDB_PIA(
IN polygon geometry,
IN tolerance numeric DEFAULT 1.0
)
RETURNS geometry AS $$
DECLARE
env geometry[];
cells geometry[];
cell geometry;
best_c geometry;
best_d numeric;
test_d numeric;
test_mx numeric;
test_h numeric;
test_cells geometry[];
width numeric;
height numeric;
h numeric;
i integer;
n integer;
sqr numeric;
p geometry;
BEGIN
sqr := |/2;
polygon := ST_Transform(polygon, 3857);
-- grid #0 cell size
height := ST_YMax(polygon) - ST_YMin(polygon);
width := ST_XMax(polygon) - ST_XMin(polygon);
h := 0.5*LEAST(height, width);
-- grid #0
with c1 as(
SELECT cdb_crankshaft.CDB_RectangleGrid(polygon, h, h) as c
)
SELECT array_agg(c) INTO cells FROM c1;
-- 1st guess: centroid
best_d := cdb_crankshaft._Signed_Dist(polygon, ST_Centroid(Polygon));
-- looping the loop
n := array_length(cells,1);
i := 1;
LOOP
EXIT WHEN i > n;
cell := cells[i];
i := i+1;
-- cell side size, it's square
test_h := ST_XMax(cell) - ST_XMin(cell) ;
-- check distance
test_d := cdb_crankshaft._Signed_Dist(polygon, ST_Centroid(cell));
IF test_d > best_d THEN
best_d := test_d;
best_c := cells[i];
END IF;
-- longest distance within the cell
test_mx := test_d + (test_h/2 * sqr);
-- if the cell has no chance to contains the desired point, continue
CONTINUE WHEN test_mx - best_d <= tolerance;
-- resample the cell
with c1 as(
SELECT cdb_crankshaft.CDB_RectangleGrid(cell, test_h/2, test_h/2) as c
)
SELECT array_agg(c) INTO test_cells FROM c1;
-- concat the new cells to the former array
cells := cells || test_cells;
-- prepare next iteration
n := array_length(cells,1);
END LOOP;
RETURN ST_transform(ST_Centroid(best_c), 4326);
END;
$$ language plpgsql IMMUTABLE;
-- signed distance point to polygon with holes
-- negative is the point is out the polygon
CREATE OR REPLACE FUNCTION _Signed_Dist(
IN polygon geometry,
IN point geometry
)
RETURNS numeric AS $$
DECLARE
i integer;
within integer;
holes integer;
dist numeric;
BEGIN
dist := 1e999;
SELECT LEAST(dist, ST_distance(point, ST_ExteriorRing(polygon))::numeric) INTO dist;
SELECT CASE WHEN ST_Within(point,polygon) THEN 1 ELSE -1 END INTO within;
SELECT ST_NumInteriorRings(polygon) INTO holes;
IF holes > 0 THEN
FOR i IN 1..holes
LOOP
SELECT LEAST(dist, ST_distance(point, ST_InteriorRingN(polygon, i))::numeric) INTO dist;
END LOOP;
END IF;
dist := dist * within::numeric;
RETURN dist;
END;
$$ language plpgsql IMMUTABLE;
src/pg/sql/14_densify.sql
+67
View File
@@ -0,0 +1,67 @@
--
-- Iterative densification of a set of points using Delaunay triangulation
-- the new points have as assigned value the average value of the 3 vertex (centroid)
--
-- @param geomin - array of geometries (points)
--
-- @param colin - array of numeric values in that points
--
-- @param iterations - integer, number of iterations
--
--
-- Returns: TABLE(geomout geometry, colout numeric)
--
--
CREATE OR REPLACE FUNCTION CDB_Densify(
IN geomin geometry[],
IN colin numeric[],
IN iterations integer
)
RETURNS TABLE(geomout geometry, colout numeric) AS $$
DECLARE
geotemp geometry[];
coltemp numeric[];
i integer;
gs geometry[];
g geometry;
vertex geometry[];
va numeric;
vb numeric;
vc numeric;
center geometry;
centerval numeric;
tmp integer;
BEGIN
geotemp := geomin;
coltemp := colin;
FOR i IN 1..iterations
LOOP
-- generate TIN
WITH a as (SELECT unnest(geotemp) AS e),
b as (SELECT ST_DelaunayTriangles(ST_Collect(a.e),0.001, 0) AS t FROM a),
c as (SELECT (ST_Dump(t)).geom AS v FROM b)
SELECT array_agg(v) INTO gs FROM c;
-- loop cells
FOREACH g IN ARRAY gs
LOOP
-- append centroid
SELECT ST_Centroid(g) INTO center;
geotemp := array_append(geotemp, center);
-- retrieve the value of each vertex
WITH a AS (SELECT (ST_DumpPoints(g)).geom AS v)
SELECT array_agg(v) INTO vertex FROM a;
WITH a AS(SELECT unnest(geotemp) as geo, unnest(coltemp) as c)
SELECT c INTO va FROM a WHERE ST_Equals(geo, vertex[1]);
WITH a AS(SELECT unnest(geotemp) as geo, unnest(coltemp) as c)
SELECT c INTO vb FROM a WHERE ST_Equals(geo, vertex[2]);
WITH a AS(SELECT unnest(geotemp) as geo, unnest(coltemp) as c)
SELECT c INTO vc FROM a WHERE ST_Equals(geo, vertex[3]);
-- calc the value at the center
centerval := (va + vb + vc) / 3;
-- append the value
coltemp := array_append(coltemp, centerval);
END LOOP;
END LOOP;
RETURN QUERY SELECT unnest(geotemp ) as geomout, unnest(coltemp ) as colout;
END;
$$ language plpgsql IMMUTABLE;
src/pg/sql/15_tinmap.sql
+43
View File
@@ -0,0 +1,43 @@
CREATE OR REPLACE FUNCTION CDB_TINmap(
IN geomin geometry[],
IN colin numeric[],
IN iterations integer
)
RETURNS TABLE(geomout geometry, colout numeric) AS $$
DECLARE
p geometry[];
vals numeric[];
gs geometry[];
g geometry;
vertex geometry[];
centerval numeric;
va numeric;
vb numeric;
vc numeric;
coltemp numeric[];
BEGIN
SELECT array_agg(dens.geomout), array_agg(dens.colout) INTO p, vals FROM cdb_crankshaft.CDB_Densify(geomin, colin, iterations) dens;
WITH a as (SELECT unnest(p) AS e),
b as (SELECT ST_DelaunayTriangles(ST_Collect(a.e),0.001, 0) AS t FROM a),
c as (SELECT (ST_Dump(t)).geom AS v FROM b)
SELECT array_agg(v) INTO gs FROM c;
FOREACH g IN ARRAY gs
LOOP
-- retrieve the vertex of each triangle
WITH a AS (SELECT (ST_DumpPoints(g)).geom AS v)
SELECT array_agg(v) INTO vertex FROM a;
-- retrieve the value of each vertex
WITH a AS(SELECT unnest(p) as geo, unnest(vals) as c)
SELECT c INTO va FROM a WHERE ST_Equals(geo, vertex[1]);
WITH a AS(SELECT unnest(p) as geo, unnest(vals) as c)
SELECT c INTO vb FROM a WHERE ST_Equals(geo, vertex[2]);
WITH a AS(SELECT unnest(p) as geo, unnest(vals) as c)
SELECT c INTO vc FROM a WHERE ST_Equals(geo, vertex[3]);
-- calc the value at the center
centerval := (va + vb + vc) / 3;
-- append the value
coltemp := array_append(coltemp, centerval);
END LOOP;
RETURN QUERY SELECT unnest(gs) as geomout, unnest(coltemp ) as colout;
END;
$$ language plpgsql IMMUTABLE;
src/pg/sql/19_contour.sql
+208
View File
@@ -0,0 +1,208 @@
CREATE OR REPLACE FUNCTION CDB_Contour(
IN geomin geometry[],
IN colin numeric[],
IN buffer numeric,
IN intmethod integer,
IN classmethod integer,
IN steps integer,
IN max_time integer DEFAULT 60000
)
RETURNS TABLE(
the_geom geometry,
bin integer,
min_value numeric,
max_value numeric,
avg_value numeric
) AS $$
DECLARE
cell_count integer;
tin geometry[];
resolution integer;
BEGIN
-- nasty trick to override issue #121
IF max_time = 0 THEN
max_time = -90;
END IF;
resolution := max_time;
max_time := -1 * resolution;
-- calc the optimal number of cells for the current dataset
SELECT
CASE intmethod
WHEN 0 THEN round(3.7745903782 * max_time - 9.4399210051 * array_length(geomin,1) - 1350.8778213073)
WHEN 1 THEN round(2.2855592156 * max_time - 87.285217133 * array_length(geomin,1) + 17255.7085601797)
WHEN 2 THEN round(0.9799471999 * max_time - 127.0334085369 * array_length(geomin,1) + 22707.9579721218)
ELSE 10000
END INTO cell_count;
-- we don't have iterative barycentric interpolation in CDB_interpolation,
-- and it's a costy function, so let's make a custom one here till
-- we update the code
-- tin := ARRAY[]::geometry[];
IF intmethod=1 THEN
WITH
a as (SELECT unnest(geomin) AS e),
b as (SELECT ST_DelaunayTriangles(ST_Collect(a.e),0.001, 0) AS t FROM a),
c as (SELECT (ST_Dump(t)).geom as v FROM b)
SELECT array_agg(v) INTO tin FROM c;
END IF;
-- Delaunay stuff performed just ONCE!!
