MoleculeNet Classification Data : Modularization, dynamic splits, Clean Code - #174

Draft

aditya0by0 wants to merge 61 commits into

devfrom

fix/molecule_net_dynamic_split

Commits on Jul 17, 2026

make dynamic dataset generic
aditya0by0
committed
010e300
add common class for molecule net
aditya0by0
committed
8927a69
remain code
aditya0by0
committed
3209b56
add group splitter class
aditya0by0
committed
6c759e6
add class for multilabel splitter
aditya0by0
committed
e492b24
update classification molecule net config
aditya0by0
committed
3f33b54
general splitter class
aditya0by0
committed
75d6301
certific dependency for ssl error
aditya0by0
committed
e87df76
fix file path error
aditya0by0
committed
05dbb4d
right column names for each data
aditya0by0
committed
c9841ad
addd certifi
aditya0by0
committed
f6c9a84

Commits on Jul 18, 2026

Commits on Jul 19, 2026

remove tox test
aditya0by0
committed
beebc7b
save raw and processed in our custom dir
aditya0by0
committed
50f033c
fix settings json
aditya0by0
committed
4555935
fix pre-commit
aditya0by0
committed
1773c00
use mol as features instead of smiles as per https://github.com/ChEB-AI/python-chebai/pull/147/
aditya0by0
committed
c608370
fix test, and splitter docstrings
aditya0by0
committed
d0804d6
rename to molNet_classification
aditya0by0
committed
3da4876
remove split size from config
aditya0by0
committed
833727f

Commits on Jul 22, 2026

move chebi version parameter from base data class to chebi class
aditya0by0
committed
99ebd14
fix test
aditya0by0
committed
bc72d04
Rename 'val' to 'validation' in data splits (#175)

sfluegel05
authored and
aditya0by0
committed
57881ad
Merge branch 'dev' into fix/molecule_net_dynamic_split
aditya0by0
authored
4c59fbd
fix validation key for all splitters
aditya0by0
committed
4f3b325

Commits on Jul 24, 2026

Commits on Jul 25, 2026

rename to moleculenet_classification
aditya0by0
committed
beff841
rename to molecule_net_classification
aditya0by0
committed
79aca00
CLASS name captilazie
aditya0by0
committed
2a624da
add multilabel avg precision
aditya0by0
committed
4c95eb3

Commits on Aug 7, 2026

add data type prop for data specific token files
aditya0by0
committed
d059192
Merge branch 'dev' into fix/molecule_net_dynamic_split
aditya0by0
committed
3597b51
suggested changes
aditya0by0
committed
5f76acb
fix ident
aditya0by0
committed
5e52f86

Commits on Aug 10, 2026

fix pr-auc link error
aditya0by0
committed
2f18c58
data type for each molenet class
aditya0by0
committed
b6b5ed0
fix diff dir for molenet issue
aditya0by0
committed
18a2f66
fix data empty error + prompt user to use split file path
aditya0by0
committed
6f6e310
pre-commit run -a
aditya0by0
committed
05492ad

Commits on Aug 11, 2026

fix duplicate entries in molnet datasets
aditya0by0
committed
7a93792

Commits on Aug 13, 2026

convert labels to bool dtype
aditya0by0
committed
e3df681

Commits on Aug 15, 2026

ignore labels with 0 weights
aditya0by0
committed
f778893
fix None label issue
aditya0by0
committed
af8645d
move missing label mask to collator func and handle cases where all labels are None [None, None]
aditya0by0
committed
bac2cd3

Commits on Aug 18, 2026

Commits on Aug 19, 2026

Commits on Aug 20, 2026

Commits on Aug 21, 2026

also return None if valid_label_mask has all valid labels
aditya0by0
committed
f4fabae