I keep failing with custom transformer implementation. I posted a question on
stackoverflow, and deleted it as I think it should be more appropriate here.
I followed the suggestions by other people. The code right now is this:
from sklearn.base import TransformerMixin, BaseEstimator
class myRegressor(BaseEstimator, TransformerMixin):
def __init__(self, list1=None, list2=None):
self.list1_ = list1
self.list2_ = list2
self.var_ = # something
print self.list1_
print self.list2_
print list1
print list2
def fit(self, X, y):
# some code
return self
def transform(self, X):
return something
I then feed the estimator into sklearn pipeline for gridsearch with CV.
1) Except for the first iteration of grid search, all other times the
lists I print from within __init__() function are None. Same thing from within
fit().
And I get the same output if I remove `None` from default argument values.
2) If I do provide get_params (I don't event know if I need to provide it
with a transformer) I get "__init__: got unexpected keworkd argument "
etcetera. I'm guessing the issue is the underscore at the end of the parameter
name. For example :
def get_params(self, deep=True):
# suppose this estimator has parameters "alpha" and "recursive"
return {"list1_": self.list1_, "list2_": self.list2_}
However, if I remove the underscore from the string names, I get
'GridSearchCV' object has no attribute 'best_estimator_'
Thanks,
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