-- magic
RETURN QUERY
WITH
convexhull as (
SELECT
ST_ConvexHull(ST_Collect(geomin)) as g,
buffer * |/ st_area(ST_ConvexHull(ST_Collect(geomin)))/PI() as r
),
envelope as (
SELECT
st_expand(a.g, a.r) as e
FROM convexhull a
),
envelope3857 as(
SELECT
ST_Transform(e, 3857) as geom
FROM envelope
),
resolution as(
SELECT
CASE WHEN resolution <= 0 THEN
round(|/ (
ST_area(geom) / abs(cell_count)
))
ELSE
resolution
END AS cell
FROM envelope3857
),
grid as(
SELECT
ST_Transform(cdb_crankshaft.CDB_RectangleGrid(e.geom, r.cell, r.cell), 4326) as geom
FROM envelope3857 e, resolution r
),
interp as(
SELECT
geom,
CASE
WHEN intmethod=1 THEN cdb_crankshaft._interp_in_tin(geomin, colin, tin, ST_Centroid(geom))
ELSE cdb_crankshaft.CDB_SpatialInterpolation(geomin, colin, ST_Centroid(geom), intmethod)
END as val
FROM grid
),
classes as(
SELECT CASE
WHEN classmethod = 0 THEN
cdb_crankshaft.CDB_EqualIntervalBins(array_agg(val), steps)
WHEN classmethod = 1 THEN
cdb_crankshaft.CDB_HeadsTailsBins(array_agg(val), steps)
WHEN classmethod = 2 THEN
cdb_crankshaft.CDB_JenksBins(array_agg(val), steps)
ELSE
cdb_crankshaft.CDB_QuantileBins(array_agg(val), steps)
END as b
FROM interp
where val is not null
),
classified as(
SELECT
i.*,
width_bucket(i.val, c.b) as bucket
FROM interp i left join classes c
ON 1=1
),
classified2 as(
SELECT
geom,
val,
CASE
WHEN bucket = steps THEN bucket - 1
ELSE bucket
END as b
FROM classified
),
final as(
SELECT
st_union(geom) as the_geom,
b as bin,
min(val) as min_value,
max(val) as max_value,
avg(val) as avg_value
FROM classified2
GROUP BY bin
)
SELECT
*
FROM final
where final.bin is not null
;
END;
$$ language plpgsql;
-- =====================================================================
-- Interp in grid, so we can use barycentric with a precalculated tin (NNI)
-- =====================================================================
CREATE OR REPLACE FUNCTION _interp_in_tin(
IN geomin geometry[],
IN colin numeric[],
IN tin geometry[],
IN point geometry
)
RETURNS numeric AS
$$
DECLARE
g geometry;
vertex geometry[];
sg numeric;
sa numeric;
sb numeric;
sc numeric;
va numeric;
vb numeric;
vc numeric;
output numeric;
BEGIN
-- get the cell the point is within
WITH
a as (SELECT unnest(tin) as v),
b as (SELECT v FROM a WHERE ST_Within(point, v))
SELECT v INTO g FROM b;
-- if we're out of the data realm,
-- return null
IF g is null THEN
RETURN null;
END IF;
-- vertex of the selected cell
WITH a AS (
SELECT (ST_DumpPoints(g)).geom AS v
)
SELECT array_agg(v) INTO vertex FROM a;
-- retrieve the value of each vertex
WITH a AS(SELECT unnest(geomin) as geo, unnest(colin) as c)
SELECT c INTO va FROM a WHERE ST_Equals(geo, vertex[1]);
WITH a AS(SELECT unnest(geomin) as geo, unnest(colin) as c)
SELECT c INTO vb FROM a WHERE ST_Equals(geo, vertex[2]);
WITH a AS(SELECT unnest(geomin) as geo, unnest(colin) as c)
SELECT c INTO vc FROM a WHERE ST_Equals(geo, vertex[3]);
-- calc the areas
SELECT
ST_area(g),
ST_area(ST_MakePolygon(ST_MakeLine(ARRAY[point, vertex[2], vertex[3], point]))),
ST_area(ST_MakePolygon(ST_MakeLine(ARRAY[point, vertex[1], vertex[3], point]))),
ST_area(ST_MakePolygon(ST_MakeLine(ARRAY[point,vertex[1],vertex[2], point]))) INTO sg, sa, sb, sc;
output := (coalesce(sa,0) * coalesce(va,0) + coalesce(sb,0) * coalesce(vb,0) + coalesce(sc,0) * coalesce(vc,0)) / coalesce(sg,1);
RETURN output;
END;
$$
language plpgsql;
src/pg/sql/cdb_utils.sql
+447
View File
@@ -0,0 +1,447 @@
--
-- Fill given extent with a rectangular coverage
--
-- @param ext Extent to fill. Only rectangles with center point falling
-- inside the extent (or at the lower or leftmost edge) will
-- be emitted. The returned hexagons will have the same SRID
-- as this extent.
--
-- @param width With of each rectangle
--
-- @param height Height of each rectangle
--
-- @param origin Optional origin to allow for exact tiling.
-- If omitted the origin will be 0,0.
-- The parameter is checked for having the same SRID
-- as the extent.
--
--
CREATE OR REPLACE FUNCTION CDB_RectangleGrid(ext GEOMETRY, width FLOAT8, height FLOAT8, origin GEOMETRY DEFAULT NULL)
RETURNS SETOF GEOMETRY
AS $$
DECLARE
h GEOMETRY; -- rectangle cell
hstep FLOAT8; -- horizontal step
vstep FLOAT8; -- vertical step
hw FLOAT8; -- half width
hh FLOAT8; -- half height
vstart FLOAT8;
hstart FLOAT8;
hend FLOAT8;
vend FLOAT8;
xoff FLOAT8;
yoff FLOAT8;
xgrd FLOAT8;
ygrd FLOAT8;
x FLOAT8;
y FLOAT8;
srid INTEGER;
BEGIN
srid := ST_SRID(ext);
xoff := 0;
yoff := 0;
IF origin IS NOT NULL THEN
IF ST_SRID(origin) != srid THEN
RAISE EXCEPTION 'SRID mismatch between extent (%) and origin (%)', srid, ST_SRID(origin);
END IF;
xoff := ST_X(origin);
yoff := ST_Y(origin);
END IF;
--RAISE DEBUG 'X offset: %', xoff;
--RAISE DEBUG 'Y offset: %', yoff;
hw := width/2.0;
hh := height/2.0;
xgrd := hw;
ygrd := hh;
--RAISE DEBUG 'X grid size: %', xgrd;
--RAISE DEBUG 'Y grid size: %', ygrd;
hstep := width;
vstep := height;
-- Tweak horizontal start on hstep grid from origin
hstart := xoff + ceil((ST_XMin(ext)-xoff)/hstep)*hstep;
--RAISE DEBUG 'hstart: %', hstart;
-- Tweak vertical start on vstep grid from origin
vstart := yoff + ceil((ST_Ymin(ext)-yoff)/vstep)*vstep;
--RAISE DEBUG 'vstart: %', vstart;
hend := ST_XMax(ext);
vend := ST_YMax(ext);
--RAISE DEBUG 'hend: %', hend;
--RAISE DEBUG 'vend: %', vend;
x := hstart;
WHILE x < hend LOOP -- over X
y := vstart;
h := ST_MakeEnvelope(x-hw, y-hh, x+hw, y+hh, srid);
WHILE y < vend LOOP -- over Y
RETURN NEXT h;
h := ST_Translate(h, 0, vstep);
y := yoff + round(((y + vstep)-yoff)/ygrd)*ygrd; -- round to grid
END LOOP;
x := xoff + round(((x + hstep)-xoff)/xgrd)*xgrd; -- round to grid
END LOOP;
RETURN;
END
$$ LANGUAGE 'plpgsql' IMMUTABLE;
--
-- Calculate the equal interval bins for a given column
--
-- @param in_array A numeric array of numbers to determine the best
-- to determine the bin boundary
--
-- @param breaks The number of bins you want to find.
--
--
-- Returns: upper edges of bins
--
--
CREATE OR REPLACE FUNCTION CDB_EqualIntervalBins ( in_array NUMERIC[], breaks INT ) RETURNS NUMERIC[] as $$
DECLARE
diff numeric;
min_val numeric;
max_val numeric;
tmp_val numeric;
i INT := 1;
reply numeric[];
BEGIN
SELECT min(e), max(e) INTO min_val, max_val FROM ( SELECT unnest(in_array) e ) x WHERE e IS NOT NULL;
diff = (max_val - min_val) / breaks::numeric;
LOOP
IF i < breaks THEN
tmp_val = min_val + i::numeric * diff;
reply = array_append(reply, tmp_val);
i := i+1;
ELSE
reply = array_append(reply, max_val);
EXIT;
END IF;
END LOOP;
RETURN reply;
END;
$$ language plpgsql IMMUTABLE;
--
-- Determine the Heads/Tails classifications from a numeric array
--
-- @param in_array A numeric array of numbers to determine the best
-- bins based on the Heads/Tails method.
