Molecular Property Prediction On Freesolv

Métriques

RMSE

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	R2	RMSE	Paper Title	Repository
GIN	0.964±0.008	0.744±0.083	How Powerful are Graph Neural Networks?
GROVER (large)	-	2.272	Self-Supervised Graph Transformer on Large-Scale Molecular Data
S-CGIB	-	1.648±0.074	Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck	-
N-GramRF	-	2.688	N-Gram Graph: Simple Unsupervised Representation for Graphs, with Applications to Molecules
Uni-Mol	-	1.620	Uni-Mol: A Universal 3D Molecular Representation Learning Framework	-
N-GramXGB	-	5.061	N-Gram Graph: Simple Unsupervised Representation for Graphs, with Applications to Molecules
ChemRL-GEM	-	1.877	ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction	-
DropGIN	0.972±0.005	0.657±0.059	DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks
SMA	-	1.09	Self-Guided Masked Autoencoders for Domain-Agnostic Self-Supervised Learning
PNA	0.951±0.009	0.870±0.081	Principal Neighbourhood Aggregation for Graph Nets
ChemBFN	-	1.418	A Bayesian Flow Network Framework for Chemistry Tasks
TokenGT	0.930±0.018	1.038±0.125	Pure Transformers are Powerful Graph Learners
ESA (Edge set attention, no positional encodings)	0.977±0.001	0.595±0.013	An end-to-end attention-based approach for learning on graphs
PretrainGNN	-	2.764	Strategies for Pre-training Graph Neural Networks
GAT	0.959±0.011	0.791±0.101	Graph Attention Networks
Graphormer	0.927±0.005	1.065±0.039	Do Transformers Really Perform Bad for Graph Representation?
D-MPNN	-	2.082	Analyzing Learned Molecular Representations for Property Prediction
GraphGPS	0.861±0.037	1.462±0.188	Recipe for a General, Powerful, Scalable Graph Transformer
GCN	0.957±0.009	0.815±0.086	Semi-Supervised Classification with Graph Convolutional Networks
GATv2	0.970±0.007	0.676±0.081	How Attentive are Graph Attention Networks?

0 of 22 row(s) selected.