Skip to content

feat: Add TransferForestClassifier #35

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

Merged
antoinedemathelin merged 11 commits into from
Mar 26, 2022
Merged

feat: Add TransferForestClassifier #35

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
bf27e4f
TransferForestClassifier
atiqm Mar 23, 2022
5ddd4c9
Merge transfer tree
atiqm Mar 23, 2022
6178a76
update comments
atiqm Mar 23, 2022
049d4eb
Update TransferForestClassifier
atiqm Mar 23, 2022
43439de
test commit : comment plot
atiqm Mar 24, 2022
96f352d
tests treeutils, TransferTreeClassifier and TransferForestClassifier
atiqm Mar 25, 2022
3709b97
Update __init__
atiqm Mar 25, 2022
6e8bc2c
test file corrections
atiqm Mar 25, 2022
537e7d6
Add BaseAdaptEstimator
antoinedemathelin Mar 25, 2022
519322b
Fix test transfertree
antoinedemathelin Mar 26, 2022
4fdcf23
Fix version of jinja2 because of bug with nbconvert
antoinedemathelin Mar 26, 2022
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
Next Next commit
TransferForestClassifier
  • Loading branch information
@atiqm
atiqm committed Mar 23, 2022
commit bf27e4f8ac9fa9fa53f33ed6712f443193660c54
29 changes: 26 additions & 3 deletions adapt/_tree_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,31 @@
import copy
import numpy as np


def _bootstrap_(size,class_wise=False,y=None):
if class_wise:
if y is None:
print("Error : need labels to apply class wise bootstrap.")
else:
inds = []
oob_inds = []
classes_ = set(y)
ind_classes_ = np.zeros(len(classes_),dtype=object)

for j,c in enumerate(classes_):
ind_classes_[j] = np.where(y==c)[0]
s = ind_classes_[j].size
inds += list(np.random.choice(ind_classes_[j], s, replace=True))
oob_inds += list(set(ind_classes_[j]) - set(inds))

inds,oob_inds = np.array(inds),np.array(oob_inds)
else:
inds = np.random.choice(np.arange(size), size, replace=True)
oob_inds = set(np.arange(size)) - set(inds)
oob_inds = np.array(list(oob_inds))

return inds, oob_inds

def depth_tree(dt,node=0):

if dt.tree_.feature[node] == -2:
Expand Down Expand Up @@ -464,8 +489,7 @@ def coherent_new_split(phi,th,rule):
return 0,1
else:
return 1,0



def all_coherent_splits(rule,all_splits):

inds = np.zeros(all_splits.shape[0],dtype=bool)
Expand Down Expand Up @@ -534,4 +558,3 @@ def bounds_rule(rule,n_features):

return bound_infs,bound_sups


Loading