--
-- @param breaks The number of bins you want to find.
--
--
CREATE OR REPLACE FUNCTION CDB_HeadsTailsBins ( in_array NUMERIC[], breaks INT) RETURNS NUMERIC[] as $$
DECLARE
element_count INT4;
arr_mean numeric;
i INT := 2;
reply numeric[];
BEGIN
-- get the total size of our row
element_count := array_upper(in_array, 1) - array_lower(in_array, 1);
-- ensure the ordering of in_array
SELECT array_agg(e) INTO in_array FROM (SELECT unnest(in_array) e ORDER BY e) x;
-- stop if no rows
IF element_count IS NULL THEN
RETURN NULL;
END IF;
-- stop if our breaks are more than our input array size
IF element_count < breaks THEN
RETURN in_array;
END IF;
-- get our mean value
SELECT avg(v) INTO arr_mean FROM ( SELECT unnest(in_array) as v ) x;
reply = Array[arr_mean];
-- slice our bread
LOOP
IF i > breaks THEN EXIT; END IF;
SELECT avg(e) INTO arr_mean FROM ( SELECT unnest(in_array) e) x WHERE e > reply[i-1];
IF arr_mean IS NOT NULL THEN
reply = array_append(reply, arr_mean);
END IF;
i := i+1;
END LOOP;
RETURN reply;
END;
$$ language plpgsql IMMUTABLE;
--
-- Determine the Jenks classifications from a numeric array
--
-- @param in_array A numeric array of numbers to determine the best
-- bins based on the Jenks method.
--
-- @param breaks The number of bins you want to find.
--
-- @param iterations The number of different starting positions to test.
--
-- @param invert Optional wheter to return the top of each bin (default)
-- or the bottom. BOOLEAN, default=FALSE.
--
--
CREATE OR REPLACE FUNCTION CDB_JenksBins ( in_array NUMERIC[], breaks INT, iterations INT DEFAULT 5, invert BOOLEAN DEFAULT FALSE) RETURNS NUMERIC[] as $$
DECLARE
element_count INT4;
arr_mean NUMERIC;
bot INT;
top INT;
tops INT[];
classes INT[][];
i INT := 1; j INT := 1;
curr_result NUMERIC[];
best_result NUMERIC[];
seedtarget TEXT;
quant NUMERIC[];
shuffles INT;
BEGIN
-- get the total size of our row
element_count := array_length(in_array, 1); --array_upper(in_array, 1) - array_lower(in_array, 1);
-- ensure the ordering of in_array
SELECT array_agg(e) INTO in_array FROM (SELECT unnest(in_array) e ORDER BY e) x;
-- stop if no rows
IF element_count IS NULL THEN
RETURN NULL;
END IF;
-- stop if our breaks are more than our input array size
IF element_count < breaks THEN
RETURN in_array;
END IF;
shuffles := LEAST(GREATEST(floor(2500000.0/(element_count::float*iterations::float)), 1), 750)::int;
-- get our mean value
SELECT avg(v) INTO arr_mean FROM ( SELECT unnest(in_array) as v ) x;
-- assume best is actually Quantile
SELECT cdb_crankshaft.CDB_QuantileBins(in_array, breaks) INTO quant;
-- if data is very very large, just return quant and be done
IF element_count > 5000000 THEN
RETURN quant;
END IF;
-- change quant into bottom, top markers
LOOP
IF i = 1 THEN
bot = 1;
ELSE
-- use last top to find this bot
bot = top+1;
END IF;
IF i = breaks THEN
top = element_count;
ELSE
SELECT count(*) INTO top FROM ( SELECT unnest(in_array) as v) x WHERE v <= quant[i];
END IF;
IF i = 1 THEN
classes = ARRAY[ARRAY[bot,top]];
ELSE
classes = ARRAY_CAT(classes,ARRAY[bot,top]);
END IF;
IF i > breaks THEN EXIT; END IF;
i = i+1;
END LOOP;
best_result = cdb_crankshaft.CDB_JenksBinsIteration( in_array, breaks, classes, invert, element_count, arr_mean, shuffles);
--set the seed so we can ensure the same results
SELECT setseed(0.4567) INTO seedtarget;
--loop through random starting positions
LOOP
IF j > iterations-1 THEN EXIT; END IF;
i = 1;
tops = ARRAY[element_count];
LOOP
IF i = breaks THEN EXIT; END IF;
SELECT array_agg(distinct e) INTO tops FROM (SELECT unnest(array_cat(tops, ARRAY[floor(random()*element_count::float)::int])) as e ORDER BY e) x WHERE e != 1;
i = array_length(tops, 1);
END LOOP;
i = 1;
LOOP
IF i > breaks THEN EXIT; END IF;
IF i = 1 THEN
bot = 1;
ELSE
bot = top+1;
END IF;
top = tops[i];
IF i = 1 THEN
classes = ARRAY[ARRAY[bot,top]];
ELSE
classes = ARRAY_CAT(classes,ARRAY[bot,top]);
END IF;
i := i+1;
END LOOP;
curr_result = cdb_crankshaft.CDB_JenksBinsIteration( in_array, breaks, classes, invert, element_count, arr_mean, shuffles);
IF curr_result[1] > best_result[1] THEN
best_result = curr_result;
j = j-1; -- if we found a better result, add one more search
END IF;
j = j+1;
END LOOP;
RETURN (best_result)[2:array_upper(best_result, 1)];
END;
$$ language plpgsql IMMUTABLE;
--
-- Perform a single iteration of the Jenks classification
--
CREATE OR REPLACE FUNCTION CDB_JenksBinsIteration ( in_array NUMERIC[], breaks INT, classes INT[][], invert BOOLEAN, element_count INT4, arr_mean NUMERIC, max_search INT DEFAULT 50) RETURNS NUMERIC[] as $$
DECLARE
tmp_val numeric;
new_classes int[][];
tmp_class int[];
i INT := 1;
j INT := 1;
side INT := 2;
sdam numeric;
gvf numeric := 0.0;
new_gvf numeric;
arr_gvf numeric[];
class_avg numeric;
class_max_i INT;
class_min_i INT;
class_max numeric;
class_min numeric;
reply numeric[];
BEGIN
-- Calculate the sum of squared deviations from the array mean (SDAM).
SELECT sum((arr_mean - e)^2) INTO sdam FROM ( SELECT unnest(in_array) as e ) x;
--Identify the breaks for the lowest GVF
LOOP
i = 1;
LOOP
-- get our mean
SELECT avg(e) INTO class_avg FROM ( SELECT unnest(in_array[classes[i][1]:classes[i][2]]) as e) x;
-- find the deviation
SELECT sum((class_avg-e)^2) INTO tmp_val FROM ( SELECT unnest(in_array[classes[i][1]:classes[i][2]]) as e ) x;
IF i = 1 THEN
arr_gvf = ARRAY[tmp_val];
-- init our min/max map for later
class_max = arr_gvf[i];
class_min = arr_gvf[i];
class_min_i = 1;
class_max_i = 1;
ELSE
arr_gvf = array_append(arr_gvf, tmp_val);
END IF;
i := i+1;
IF i > breaks THEN EXIT; END IF;
END LOOP;
-- calculate our new GVF
SELECT sdam-sum(e) INTO new_gvf FROM ( SELECT unnest(arr_gvf) as e ) x;
-- if no improvement was made, exit
IF new_gvf < gvf THEN EXIT; END IF;
gvf = new_gvf;
IF j > max_search THEN EXIT; END IF;
j = j+1;
i = 1;
LOOP
--establish directionality (uppward through classes or downward)
IF arr_gvf[i] < class_min THEN
class_min = arr_gvf[i];
class_min_i = i;
END IF;
IF arr_gvf[i] > class_max THEN
class_max = arr_gvf[i];
class_max_i = i;
END IF;
i := i+1;
IF i > breaks THEN EXIT; END IF;
END LOOP;
IF class_max_i > class_min_i THEN
class_min_i = class_max_i - 1;
ELSE
class_min_i = class_max_i + 1;
END IF;
--Move from higher class to a lower gid order
IF class_max_i > class_min_i THEN
classes[class_max_i][1] = classes[class_max_i][1] + 1;
classes[class_min_i][2] = classes[class_min_i][2] + 1;
ELSE -- Move from lower class UP into a higher class by gid
classes[class_max_i][2] = classes[class_max_i][2] - 1;
classes[class_min_i][1] = classes[class_min_i][1] - 1;
END IF;
END LOOP;
i = 1;
LOOP
IF invert = TRUE THEN
side = 1; --default returns bottom side of breaks, invert returns top side
END IF;
reply = array_append(reply, in_array[classes[i][side]]);
i = i+1;
IF i > breaks THEN EXIT; END IF;
END LOOP;
RETURN array_prepend(gvf, reply);
END;
$$ language plpgsql IMMUTABLE;
--
-- Determine the Quantile classifications from a numeric array
--
-- @param in_array A numeric array of numbers to determine the best
-- bins based on the Quantile method.
--
-- @param breaks The number of bins you want to find.
