الرئيسية الأخبار أحدث الأوراق البحثية الدروس مجموعات البيانات الموسوعة SOTA نماذج LLM لوحة الأداء GPU الفعاليات

العربية

Formation Energy On Qm9

المقاييس

MAE

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

اسم النموذج	MAE	Paper Title	Repository
PhysNet	0.19	PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges	-
TensorNet	0.09	TensorNet: Cartesian Tensor Representations for Efficient Learning of Molecular Potentials	-
SchNet	0.314	Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials	-
HIP-NN	0.256	Hierarchical modeling of molecular energies using a deep neural network	-
Wigner Kernels	0.100 ± 0.003	Wigner kernels: body-ordered equivariant machine learning without a basis	-
PAMNet	0.136	A Universal Framework for Accurate and Efficient Geometric Deep Learning of Molecular Systems
MXMNet	0.137	Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures	-
MEGNet-Full	0.21	Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals	-
HMGNN	0.138	Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties	-
ALIGNN	0.30	Atomistic Line Graph Neural Network for Improved Materials Property Predictions	-
DimeNet	0.185	Directional Message Passing for Molecular Graphs	-
SchNet-edge-update	0.242	Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials	-
MPNN	0.49	Neural Message Passing for Quantum Chemistry	-
HDAD+KRR	0.58	Machine learning prediction errors better than DFT accuracy	-
DeepMoleNet	0.141	Transferable Multi-level Attention Neural Network for Accurate Prediction of Quantum Chemistry Properties via Multi-task Learning	-
MEGNet-simple	0.28	Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals	-
PhysNet-ens5	0.14	PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges	-
xGPR -- Gaussian process, graph convolution kernel	0.167	Linear-scaling kernels for protein sequences and small molecules outperform deep learning while providing uncertainty quantitation and improved interpretability	-

0 of 18 row(s) selected.

Formation Energy On Qm9 | SOTA | HyperAI