assign new values to a column of H2o dataframe in python

[Paul Xiao]

One column of a H2oFrame contains integers 0-10, I'd like to assign 0 to those <=5, and 1 to those > 5. What is the fastest way to do it. Using a loop takes a long time.

[Lauren DiPerna]

Which language are you working in?

On Thu, May 19, 2016 at 4:46 PM, Paul Xiao <yx7...@gmail.com> wrote:

One column of a H2oFrame contains integers 0-10, I'd like to assign 0 to those <=5, and 1 to those > 5. What is the fastest way to do it. Using a loop takes a long time.

--
You received this message because you are subscribed to the Google Groups "H2O Open Source Scalable Machine Learning - h2ostream" group.
To unsubscribe from this group and stop receiving emails from it, send an email to h2ostream+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Lauren DiPerna]

Sure, for Python or R you can use ifelse()

Here's the example in Python:

test_h2o = h2o.H2OFrame({'col_1':range(11), 'col_2':range(11,22)})
test_h2o['col_1'] = h2o.H2OFrame.ifelse(test_h2o['col_1'] <=5,0,1)
test_h2o

[Lauren DiPerna]

or you can do

test_h2o['col_1'] = (test_h2o['col_1'] <=5).ifelse(1,0)