--
--
CREATE OR REPLACE FUNCTION CDB_QuantileBins ( in_array NUMERIC[], breaks INT) RETURNS NUMERIC[] as $$
DECLARE
element_count INT4;
break_size numeric;
tmp_val numeric;
i INT := 1;
reply numeric[];
BEGIN
-- sort our values
SELECT array_agg(e) INTO in_array FROM (SELECT unnest(in_array) e ORDER BY e ASC) x;
-- get the total size of our data
element_count := array_length(in_array, 1);
break_size := element_count::numeric / breaks;
-- slice our bread
LOOP
IF i < breaks THEN
IF break_size * i % 1 > 0 THEN
SELECT e INTO tmp_val FROM ( SELECT unnest(in_array) e LIMIT 1 OFFSET ceil(break_size * i) - 1) x;
ELSE
SELECT avg(e) INTO tmp_val FROM ( SELECT unnest(in_array) e LIMIT 2 OFFSET ceil(break_size * i) - 1 ) x;
END IF;
ELSIF i = breaks THEN
-- select the last value
SELECT max(e) INTO tmp_val FROM ( SELECT unnest(in_array) e ) x;
ELSE
EXIT;
END IF;
reply = array_append(reply, tmp_val);
i := i+1;
END LOOP;
RETURN reply;
END;
$$ language plpgsql IMMUTABLE;
src/pg/test/expected/02_moran_test.out
+6
View File
@@ -5,6 +5,12 @@ SET client_min_messages TO WARNING;
\set ECHO none
_cdb_random_seeds
(1 row)
moran|significance
0.3399|-0.0196
(1 row)
_cdb_random_seeds
(1 row)
code|quads
01|HH
src/pg/test/expected/08_interpolation_test.out
+3 -3
View File
@@ -1,7 +1,7 @@
SET client_min_messages TO WARNING;
\set ECHO none
nn | nni | idw
-----+--------------------------+-----------------
200 | 238.41059602632179224595 | 341.46260750526
nn | nni | idw
----------------------+--------------------------+-----------------
200.0000000000000000 | 238.41059602632179224595 | 341.46260750526
(1 row)
src/pg/test/expected/13_pia_test.out
+7
View File
@@ -0,0 +1,7 @@
SET client_min_messages TO WARNING;
\set ECHO none
st_astext
-------------------------------------------
POINT(-3.67484492582767 40.4395084885993)
(1 row)
src/pg/test/expected/14_densify_test.out
+50
View File
@@ -0,0 +1,50 @@
SET client_min_messages TO WARNING;
\set ECHO none
cdb_densify
-----------------------------------------------------------------
(01010000001361C3D32B6501403255302AA9B34440,7.0)
(01010000002497FF907EFB0040F085C954C1B04440,8.0)
(0101000000A167B3EA73350140E4141DC9E5AF4440,1.0)
(010100000062A1D634EF38014014D044D8F0B44440,2.0)
(010100000052B81E85EB510140EEEBC03923B24440,3.0)
(0101000000C286A757CA3201409D8026C286AF4440,5.0)
(01010000007DD0B359F5390140F38E537424AF4440,6.0)
(0101000000D237691A140D0140014EEFE2FDB44440,4.0)
(01010000003312B4DCAC14014047F8F1AAE1B14440,4.3333333333333333)
(010100000048C0FBBD27290140A9BBC5D646B34440,2.3333333333333333)
(01010000001DEBE2361A400140F79A0B4953B24440,2.0000000000000000)
(010100000098933D2F02500140115B676994B34440,4.0000000000000000)
(01010000004BA3DC90595001405C456C9DA5B14440,5.3333333333333333)
(0101000000D1FA8198714001404285107D64B04440,3.3333333333333333)
(01010000004AEA043411360140D2B687AA85AF4440,4.0000000000000000)
(0101000000CCA4736BBF2201402B8716D9CEAF4440,6.3333333333333333)
(0101000000832C1EF13E2101402609AF4A0FB04440,4.6666666666666667)
(010100000063A009D8BF090140BEEE38F68AB24440,5.4444444444444444)
(010100000019AE5D3CF8180140A5008D2162B34440,3.5555555555555555)
(01010000007E88BE590E250140EB9DA83067B44440,2.7777777777777778)
(0101000000C05105B65D3B014044A2D0B2EEB34440,2.7777777777777778)
(010100000005329D125F4F01401C8049790FB44440,4.3333333333333333)
(010100000054E45F2DB3570140BBDE724420B34440,4.6666666666666667)
(0101000000E53EEA4DD05701407FD7C9557BB24440,5.1111111111111111)
(01010000004A9CC694D34F01402B63A5BE7BB14440,6.1111111111111111)
(0101000000DD24068195430140317345DA64B04440,4.8888888888888889)
(010100000033E768B7D23A014002994E89AFAF4440,4.4444444444444444)
(010100000083C0CAA145360140CAEC55A065AF4440,5.0000000000000000)
(010100000004549A09D52F01403E3230057EAF4440,5.7777777777777778)
(010100000024040D72661D0140B0DEBBE0E6AF4440,6.7777777777777778)
(01010000007CCD85A42915014017B22F2835B04440,6.3333333333333333)
(0101000000F5F145CA781001401F2DCE18E6B04440,5.6666666666666667)
(0101000000F10DE756572701402134E4146CB14440,3.6666666666666667)
(01010000008894DB45FA290140F8C4EB987EB24440,2.8888888888888889)
(0101000000AABF5E61C13D0140E6E512830FB34440,2.7777777777777778)
(0101000000581215F9574B0140FDF5BB4EAEB24440,3.0000000000000000)
(0101000000EA6C9F19754B0140C0EE126009B24440,3.4444444444444444)
(01010000001383C0CAA145014088CC827674B14440,3.5555555555555555)
(0101000000D0B02B40EE350140C90D9905EDB04440,3.3333333333333333)
(010100000051DA1B7C613201406F36F4DA1DB04440,3.0000000000000000)
(0101000000EA6E133D52390140FD1AE7FAEFAF4440,2.7777777777777778)
(01010000003A487527C5340140C76EEE11A6AF4440,3.3333333333333333)
(0101000000BADB448F542E01403AB4C876BEAF4440,4.1111111111111111)
(0101000000FB12176D7B280140BDE1A0F9EBAF4440,4.0000000000000000)
(44 rows)
src/pg/test/expected/15_tinmap_test.out
+83
View File
@@ -0,0 +1,83 @@
SET client_min_messages TO WARNING;
\set ECHO none
cdb_tinmap
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
(01030000000100000004000000D237691A140D0140014EEFE2FDB444402497FF907EFB0040F085C954C1B0444063A009D8BF090140BEEE38F68AB24440D237691A140D0140014EEFE2FDB44440,5.8148148148148148)
(01030000000100000004000000D237691A140D0140014EEFE2FDB4444063A009D8BF090140BEEE38F68AB2444019AE5D3CF8180140A5008D2162B34440D237691A140D0140014EEFE2FDB44440,4.3333333333333333)
(01030000000100000004000000D237691A140D0140014EEFE2FDB4444019AE5D3CF8180140A5008D2162B344407E88BE590E250140EB9DA83067B44440D237691A140D0140014EEFE2FDB44440,3.4444444444444444)
(01030000000100000004000000D237691A140D0140014EEFE2FDB444407E88BE590E250140EB9DA83067B4444062A1D634EF38014014D044D8F0B44440D237691A140D0140014EEFE2FDB44440,2.9259259259259259)
(0103000000010000000400000062A1D634EF38014014D044D8F0B444407E88BE590E250140EB9DA83067B44440C05105B65D3B014044A2D0B2EEB3444062A1D634EF38014014D044D8F0B44440,2.5185185185185185)
(0103000000010000000400000062A1D634EF38014014D044D8F0B44440C05105B65D3B014044A2D0B2EEB3444005329D125F4F01401C8049790FB4444062A1D634EF38014014D044D8F0B44440,3.0370370370370370)
(0103000000010000000400000062A1D634EF38014014D044D8F0B4444005329D125F4F01401C8049790FB444401361C3D32B6501403255302AA9B3444062A1D634EF38014014D044D8F0B44440,4.4444444444444444)
(010300000001000000040000001361C3D32B6501403255302AA9B3444005329D125F4F01401C8049790FB4444098933D2F02500140115B676994B344401361C3D32B6501403255302AA9B34440,5.1111111111111111)
(010300000001000000040000001361C3D32B6501403255302AA9B3444098933D2F02500140115B676994B3444054E45F2DB3570140BBDE724420B344401361C3D32B6501403255302AA9B34440,5.2222222222222222)
(010300000001000000040000001361C3D32B6501403255302AA9B3444054E45F2DB3570140BBDE724420B34440E53EEA4DD05701407FD7C9557BB244401361C3D32B6501403255302AA9B34440,5.5925925925925926)
(010300000001000000040000001361C3D32B6501403255302AA9B34440E53EEA4DD05701407FD7C9557BB244404A9CC694D34F01402B63A5BE7BB144401361C3D32B6501403255302AA9B34440,6.0740740740740741)
(010300000001000000040000007DD0B359F5390140F38E537424AF44404A9CC694D34F01402B63A5BE7BB14440DD24068195430140317345DA64B044407DD0B359F5390140F38E537424AF4440,5.6666666666666667)
(010300000001000000040000007DD0B359F5390140F38E537424AF4440DD24068195430140317345DA64B0444033E768B7D23A014002994E89AFAF44407DD0B359F5390140F38E537424AF4440,5.1111111111111111)
(010300000001000000040000007DD0B359F5390140F38E537424AF444033E768B7D23A014002994E89AFAF44404AEA043411360140D2B687AA85AF44407DD0B359F5390140F38E537424AF4440,4.8148148148148148)
(010300000001000000040000007DD0B359F5390140F38E537424AF44404AEA043411360140D2B687AA85AF4440C286A757CA3201409D8026C286AF44407DD0B359F5390140F38E537424AF4440,5.0000000000000000)
