am I right when I just use cPickle for seriailzation a model (see code below)
or didnt I get it with PickleSerializer (from pyspark.serializers import
PickleSerializer)
...
model = LogisticRegressionWithSGD.train(parsedData)
mm = open("mm.txt","wb")
import cPickle
cPickle.dump(model,mm)
mm.close()
... using another ipython session ....
import cPickle
mm = open("mm.txt","rb")
model = cPickle.load(mm)
model._coeff
->
Out[4]:
array([-0.19081369, 0.08550208, 0.01062124, 0.36476343, -0.07904827,
0.00820299, 0.23729565, -0.2596712 , 0.08584212, 0.08584212,
0.63061877, -0.06770867, 1.00000382, 0.00630704, 0.26766591, 1.
])
--
View this message in context:
http://apache-spark-user-list.1001560.n3.nabble.com/serialization-a-model-tp7177.html
Sent from the Apache Spark User List mailing list archive at Nabble.com.
