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

adds fuller test suite for segmentation

Browse Source
This commit is contained in:
Andy Eschbacher
2018-03-15 11:41:47 -04:00
parent bc9cae1a63
commit 6260a080a6
5 changed files with 207 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
src/pg/test/expected/06_segmentation_test.out
View File

@@ -25,3 +25,53 @@ t
t
t
(20 rows)
_cdb_random_seeds
(1 row)
within_tolerance
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
(20 rows)
_cdb_random_seeds
(1 row)
within_tolerance
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
(20 rows)

4
src/pg/test/fixtures/ml_values.sql vendored
View File

@@ -1,7 +1,7 @@
SET client_min_messages TO WARNING;
\set ECHO none
CREATE TABLE ml_values (cartodb_id integer, target float, the_geom geometry, x1 float , x2 float, x3 float, class text);
INSERT INTO ml_values(cartodb_id, target,x1,x2,x3, class) VALUES
CREATE TABLE ml_values (cartodb_id integer, target float, the_geom geometry, x1 float, x2 float, x3 float, class text);
INSERT INTO ml_values(cartodb_id, target, x1, x2, x3, class) VALUES
(0,1.24382137034,0.811403626309,0.657584780869,0,'train'),
(1,1.72727475342,0.447764244847,0.528687533966,1,'train'),
(2,3.32104694099,0.62774565606,0.832647155118,2,'train'),

160
src/pg/test/sql/06_segmentation_test.sql
View File

@@ -3,31 +3,141 @@
\i test/fixtures/ml_values.sql
SELECT cdb_crankshaft._cdb_random_seeds(1234);
-- second version (query, not specifying features)
WITH expected AS (
SELECT generate_series(1000,1020) AS id, unnest(ARRAY[
4.5656517130822492,
1.7928053473230694,
1.0283378773916563,
2.6586517814904593,
2.9699056242935944,
3.9550646059951347,
4.1662572444459745,
3.8126334839264162,
1.8809821053623488,
1.6349065129019873,
3.0391288591472954,
3.3035970359672553,
1.5835471589451968,
3.7530378537263638,
1.0833589653009252,
3.8104965452882897,
2.665217959294802,
1.5850334252802472,
3.679401198805563,
3.5332033186588636
]) AS expected LIMIT 20
SELECT
generate_series(1000, 1020) AS id,
unnest(ARRAY[4.5656517130822492,
1.7928053473230694,
1.0283378773916563,
2.6586517814904593,
2.9699056242935944,
3.9550646059951347,
4.1662572444459745,
3.8126334839264162,
1.8809821053623488,
1.6349065129019873,
3.0391288591472954,
3.3035970359672553,
1.5835471589451968,
3.7530378537263638,
1.0833589653009252,
3.8104965452882897,
2.665217959294802,
1.5850334252802472,
3.679401198805563,
3.5332033186588636 ]) AS expected
LIMIT 20
), training as (
SELECT
array_agg(target)::numeric[] as target,
cdb_crankshaft.CDB_PyAgg(Array[x1, x2, x3]::numeric[]) as features
FROM (SELECT * FROM ml_values ORDER BY cartodb_id asc) as _w
WHERE class = 'train'
), testing As (
SELECT
cdb_crankshaft.CDB_PyAgg(Array[x1, x2, x3]::numeric[]) as features,
array_agg(cartodb_id)::numeric[] as cartodb_ids
FROM (SELECT * FROM ml_values ORDER BY cartodb_id asc) as _w
WHERE class = 'test'
), prediction AS (
SELECT cartodb_id::integer id, prediction
FROM cdb_crankshaft.CDB_CreateAndPredictSegment('SELECT target, x1, x2, x3 FROM ml_values WHERE class = $$train$$','target', Array['x1', 'x2', 'x3'], 'SELECT cartodb_id, x1, x2, x3 FROM ml_values WHERE class = $$test$$')
SELECT
*
FROM
cdb_crankshaft.CDB_CreateAndPredictSegment(
(SELECT target FROM training),
(SELECT features FROM training),
(SELECT features FROM testing),
(SELECT cartodb_ids FROM testing)
)
)
SELECT
abs(e.expected - p.prediction) <= 1e-1 AS within_tolerance
FROM expected e, prediction p
WHERE e.id = p.cartodb_id
LIMIT 20;
SELECT cdb_crankshaft._cdb_random_seeds(1234);
-- second version (query, not specifying features)
WITH expected AS (
SELECT
generate_series(1000, 1020) AS id,
unnest(ARRAY[4.5656517130822492,
1.7928053473230694,
1.0283378773916563,
2.6586517814904593,
2.9699056242935944,
3.9550646059951347,
4.1662572444459745,
3.8126334839264162,
1.8809821053623488,
1.6349065129019873,
3.0391288591472954,
3.3035970359672553,
1.5835471589451968,
3.7530378537263638,
1.0833589653009252,
3.8104965452882897,
2.665217959294802,
1.5850334252802472,
3.679401198805563,
3.5332033186588636 ]) AS expected
LIMIT 20
), prediction AS (
SELECT
cartodb_id::integer id,
prediction
FROM cdb_crankshaft.CDB_CreateAndPredictSegment(
'SELECT target, x1, x2, x3 FROM ml_values WHERE class = $$train$$ ORDER BY cartodb_id asc',
'target',
'SELECT cartodb_id, x1, x2, x3 FROM ml_values WHERE class = $$test$$ ORDER BY cartodb_id asc'
)
LIMIT 20
) SELECT abs(e.expected - p.prediction) <= 1e-1 AS within_tolerance FROM expected e, prediction p WHERE e.id = p.id;
)
SELECT
abs(e.expected - p.prediction) <= 1e-1 AS within_tolerance
FROM expected e, prediction p
WHERE e.id = p.id;
SELECT cdb_crankshaft._cdb_random_seeds(1234);
-- third version (query, specifying features)
WITH expected AS (
SELECT
generate_series(1000, 1020) AS id,
unnest(ARRAY[4.5656517130822492,
1.7928053473230694,
1.0283378773916563,
2.6586517814904593,
2.9699056242935944,
3.9550646059951347,
4.1662572444459745,
3.8126334839264162,
1.8809821053623488,
1.6349065129019873,
3.0391288591472954,
3.3035970359672553,
1.5835471589451968,
3.7530378537263638,
1.0833589653009252,
3.8104965452882897,
2.665217959294802,
1.5850334252802472,
3.679401198805563,
3.5332033186588636 ]) AS expected
LIMIT 20
), prediction AS (
SELECT
cartodb_id::integer id,
prediction
FROM cdb_crankshaft.CDB_CreateAndPredictSegment(
'SELECT target, x1, x2, x3 FROM ml_values WHERE class = $$train$$',
'target',
Array['x1', 'x2', 'x3'],
'SELECT cartodb_id, x1, x2, x3 FROM ml_values WHERE class = $$test$$'
)
LIMIT 20
)
SELECT
abs(e.expected - p.prediction) <= 1e-1 AS within_tolerance
FROM expected e, prediction p
WHERE e.id = p.id;