(010300000001000000040000007DD0B359F5390140F38E537424AF4440C286A757CA3201409D8026C286AF444004549A09D52F01403E3230057EAF44407DD0B359F5390140F38E537424AF4440,5.5925925925925926)
(010300000001000000040000007DD0B359F5390140F38E537424AF444004549A09D52F01403E3230057EAF4440CCA4736BBF2201402B8716D9CEAF44407DD0B359F5390140F38E537424AF4440,6.0370370370370370)
(010300000001000000040000007DD0B359F5390140F38E537424AF4440CCA4736BBF2201402B8716D9CEAF444024040D72661D0140B0DEBBE0E6AF44407DD0B359F5390140F38E537424AF4440,6.3703703703703704)
(0103000000010000000400000024040D72661D0140B0DEBBE0E6AF4440CCA4736BBF2201402B8716D9CEAF4440832C1EF13E2101402609AF4A0FB0444024040D72661D0140B0DEBBE0E6AF4440,5.9259259259259259)
(0103000000010000000400000024040D72661D0140B0DEBBE0E6AF4440832C1EF13E2101402609AF4A0FB044407CCD85A42915014017B22F2835B0444024040D72661D0140B0DEBBE0E6AF4440,5.9259259259259259)
(0103000000010000000400000024040D72661D0140B0DEBBE0E6AF44407CCD85A42915014017B22F2835B044402497FF907EFB0040F085C954C1B0444024040D72661D0140B0DEBBE0E6AF4440,7.0370370370370370)
(010300000001000000040000002497FF907EFB0040F085C954C1B044407CCD85A42915014017B22F2835B04440F5F145CA781001401F2DCE18E6B044402497FF907EFB0040F085C954C1B04440,6.6666666666666667)
(010300000001000000040000002497FF907EFB0040F085C954C1B04440F5F145CA781001401F2DCE18E6B0444063A009D8BF090140BEEE38F68AB244402497FF907EFB0040F085C954C1B04440,6.3703703703703704)
(0103000000010000000400000063A009D8BF090140BEEE38F68AB24440F5F145CA781001401F2DCE18E6B044403312B4DCAC14014047F8F1AAE1B1444063A009D8BF090140BEEE38F68AB24440,5.1481481481481481)
(0103000000010000000400000063A009D8BF090140BEEE38F68AB244403312B4DCAC14014047F8F1AAE1B1444019AE5D3CF8180140A5008D2162B3444063A009D8BF090140BEEE38F68AB24440,4.4444444444444444)
(0103000000010000000400000019AE5D3CF8180140A5008D2162B344403312B4DCAC14014047F8F1AAE1B144408894DB45FA290140F8C4EB987EB2444019AE5D3CF8180140A5008D2162B34440,3.5925925925925926)
(0103000000010000000400000019AE5D3CF8180140A5008D2162B344408894DB45FA290140F8C4EB987EB2444048C0FBBD27290140A9BBC5D646B3444019AE5D3CF8180140A5008D2162B34440,2.9259259259259259)
(0103000000010000000400000019AE5D3CF8180140A5008D2162B3444048C0FBBD27290140A9BBC5D646B344407E88BE590E250140EB9DA83067B4444019AE5D3CF8180140A5008D2162B34440,2.8888888888888889)
(010300000001000000040000007E88BE590E250140EB9DA83067B4444048C0FBBD27290140A9BBC5D646B34440C05105B65D3B014044A2D0B2EEB344407E88BE590E250140EB9DA83067B44440,2.6296296296296296)
(01030000000100000004000000C05105B65D3B014044A2D0B2EEB3444048C0FBBD27290140A9BBC5D646B34440AABF5E61C13D0140E6E512830FB34440C05105B65D3B014044A2D0B2EEB34440,2.6296296296296296)
(01030000000100000004000000C05105B65D3B014044A2D0B2EEB34440AABF5E61C13D0140E6E512830FB3444098933D2F02500140115B676994B34440C05105B65D3B014044A2D0B2EEB34440,3.1851851851851852)
(01030000000100000004000000C05105B65D3B014044A2D0B2EEB3444098933D2F02500140115B676994B3444005329D125F4F01401C8049790FB44440C05105B65D3B014044A2D0B2EEB34440,3.7037037037037037)
(0103000000010000000400000098933D2F02500140115B676994B34440AABF5E61C13D0140E6E512830FB34440581215F9574B0140FDF5BB4EAEB2444098933D2F02500140115B676994B34440,3.2592592592592593)
(0103000000010000000400000098933D2F02500140115B676994B34440581215F9574B0140FDF5BB4EAEB2444054E45F2DB3570140BBDE724420B3444098933D2F02500140115B676994B34440,3.8888888888888889)
(0103000000010000000400000054E45F2DB3570140BBDE724420B34440581215F9574B0140FDF5BB4EAEB24440E53EEA4DD05701407FD7C9557BB2444054E45F2DB3570140BBDE724420B34440,4.2592592592592593)
(01030000000100000004000000E53EEA4DD05701407FD7C9557BB24440581215F9574B0140FDF5BB4EAEB2444052B81E85EB510140EEEBC03923B24440E53EEA4DD05701407FD7C9557BB24440,3.7037037037037037)
(01030000000100000004000000E53EEA4DD05701407FD7C9557BB2444052B81E85EB510140EEEBC03923B244404BA3DC90595001405C456C9DA5B14440E53EEA4DD05701407FD7C9557BB24440,4.4814814814814815)
(01030000000100000004000000E53EEA4DD05701407FD7C9557BB244404BA3DC90595001405C456C9DA5B144404A9CC694D34F01402B63A5BE7BB14440E53EEA4DD05701407FD7C9557BB24440,5.5185185185185185)
(010300000001000000040000004A9CC694D34F01402B63A5BE7BB144404BA3DC90595001405C456C9DA5B144401383C0CAA145014088CC827674B144404A9CC694D34F01402B63A5BE7BB14440,5.0000000000000000)
(010300000001000000040000004A9CC694D34F01402B63A5BE7BB144401383C0CAA145014088CC827674B14440DD24068195430140317345DA64B044404A9CC694D34F01402B63A5BE7BB14440,4.8518518518518518)
(01030000000100000004000000DD24068195430140317345DA64B044401383C0CAA145014088CC827674B14440D1FA8198714001404285107D64B04440DD24068195430140317345DA64B04440,3.9259259259259259)
(01030000000100000004000000DD24068195430140317345DA64B04440D1FA8198714001404285107D64B0444033E768B7D23A014002994E89AFAF4440DD24068195430140317345DA64B04440,4.2222222222222222)
(0103000000010000000400000033E768B7D23A014002994E89AFAF4440D1FA8198714001404285107D64B04440EA6E133D52390140FD1AE7FAEFAF444033E768B7D23A014002994E89AFAF4440,3.5185185185185185)
(0103000000010000000400000033E768B7D23A014002994E89AFAF4440EA6E133D52390140FD1AE7FAEFAF4440A167B3EA73350140E4141DC9E5AF444033E768B7D23A014002994E89AFAF4440,2.7407407407407407)
(0103000000010000000400000033E768B7D23A014002994E89AFAF4440A167B3EA73350140E4141DC9E5AF44403A487527C5340140C76EEE11A6AF444033E768B7D23A014002994E89AFAF4440,2.9259259259259259)
(0103000000010000000400000033E768B7D23A014002994E89AFAF44403A487527C5340140C76EEE11A6AF44404AEA043411360140D2B687AA85AF444033E768B7D23A014002994E89AFAF4440,3.9259259259259259)
(010300000001000000040000004AEA043411360140D2B687AA85AF44403A487527C5340140C76EEE11A6AF4440C286A757CA3201409D8026C286AF44404AEA043411360140D2B687AA85AF4440,4.1111111111111111)
(01030000000100000004000000C286A757CA3201409D8026C286AF44403A487527C5340140C76EEE11A6AF4440BADB448F542E01403AB4C876BEAF4440C286A757CA3201409D8026C286AF4440,4.1481481481481481)
(01030000000100000004000000C286A757CA3201409D8026C286AF4440BADB448F542E01403AB4C876BEAF444004549A09D52F01403E3230057EAF4440C286A757CA3201409D8026C286AF4440,4.9629629629629630)
(0103000000010000000400000004549A09D52F01403E3230057EAF4440BADB448F542E01403AB4C876BEAF4440FB12176D7B280140BDE1A0F9EBAF444004549A09D52F01403E3230057EAF4440,4.6296296296296296)
(0103000000010000000400000004549A09D52F01403E3230057EAF4440FB12176D7B280140BDE1A0F9EBAF4440CCA4736BBF2201402B8716D9CEAF444004549A09D52F01403E3230057EAF4440,5.3703703703703704)
(01030000000100000004000000CCA4736BBF2201402B8716D9CEAF4440FB12176D7B280140BDE1A0F9EBAF4440832C1EF13E2101402609AF4A0FB04440CCA4736BBF2201402B8716D9CEAF4440,5.0000000000000000)
(01030000000100000004000000832C1EF13E2101402609AF4A0FB04440FB12176D7B280140BDE1A0F9EBAF4440F10DE756572701402134E4146CB14440832C1EF13E2101402609AF4A0FB04440,4.1111111111111111)
(01030000000100000004000000832C1EF13E2101402609AF4A0FB04440F10DE756572701402134E4146CB14440F5F145CA781001401F2DCE18E6B04440832C1EF13E2101402609AF4A0FB04440,4.6666666666666667)
(01030000000100000004000000832C1EF13E2101402609AF4A0FB04440F5F145CA781001401F2DCE18E6B044407CCD85A42915014017B22F2835B04440832C1EF13E2101402609AF4A0FB04440,5.5555555555555556)
(01030000000100000004000000F5F145CA781001401F2DCE18E6B04440F10DE756572701402134E4146CB144403312B4DCAC14014047F8F1AAE1B14440F5F145CA781001401F2DCE18E6B04440,4.5555555555555556)
(010300000001000000040000003312B4DCAC14014047F8F1AAE1B14440F10DE756572701402134E4146CB144408894DB45FA290140F8C4EB987EB244403312B4DCAC14014047F8F1AAE1B14440,3.6296296296296296)
