def test_flights():
client = Client()
cmds = []
cmds.append('drop ptable jayt;')
cmds.append('create ptable jayt from /home/sgeadmin/tabular_predDB/Examples/flight_data_subset.csv;')
cmds.append('create 2 models for jayt;')
cmds.append('analyze jayt for 1 iterations;')
cmds.append('select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20;')
cmds.append('select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0;')
cmds.append('select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0 with respect to actualelapsedtime;')
cmds.append('select dayofweek, actualelapsedtime, similarity to 0 with respect to actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0 with respect to actualelapsedtime, dayofweek;')
cmds.append('select dayofweek, actualelapsedtime, similarity to 0 from jayt where distance > 800 limit 5;')
cmds.append('select dayofweek, actualelapsedtime, arrtime, similarity to 0 with respect to arrtime from jayt where distance > 800 limit 5 order by similarity to 0 with respect to arrtime, dayofweek;')
cmds.append('select probability(actualelapsedtime=200) from jayt where distance > 800 limit 20;')
# cmds.append('select * from jayt limit 5;')
#cmds.append('infer actualelapsedtime from jayt with confidence 0.8 limit 20;')
cmds.append('simulate dayofweek, deptime, crsdeptime FROM jayt where dayofweek = 7 TIMES 3;')
cmds.append('estimate dependence probabilities from jayt;')
cmds.append('estimate dependence probabilities from jayt referencing actualelapsedtime limit 6 save to fz;')
cmds.append('estimate dependence probabilities from jayt referencing actualelapsedtime with confidence 0.5;')
#cmds.append('drop ptable jayt;')
#cmds.append('estimate dependence probabilities from dan_kiva referencing activity limit 10 save to activity_z;')
#cmds.append('select * from dha_small;')
#cmds.append('select probability(mdcr_spnd_outp=1), probability(mdcr_spnd_outp=2), probability(mdcr_spnd_outp=3) from dha_small;')
for cmd in cmds:
print '>>> %s' % cmd
result = client.execute(cmd, timing=True)
print result
def test_flights():
client = Client()
cmds = []
cmds.append('drop ptable jayt;')
cmds.append(
'create ptable jayt from /home/sgeadmin/tabular_predDB/Examples/flight_data_subset.csv;'
)
cmds.append('create 2 models for jayt;')
cmds.append('analyze jayt for 1 iterations;')
cmds.append(
'select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20;'
)
cmds.append(
'select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0;'
)
cmds.append(
'select dayofweek, deptime, crsdeptime, actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0 with respect to actualelapsedtime;'
)
cmds.append(
'select dayofweek, actualelapsedtime, similarity to 0 with respect to actualelapsedtime from jayt where distance > 800 limit 20 order by similarity to 0 with respect to actualelapsedtime, dayofweek;'
)
cmds.append(
'select dayofweek, actualelapsedtime, similarity to 0 from jayt where distance > 800 limit 5;'
)
cmds.append(
'select dayofweek, actualelapsedtime, arrtime, similarity to 0 with respect to arrtime from jayt where distance > 800 limit 5 order by similarity to 0 with respect to arrtime, dayofweek;'
)
cmds.append(
'select probability(actualelapsedtime=200) from jayt where distance > 800 limit 20;'
)
# cmds.append('select * from jayt limit 5;')
#cmds.append('infer actualelapsedtime from jayt with confidence 0.8 limit 20;')
cmds.append(
'simulate dayofweek, deptime, crsdeptime FROM jayt where dayofweek = 7 TIMES 3;'
)
cmds.append('estimate dependence probabilities from jayt;')
cmds.append(
'estimate dependence probabilities from jayt referencing actualelapsedtime limit 6 save to fz;'
)
cmds.append(
'estimate dependence probabilities from jayt referencing actualelapsedtime with confidence 0.5;'
)
#cmds.append('drop ptable jayt;')
#cmds.append('estimate dependence probabilities from dan_kiva referencing activity limit 10 save to activity_z;')
#cmds.append('select * from dha_small;')
#cmds.append('select probability(mdcr_spnd_outp=1), probability(mdcr_spnd_outp=2), probability(mdcr_spnd_outp=3) from dha_small;')
for cmd in cmds:
print '>>> %s' % cmd
result = client.execute(cmd, timing=True)
print result
def test_dha_story_demo():
client = Client()
