Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[SPARK-6080] [PySpark] correct LogisticRegressionWithLBFGS regType parameter for pyspark #4831

Closed
wants to merge 1 commit into from
+1 −1
Closed

[SPARK-6080] [PySpark] correct LogisticRegressionWithLBFGS regType parameter for pyspark #4831

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion python/pyspark/mllib/classification.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -207,7 +207,7 @@ def train(cls, data, iterations=100, initialWeights=None, regParam=0.01, regType
"""
def train(rdd, i):
return callMLlibFunc("trainLogisticRegressionModelWithLBFGS", rdd, int(iterations), i,
float(regParam), str(regType), bool(intercept), int(corrections),
float(regParam), regType, bool(intercept), int(corrections),
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shall we use str(regType) if regType else None?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think directly use regType is enough, because py4j can translate "l1","l2",None in Python to "l1","l2",null in Java/Scala smoothly.
I found the peer functions LogisticRegressionWithSGD and SVMWithSGD also directly use regType and it can work well.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The difference is the error message. If we use str(...), the Java function call would be successful and inside the Java function, we say the input regType is not recognized.

If we don't use str(..), if users input something like 1, 2. Py4j will throw an error saying there is no Java function matching the input signature, which usually confuses users.

Anyway, this is a minor issue for regType.

float(tolerance))

return _regression_train_wrapper(train, LogisticRegressionModel, data, initialWeights)
Expand Down