(010300000001000000040000008894DB45FA290140F8C4EB987EB24440F10DE756572701402134E4146CB144401DEBE2361A400140F79A0B4953B244408894DB45FA290140F8C4EB987EB24440,2.8518518518518519)
(010300000001000000040000008894DB45FA290140F8C4EB987EB244401DEBE2361A400140F79A0B4953B24440AABF5E61C13D0140E6E512830FB344408894DB45FA290140F8C4EB987EB24440,2.5555555555555556)
(010300000001000000040000008894DB45FA290140F8C4EB987EB24440AABF5E61C13D0140E6E512830FB3444048C0FBBD27290140A9BBC5D646B344408894DB45FA290140F8C4EB987EB24440,2.6666666666666667)
(01030000000100000004000000AABF5E61C13D0140E6E512830FB344401DEBE2361A400140F79A0B4953B24440581215F9574B0140FDF5BB4EAEB24440AABF5E61C13D0140E6E512830FB34440,2.5925925925925926)
(01030000000100000004000000581215F9574B0140FDF5BB4EAEB244401DEBE2361A400140F79A0B4953B24440EA6C9F19754B0140C0EE126009B24440581215F9574B0140FDF5BB4EAEB24440,2.8148148148148148)
(01030000000100000004000000581215F9574B0140FDF5BB4EAEB24440EA6C9F19754B0140C0EE126009B2444052B81E85EB510140EEEBC03923B24440581215F9574B0140FDF5BB4EAEB24440,3.1481481481481481)
(0103000000010000000400000052B81E85EB510140EEEBC03923B24440EA6C9F19754B0140C0EE126009B244404BA3DC90595001405C456C9DA5B1444052B81E85EB510140EEEBC03923B24440,3.9259259259259259)
(010300000001000000040000004BA3DC90595001405C456C9DA5B14440EA6C9F19754B0140C0EE126009B244401383C0CAA145014088CC827674B144404BA3DC90595001405C456C9DA5B14440,4.1111111111111111)
(010300000001000000040000001383C0CAA145014088CC827674B14440EA6C9F19754B0140C0EE126009B244401DEBE2361A400140F79A0B4953B244401383C0CAA145014088CC827674B14440,3.0000000000000000)
(010300000001000000040000001383C0CAA145014088CC827674B144401DEBE2361A400140F79A0B4953B24440D0B02B40EE350140C90D9905EDB044401383C0CAA145014088CC827674B14440,2.9629629629629629)
(010300000001000000040000001383C0CAA145014088CC827674B14440D0B02B40EE350140C90D9905EDB04440D1FA8198714001404285107D64B044401383C0CAA145014088CC827674B14440,3.4074074074074074)
(01030000000100000004000000D1FA8198714001404285107D64B04440D0B02B40EE350140C90D9905EDB0444051DA1B7C613201406F36F4DA1DB04440D1FA8198714001404285107D64B04440,3.2222222222222222)
(01030000000100000004000000D1FA8198714001404285107D64B0444051DA1B7C613201406F36F4DA1DB04440EA6E133D52390140FD1AE7FAEFAF4440D1FA8198714001404285107D64B04440,3.0370370370370370)
(01030000000100000004000000EA6E133D52390140FD1AE7FAEFAF444051DA1B7C613201406F36F4DA1DB04440A167B3EA73350140E4141DC9E5AF4440EA6E133D52390140FD1AE7FAEFAF4440,2.2592592592592593)
(01030000000100000004000000A167B3EA73350140E4141DC9E5AF444051DA1B7C613201406F36F4DA1DB04440BADB448F542E01403AB4C876BEAF4440A167B3EA73350140E4141DC9E5AF4440,2.7037037037037037)
(01030000000100000004000000A167B3EA73350140E4141DC9E5AF4440BADB448F542E01403AB4C876BEAF44403A487527C5340140C76EEE11A6AF4440A167B3EA73350140E4141DC9E5AF4440,2.8148148148148148)
(01030000000100000004000000BADB448F542E01403AB4C876BEAF444051DA1B7C613201406F36F4DA1DB04440FB12176D7B280140BDE1A0F9EBAF4440BADB448F542E01403AB4C876BEAF4440,3.7037037037037037)
(01030000000100000004000000FB12176D7B280140BDE1A0F9EBAF444051DA1B7C613201406F36F4DA1DB04440F10DE756572701402134E4146CB14440FB12176D7B280140BDE1A0F9EBAF4440,3.5555555555555556)
(01030000000100000004000000F10DE756572701402134E4146CB1444051DA1B7C613201406F36F4DA1DB04440D0B02B40EE350140C90D9905EDB04440F10DE756572701402134E4146CB14440,3.3333333333333333)
(01030000000100000004000000F10DE756572701402134E4146CB14440D0B02B40EE350140C90D9905EDB044401DEBE2361A400140F79A0B4953B24440F10DE756572701402134E4146CB14440,3.0000000000000000)
(77 rows)
src/pg/test/expected/19_contour_test.out
+9
View File
@@ -0,0 +1,9 @@
SET client_min_messages TO WARNING;
\set ECHO none
bin|avg_value
0|280.23070673030491816424178
1|413.81702914846213479025305
2|479.6334491374486884098328
3|529.1545236882183479447113
4|614.1132081424930103122037
(5 rows)
src/pg/test/sql/02_moran_test.sql
+8
View File
@@ -6,6 +6,14 @@
-- Areas of Interest functions perform some nondeterministic computations
-- (to estimate the significance); we will set the seeds for the RNGs
-- that affect those results to have repeateble results
-- Moran's I Global
SELECT cdb_crankshaft._cdb_random_seeds(1234);
SELECT round(moran, 4) As moran, round(significance, 4) As significance
FROM cdb_crankshaft.CDB_AreasOfInterestGlobal('SELECT * FROM ppoints', 'value') m(moran, significance);
-- Moran's I Local
SELECT cdb_crankshaft._cdb_random_seeds(1234);
SELECT ppoints.code, m.quads
src/pg/test/sql/13_pia_test.sql
+7
View File
@@ -0,0 +1,7 @@
SET client_min_messages TO WARNING;
\set ECHO none
with a as(
select st_geomfromtext('POLYGON((-432540.453078056 4949775.20452642,-432329.947920966 4951361.232584,-431245.028163694 4952223.31516671,-429131.071033529 4951768.00415574,-424622.07505895 4952843.13503987,-423688.327170174 4953499.20752423,-424086.294349759 4954968.38274191,-423068.388925945 4954378.63345336,-423387.653225542 4953355.67417084,-420594.869840519 4953781.00230592,-416026.095299382 4951484.06849063,-412483.018546414 4951024.5410983,-410490.399661215 4954502.24032205,-408186.197521284 4956398.91417441,-407627.262358013 4959300.94633864,-406948.770061627 4959874.85407739,-404949.583326472 4959047.74518163,-402570.908447199 4953743.46829807,-400971.358683991 4952193.11680804,-403533.488084088 4949649.89857885,-406335.177028373 4950193.19571096,-407790.456731515 4952391.46015616,-412060.672398345 4950381.2389307,-410716.93482498 4949156.7509561,-408464.162289794 4943912.8940387,-409350.599394983 4942819.84896006,-408087.791091424 4942451.6711778,-407274.045613725 4940572.4807777,-404446.196589102 4939976.71501489,-402422.964843936 4940450.3670813,-401010.654464241 4939054.8061663,-397647.247369412 4940679.80737878,-395658.413346901 4940528.84765185,-395536.852462953 4938829.79565997,-394268.923462818 4938003.7277717,-393388.720249116 4934757.80596815,-392393.301362444 4934326.71675815,-392573.527618037 4932323.40974412,-393464.640141837 4931903.10653605,-393085.597275686 4931094.7353605,-398426.261165985 4929156.87541607,-398261.174361137 4926238.00816416,-394045.059966834 4925765.18668498,-392982.960705174 4926391.81893628,-393090.272694301 4927176.84692181,-391648.240010564 4924626.06386961,-391889.914625075 4923086.14787613,-394345.177314013 4923235.086036,-395550.878718795 4917812.79243978,-399009.463978251 4912927.7157945,-398948.794855767 4911941.91010796,-398092.636652078 4911806.57392519,-401991.601817112 4911722.9204501,-406225.972607907 4914505.47286319,-411104.994569885 4912569.26941163,-412925.513522316 4913030.3608866,-414630.148884835 4914436.69169949,-414207.691417276 4919205.78028405,-418306.141109809 4917994.9580478,-424184.700779621 4918938.12432889,-426816.961458921 4923664.37379373,-420956.324227126 4923381.98014807,-420186.661267781 4924286.48693378,-420943.411166194 4926812.76394433,-419779.45457046 4928527.43466337,-419768.767899344 4930681.94459216,-421911.668097113 4930432.40620397,-423482.386112205 4933451.28047252,-427272.814773717 4934151.56473242,-427144.908678797 4939731.77191996,-428982.125554848 4940522.84445172,-428986.133056516 4942437.17281266,-431237.792396792 4947309.68284815,-432476.889648814 4947791.74800037,-432540.453078056 4949775.20452642))', 3857) as g
)
SELECT st_astext(cdb_crankshaft.CDB_PIA(g)) from a;
src/pg/test/sql/14_densify_test.sql
+9
View File
@@ -0,0 +1,9 @@
SET client_min_messages TO WARNING;
\set ECHO none
with data as (
select
ARRAY[7.0,8.0,1.0,2.0,3.0,5.0,6.0,4.0] as colin,