tests_dir = os.path.split(os.path.realpath(__file__))[0]
dha_csv_path = os.path.join(tests_dir, 'data/dha.csv')
dha_samples_path = os.path.join(tests_dir, 'samples/dha_samples.pkl.gz')
test_results_path = os.path.join(
tests_dir, 'regression_test_output/dha_story_results_record.pkl')
cmd_list = [
'DROP BTABLE dha_demo;',
'CREATE BTABLE dha_demo FROM %s;' % dha_csv_path,
'IMPORT SAMPLES %s INTO dha_demo;' % dha_samples_path,
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo LIMIT 10;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING qual_score LIMIT 6;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING qual_score WITH CONFIDENCE 0.9;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING pymt_p_md_visit LIMIT 6;',
# 'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\') LIMIT 10;',
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\', qual_score), ami_score LIMIT 10;',
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\', pymt_p_visit_ratio), ttl_mdcr_spnd LIMIT 10;',
# 'SIMULATE name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo WHERE ami_score=95.0 TIMES 10;',
# 'SIMULATE name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo WHERE ttl_mdcr_spnd=50000 TIMES 10;',
]
dha_story_results = []
if len(sys.argv) > 1 and sys.argv[1] == 'record':
print 'Recording new dha_story_results to %s' % test_results_path
record = True
else:
## Testing
dha_story_results = pickle.load(open(test_results_path, 'r'))
record = False
for i, cmd in enumerate(cmd_list):
print cmd
result = client.execute(cmd, timing=False, pretty=True)
if record:
dha_story_results.append(result)
else:
if type(result) == dict:
for k, v in result.iteritems():
if isinstance(v, numpy.ndarray):
assert (v == dha_story_results[i][k]).all(), (
v, dha_story_results[i][k])
else:
assert v == dha_story_results[i][k], (
v, dha_story_results[i][k])
else:
#assert result == dha_story_results[i], (result, dha_story_results[i])
pass
if record:
pickle.dump(dha_story_results, open(filename, 'w'))
def test_dha_story_demo():
client = Client()
tests_dir = os.path.split(os.path.realpath(__file__))[0]
dha_csv_path = os.path.join(tests_dir, 'data/dha.csv')
dha_samples_path = os.path.join(tests_dir, 'samples/dha_samples.pkl.gz')
test_results_path = os.path.join(tests_dir, 'regression_test_output/dha_story_results_record.pkl')
cmd_list = [
'DROP BTABLE dha_demo;',
'CREATE BTABLE dha_demo FROM %s;' % dha_csv_path,
'IMPORT SAMPLES %s INTO dha_demo;' % dha_samples_path,
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo LIMIT 10;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING qual_score LIMIT 6;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING qual_score WITH CONFIDENCE 0.9;',
'ESTIMATE DEPENDENCE PROBABILITIES FROM dha_demo REFERENCING pymt_p_md_visit LIMIT 6;',
# 'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\') LIMIT 10;',
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\', qual_score), ami_score LIMIT 10;',
'SELECT name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo ORDER BY similarity_to(name=\'Albany NY\', pymt_p_visit_ratio), ttl_mdcr_spnd LIMIT 10;',
# 'SIMULATE name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo WHERE ami_score=95.0 TIMES 10;',
# 'SIMULATE name, qual_score, ami_score, pymt_p_visit_ratio, ttl_mdcr_spnd, hosp_reimb_ratio, hosp_reimb_p_dcd, md_copay_p_dcd, ttl_copay_p_dcd FROM dha_demo WHERE ttl_mdcr_spnd=50000 TIMES 10;',
]
dha_story_results = []
if len(sys.argv) > 1 and sys.argv[1] == 'record':
print 'Recording new dha_story_results to %s' % test_results_path
record = True
else:
## Testing
dha_story_results = pickle.load(open(test_results_path, 'r'))
record = False
for i, cmd in enumerate(cmd_list):
print cmd
result = client.execute(cmd, timing=False, pretty=True)
if record:
dha_story_results.append(result)
else:
if type(result) == dict:
for k,v in result.iteritems():
if isinstance(v, numpy.ndarray):
assert (v == dha_story_results[i][k]).all(), (v, dha_story_results[i][k])
else:
assert v == dha_story_results[i][k], (v, dha_story_results[i][k])
else:
#assert result == dha_story_results[i], (result, dha_story_results[i])
pass
if record:
pickle.dump(dha_story_results, open(filename, 'w'))
