madlib-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dmitry Dorofeev <d...@luxmsbi.com>
Subject greenplum: Memory allocation failed. madlib.__bernoulli_vector
Date Tue, 16 May 2017 09:58:59 GMT
Hello, we are getting

psql:06_svm_train.sql:11: ERROR:  plpy.SPIError: plpy.SPIError: Function "madlib.__bernoulli_vector(integer,double
precision,double precision,double precision,integer)": Memory allocation failed. Typically,
this indicates that Greenplum Database limits the available memory to less than what is needed
for this input.  (entry db greenplum.luxms:5432 pid=12385) (plpython.c:4648)
CONTEXT:  Traceback (most recent call last):
  PL/Python function "svm_classification", line 26, in <module>
    return svm.svm(**globals())
  PL/Python function "svm_classification", line 983, in svm
  PL/Python function "svm_classification", line 1103, in _transform_w_kernel
  PL/Python function "svm_classification", line 277, in fit
PL/Python function "svm_classification"

for the following SQL:

SELECT madlib.svm_classification ('stanford.train_input',
                                  'stanford.model_table',
                                  'label=4',
                                  'ind',
                                  'polynomial',   --linear | gaussian
                                  'coef0=0',             -- kernel params
                                  '',             -- grouping_col
                                  'max_iter=1,validation_result=stanford.validation_result');
  -- max_iter=200


MADlib 1.9.1
Greenplum DB version 4.3ORCA
CentOS Linux release 7.2.1511

Any hint how we can tune memory settings to avoid such errors ?

Thanks.

Mime
View raw message