ARRAY[ST_GeomFromText('POINT(2.1744 41.4036)'),ST_GeomFromText('POINT(2.1228 41.3809)'),ST_GeomFromText('POINT(2.1511 41.3742)'),ST_GeomFromText('POINT(2.1528 41.4136)'),ST_GeomFromText('POINT(2.165 41.3917)'),ST_GeomFromText('POINT(2.1498 41.3713)'),ST_GeomFromText('POINT(2.1533 41.3683)'),ST_GeomFromText('POINT(2.131386 41.413998)')] as geomin
)
select cdb_crankshaft.CDB_Densify(geomin, colin, 2) from data;
src/pg/test/sql/15_tinmap_test.sql
+9
View File
@@ -0,0 +1,9 @@
SET client_min_messages TO WARNING;
\set ECHO none
with data as (
select
ARRAY[7.0,8.0,1.0,2.0,3.0,5.0,6.0,4.0] as colin,
ARRAY[ST_GeomFromText('POINT(2.1744 41.4036)'),ST_GeomFromText('POINT(2.1228 41.3809)'),ST_GeomFromText('POINT(2.1511 41.3742)'),ST_GeomFromText('POINT(2.1528 41.4136)'),ST_GeomFromText('POINT(2.165 41.3917)'),ST_GeomFromText('POINT(2.1498 41.3713)'),ST_GeomFromText('POINT(2.1533 41.3683)'),ST_GeomFromText('POINT(2.131386 41.413998)')] as geomin
)
select cdb_crankshaft.CDB_TINmap(geomin, colin, 2) from data;
src/pg/test/sql/19_contour_test.sql
+17
View File
@@ -0,0 +1,17 @@
SET client_min_messages TO WARNING;
\set ECHO none
\pset format unaligned
WITH a AS (
SELECT
ARRAY[800, 700, 600, 500, 400, 300, 200, 100]::numeric[] AS vals,
ARRAY[ST_GeomFromText('POINT(2.1744 41.403)',4326),ST_GeomFromText('POINT(2.1228 41.380)',4326),ST_GeomFromText('POINT(2.1511 41.374)',4326),ST_GeomFromText('POINT(2.1528 41.413)',4326),ST_GeomFromText('POINT(2.165 41.391)',4326),ST_GeomFromText('POINT(2.1498 41.371)',4326),ST_GeomFromText('POINT(2.1533 41.368)',4326),ST_GeomFromText('POINT(2.131386 41.41399)',4326)] AS g
),
b as(
SELECT
foo.*
FROM
a,
cdb_crankshaft.CDB_contour(a.g, a.vals, 0.0, 1, 3, 5, -60) foo
)
SELECT bin, avg_value from b order by bin;
src/py/crankshaft/crankshaft/clustering/moran.py
+23 -35
View File
@@ -14,6 +14,7 @@ import crankshaft.pysal_utils as pu
# High level interface ---------------------------------------
def moran(subquery, attr_name,
w_type, num_ngbrs, permutations, geom_col, id_col):
"""
@@ -30,32 +31,28 @@ def moran(subquery, attr_name,
query = pu.construct_neighbor_query(w_type, qvals)
plpy.notice('** Query: %s' % query)
try:
result = plpy.execute(query)
# if there are no neighbors, exit
if len(result) == 0:
return pu.empty_zipped_array(2)
plpy.notice('** Query returned with %d rows' % len(result))
except plpy.SPIError:
plpy.error('Error: areas of interest query failed, check input parameters')
plpy.notice('** Query failed: "%s"' % query)
plpy.notice('** Error: %s' % plpy.SPIError)
except plpy.SPIError, e:
plpy.error('Analysis failed: %s' % e)
return pu.empty_zipped_array(2)
## collect attributes
# collect attributes
attr_vals = pu.get_attributes(result)
## calculate weights
# calculate weights
weight = pu.get_weight(result, w_type, num_ngbrs)
## calculate moran global
# calculate moran global
moran_global = ps.esda.moran.Moran(attr_vals, weight,
permutations=permutations)
return zip([moran_global.I], [moran_global.EI])
def moran_local(subquery, attr,
w_type, num_ngbrs, permutations, geom_col, id_col):
"""
@@ -79,9 +76,8 @@ def moran_local(subquery, attr,
# if there are no neighbors, exit
if len(result) == 0:
return pu.empty_zipped_array(5)
except plpy.SPIError:
plpy.error('Error: areas of interest query failed, check input parameters')
plpy.notice('** Query failed: "%s"' % query)
except plpy.SPIError, e:
plpy.error('Analysis failed: %s' % e)
return pu.empty_zipped_array(5)
attr_vals = pu.get_attributes(result)
@@ -96,6 +92,7 @@ def moran_local(subquery, attr,
return zip(lisa.Is, quads, lisa.p_sim, weight.id_order, lisa.y)
def moran_rate(subquery, numerator, denominator,
w_type, num_ngbrs, permutations, geom_col, id_col):
"""
@@ -111,32 +108,28 @@ def moran_rate(subquery, numerator, denominator,
query = pu.construct_neighbor_query(w_type, qvals)
plpy.notice('** Query: %s' % query)
try:
result = plpy.execute(query)
# if there are no neighbors, exit
if len(result) == 0:
return pu.empty_zipped_array(2)
plpy.notice('** Query returned with %d rows' % len(result))
except plpy.SPIError:
plpy.error('Error: areas of interest query failed, check input parameters')
plpy.notice('** Query failed: "%s"' % query)
plpy.notice('** Error: %s' % plpy.SPIError)
except plpy.SPIError, e:
plpy.error('Analysis failed: %s' % e)
return pu.empty_zipped_array(2)
## collect attributes
# collect attributes
numer = pu.get_attributes(result, 1)
denom = pu.get_attributes(result, 2)
weight = pu.get_weight(result, w_type, num_ngbrs)
## calculate moran global rate
# calculate moran global rate
lisa_rate = ps.esda.moran.Moran_Rate(numer, denom, weight,
permutations=permutations)
return zip([lisa_rate.I], [lisa_rate.EI])
def moran_local_rate(subquery, numerator, denominator,
w_type, num_ngbrs, permutations, geom_col, id_col):
"""
@@ -160,13 +153,11 @@ def moran_local_rate(subquery, numerator, denominator,
# if there are no neighbors, exit
if len(result) == 0:
return pu.empty_zipped_array(5)
except plpy.SPIError:
plpy.error('Error: areas of interest query failed, check input parameters')
plpy.notice('** Query failed: "%s"' % query)
plpy.notice('** Error: %s' % plpy.SPIError)
except plpy.SPIError, e:
plpy.error('Analysis failed: %s' % e)
return pu.empty_zipped_array(5)
## collect attributes
# collect attributes
numer = pu.get_attributes(result, 1)
denom = pu.get_attributes(result, 2)
@@ -181,12 +172,12 @@ def moran_local_rate(subquery, numerator, denominator,
return zip(lisa.Is, quads, lisa.p_sim, weight.id_order, lisa.y)
def moran_local_bv(subquery, attr1, attr2,
permutations, geom_col, id_col, w_type, num_ngbrs):
"""
Moran's I (local) Bivariate (untested)
"""
plpy.notice('** Constructing query')
qvals = OrderedDict([("id_col", id_col),
("attr1", attr1),
@@ -203,12 +194,11 @@ def moran_local_bv(subquery, attr1, attr2,
if len(result) == 0:
return pu.empty_zipped_array(4)
except plpy.SPIError:
plpy.error("Error: areas of interest query failed, " \
plpy.error("Error: areas of interest query failed, "
"check input parameters")
plpy.notice('** Query failed: "%s"' % query)
return pu.empty_zipped_array(4)
## collect attributes
# collect attributes
attr1_vals = pu.get_attributes(result, 1)
attr2_vals = pu.get_attributes(result, 2)
@@ -219,17 +209,14 @@ def moran_local_bv(subquery, attr1, attr2,
lisa = ps.esda.moran.Moran_Local_BV(attr1_vals, attr2_vals, weight,
permutations=permutations)
plpy.notice("len of Is: %d" % len(lisa.Is))
# find clustering of significance
lisa_sig = quad_position(lisa.q)
plpy.notice('** Finished calculations')
return zip(lisa.Is, lisa_sig, lisa.p_sim, weight.id_order)
# Low level functions ----------------------------------------
def map_quads(coord):
"""
Map a quadrant number to Moran's I designation
@@ -250,6 +237,7 @@ def map_quads(coord):
else:
return None
def quad_position(quads):
"""
Produce Moran's I classification based of n
src/py/crankshaft/crankshaft/pysal_utils/pysal_utils.py
+25 -12
View File
@@ -6,6 +6,7 @@
import numpy as np
import pysal as ps
def construct_neighbor_query(w_type, query_vals):
"""Return query (a string) used for finding neighbors
@param w_type text: type of neighbors to calculate ('knn' or 'queen')
@@ -17,7 +18,8 @@ def construct_neighbor_query(w_type, query_vals):
else:
return queen(query_vals)
## Build weight object
# Build weight object
def get_weight(query_res, w_type='knn', num_ngbrs=5):
"""
Construct PySAL weight from return value of query
@@ -39,6 +41,7 @@ def get_weight(query_res, w_type='knn', num_ngbrs=5):
return built_weight
def query_attr_select(params):
"""
Create portion of SELECT statement for attributes inolved in query.
