issues
search
Sotera
/
webpageclassifier
Categorizes a website given URL into one of blog|wiki|news|forum|classified|shopping|undecided.
Apache License 2.0
8
stars
3
forks
source link
Simplify the JPL_Classifier
#16
Closed
ctwardy
closed
7 years ago
ctwardy
commented
7 years ago
Remove Lemmatizer -> not our job to clean your labels
Drop
df
+colname guessing in favor of
X
and
y
.
Drop
categories
parameter: JPL has a fixed built-in set.
Drop
pagetypes
: use
y
for the actual labels
df
+colname guessing in favor ofX
andy
.categories
parameter: JPL has a fixed built-in set.pagetypes
: usey
for the actual labels