@@ -57,21 +60,24 @@ def query_attr_select(params):
template = "i.\"%(col)s\"::numeric As attr%(alias_num)s, "
if 'time_cols' in params:
## if markov analysis
# if markov analysis
attrs = params['time_cols']
for idx, val in enumerate(attrs):
attr_string += template % {"col": val, "alias_num": idx + 1}
else:
## if moran's analysis
# if moran's analysis
attrs = [k for k in params
if k not in ('id_col', 'geom_col', 'subquery', 'num_ngbrs')]
if k not in ('id_col', 'geom_col', 'subquery',
'num_ngbrs', 'subquery')]
for idx, val in enumerate(attrs):
attr_string += template % {"col": params[val], "alias_num": idx + 1}
attr_string += template % {"col": params[val],
"alias_num": idx + 1}
return attr_string
def query_attr_where(params):
"""
Construct where conditions when building neighbors query
@@ -81,7 +87,8 @@ def query_attr_where(params):
'numerator': 'data1',
'denominator': 'data2',
'': ...}
Output: 'idx_replace."data1" IS NOT NULL AND idx_replace."data2" IS NOT NULL'
Output: 'idx_replace."data1" IS NOT NULL AND idx_replace."data2"
IS NOT NULL'
Input:
{'subquery': ...,
'time_cols': ['time1', 'time2', 'time3'],
@@ -93,17 +100,18 @@ def query_attr_where(params):
template = "idx_replace.\"%s\" IS NOT NULL"
if 'time_cols' in params:
## markov where clauses
# markov where clauses
attrs = params['time_cols']
# add values to template
for attr in attrs:
attr_string.append(template % attr)
else:
## moran where clauses
# moran where clauses
# get keys
attrs = sorted([k for k in params
if k not in ('id_col', 'geom_col', 'subquery', 'num_ngbrs', 'subquery')])
if k not in ('id_col', 'geom_col', 'subquery',
'num_ngbrs', 'subquery')])
# add values to template
for attr in attrs:
attr_string.append(template % params[attr])
@@ -115,6 +123,7 @@ def query_attr_where(params):
return out
def knn(params):
"""SQL query for k-nearest neighbors.
@param vars: dict of values to fill template
@@ -146,7 +155,8 @@ def knn(params):
return query.format(**params)
## SQL query for finding queens neighbors (all contiguous polygons)
# SQL query for finding queens neighbors (all contiguous polygons)
def queen(params):
"""SQL query for queen neighbors.
@param params dict: information to fill query
@@ -174,14 +184,17 @@ def queen(params):
return query.format(**params)
## to add more weight methods open a ticket or pull request
# to add more weight methods open a ticket or pull request
def get_attributes(query_res, attr_num=1):
"""
@param query_res: query results with attributes and neighbors
@param attr_num: attribute number (1, 2, ...)
"""
return np.array([x['attr' + str(attr_num)] for x in query_res], dtype=np.float)
return np.array([x['attr' + str(attr_num)] for x in query_res],
dtype=np.float)
def empty_zipped_array(num_nones):
"""
src/py/crankshaft/crankshaft/space_time_dynamics/markov.py
+2 -2
View File
@@ -56,9 +56,9 @@ def spatial_markov_trend(subquery, time_cols, num_classes=7,
)
if len(query_result) == 0:
return zip([None], [None], [None], [None], [None])
except plpy.SPIError, err:
except plpy.SPIError, e:
plpy.debug('Query failed with exception %s: %s' % (err, pu.construct_neighbor_query(w_type, qvals)))
plpy.error('Query failed, check the input parameters')
plpy.error('Analysis failed: %s' % e)
return zip([None], [None], [None], [None], [None])
## build weight
src/py/crankshaft/test/test_clustering_moran.py
+33 -15
View File
@@ -14,6 +14,7 @@ import crankshaft.pysal_utils as pu
from crankshaft import random_seeds
import json
class MoranTest(unittest.TestCase):
"""Testing class for Moran's I functions"""
@@ -26,12 +27,15 @@ class MoranTest(unittest.TestCase):
"geom_col": "the_geom",
"num_ngbrs": 321}
self.params_markov = {"id_col": "cartodb_id",
"time_cols": ["_2013_dec", "_2014_jan", "_2014_feb"],
"time_cols": ["_2013_dec", "_2014_jan",
"_2014_feb"],
"subquery": "SELECT * FROM a_list",
"geom_col": "the_geom",
"num_ngbrs": 321}
self.neighbors_data = json.loads(open(fixture_file('neighbors.json')).read())
self.moran_data = json.loads(open(fixture_file('moran.json')).read())
self.neighbors_data = json.loads(
open(fixture_file('neighbors.json')).read())
self.moran_data = json.loads(
open(fixture_file('moran.json')).read())
def test_map_quads(self):
"""Test map_quads"""
@@ -54,35 +58,49 @@ class MoranTest(unittest.TestCase):
def test_moran_local(self):
"""Test Moran's I local"""
data = [ { 'id': d['id'], 'attr1': d['value'], 'neighbors': d['neighbors'] } for d in self.neighbors_data]
data = [{'id': d['id'],
'attr1': d['value'],
'neighbors': d['neighbors']} for d in self.neighbors_data]
plpy._define_result('select', data)
random_seeds.set_random_seeds(1234)
result = cc.moran_local('subquery', 'value', 'knn', 5, 99, 'the_geom', 'cartodb_id')
result = cc.moran_local('subquery', 'value',
'knn', 5, 99, 'the_geom', 'cartodb_id')
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):
zipped_values = zip(result, self.moran_data)
for ([res_val, res_quad], [exp_val, exp_quad]) in zipped_values:
self.assertAlmostEqual(res_val, exp_val)
self.assertEqual(res_quad, exp_quad)
def test_moran_local_rate(self):
"""Test Moran's I rate"""
data = [ { 'id': d['id'], 'attr1': d['value'], 'attr2': 1, 'neighbors': d['neighbors'] } for d in self.neighbors_data]
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('subquery', 'numerator', 'denominator', 'knn', 5, 99, 'the_geom', 'cartodb_id')
print 'result == None? ', result == None
result = cc.moran_local_rate('subquery', 'numerator', 'denominator',
'knn', 5, 99, 'the_geom', 'cartodb_id')
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):
zipped_values = zip(result, self.moran_data)
for ([res_val, res_quad], [exp_val, exp_quad]) in zipped_values:
self.assertAlmostEqual(res_val, exp_val)
def test_moran(self):
"""Test Moran's I global"""
data = [{ 'id': d['id'], 'attr1': d['value'], 'neighbors': d['neighbors'] } for d in self.neighbors_data]
data = [{'id': d['id'],
'attr1': d['value'],
'neighbors': d['neighbors']} for d in self.neighbors_data]
plpy._define_result('select', data)
random_seeds.set_random_seeds(1235)
result = cc.moran('table', 'value', 'knn', 5, 99, 'the_geom', 'cartodb_id')
print 'result == None?', result == None
result = cc.moran('table', 'value',
'knn', 5, 99, 'the_geom', 'cartodb_id')
result_moran = result[0][0]
expected_moran = np.array([row[0] for row in self.moran_data]).mean()
self.assertAlmostEqual(expected_moran, result_moran, delta=10e-2)