HyperAIHyperAI

Command Palette

Search for a command to run...

Console
6 months ago

CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning

View Paper Details

Ulrik Friis-Jensen Frederik L. Johansen Andy S. Anker Erik B. Dam Kirsten M. Ø. Jensen Raghavendra Selvan

CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning

Abstract

Code Repositories

UlrikFriisJensen/CHILI
Official
pytorch
Mentioned in GitHub

Benchmarks

BenchmarkMethodologyMetrics
atomic-number-classification-on-chili-100kGAT
F1-score (Weighted): 0.192 +/- 0.000
atomic-number-classification-on-chili-100kPMLP
F1-score (Weighted): 0.191 +/- 0.000
atomic-number-classification-on-chili-100kGIN
F1-score (Weighted): 0.336 +/- 0.005
atomic-number-classification-on-chili-100kGraphSAGE
F1-score (Weighted): 0.195 +/- 0.007
atomic-number-classification-on-chili-100kGraphUNet
F1-score (Weighted): 0.287 +/- 0.004
atomic-number-classification-on-chili-100kGCN
F1-score (Weighted): 0.275 +/- 0.002
atomic-number-classification-on-chili-100kEdgeCNN
F1-score (Weighted): 0.572 +/- 0.017
atomic-number-classification-on-chili-100kRandom
F1-score (Weighted): 0.015 +/- 0.000
atomic-number-classification-on-chili-100kMost Frequent Class
F1-score (Weighted): 0.192
atomic-number-classification-on-chili-3kRandom
F1-score (Weighted): 0.016 +/- 0.000
atomic-number-classification-on-chili-3kGIN
F1-score (Weighted): 0.587 +/- 0.002
atomic-number-classification-on-chili-3kPMLP
F1-score (Weighted): 0.461 +/- 0.000
atomic-number-classification-on-chili-3kEdgeCNN
F1-score (Weighted): 0.632 +/- 0.009
atomic-number-classification-on-chili-3kGCN
F1-score (Weighted): 0.496 +/- 0.001
atomic-number-classification-on-chili-3kGraphUNet
F1-score (Weighted): 0.552 +/- 0.079
atomic-number-classification-on-chili-3kMost Frequent Class
F1-score (Weighted): 0.461
atomic-number-classification-on-chili-3kGraphSAGE
F1-score (Weighted): 0.491 +/- 0.004
atomic-number-classification-on-chili-3kGAT
F1-score (Weighted): 0.461 +/- 0.000
crystal-system-classification-on-chili-100kGIN
F1-score (Weighted): 0.069 +/- 0.040
crystal-system-classification-on-chili-100kGraphSAGE
F1-score (Weighted): 0.061 +/- 0.019
crystal-system-classification-on-chili-100kGAT
F1-score (Weighted): 0.110 +/- 0.029
crystal-system-classification-on-chili-100kMost Frequent Class
F1-score (Weighted): 0.046
crystal-system-classification-on-chili-100kGCN
F1-score (Weighted): 0.069 +/- 0.023
crystal-system-classification-on-chili-100kGraphUNet
F1-score (Weighted): 0.068 +/- 0.006
crystal-system-classification-on-chili-100kPMLP
F1-score (Weighted): 0.124 +/- 0.036
crystal-system-classification-on-chili-100kEdgeCNN
F1-score (Weighted): 0.072 +/- 0.047
crystal-system-classification-on-chili-100kRandom
F1-score (Weighted): 0.168 +/- 0.014
crystal-system-classification-on-chili-3kGAT
F1-score (Weighted): 0.504 +/- 0.076
crystal-system-classification-on-chili-3kRandom
F1-score (Weighted): 0.191 +/- 0.008
crystal-system-classification-on-chili-3kGCN
F1-score (Weighted): 0.367 +/- 0.127
crystal-system-classification-on-chili-3kPMLP
F1-score (Weighted): 0.440 +/- 0.036
crystal-system-classification-on-chili-3kGIN
F1-score (Weighted): 0.438 +/- 0.004
crystal-system-classification-on-chili-3kGraphSAGE
F1-score (Weighted): 0.422 +/- 0.037
crystal-system-classification-on-chili-3kEdgeCNN
F1-score (Weighted): 0.657 +/- 0.196
crystal-system-classification-on-chili-3kGraphUNet
F1-score (Weighted): 0.431 +/- 0.014
crystal-system-classification-on-chili-3kMost Frequent Class
F1-score (Weighted): 0.440
distance-regression-on-chili-100kGAT
MSE : 0.252 +/- 0.003
distance-regression-on-chili-100kMean
MSE : 0.307
distance-regression-on-chili-100kGCN
MSE : 0.090 +/- 0.002
distance-regression-on-chili-100kEdgeCNN
MSE : 0.030 +/- 0.001
distance-regression-on-chili-100kPMLP
MSE : 0.486 +/- 0.014
distance-regression-on-chili-100kGraphUNet
MSE : 0.085 +/- 0.002
distance-regression-on-chili-100kGIN
MSE : 0.491 +/- 0.038
distance-regression-on-chili-100kGraphSAGE
MSE : 0.064 +/- 0.001
distance-regression-on-chili-3kGCN
MSE : 0.056 +/- 0.006
distance-regression-on-chili-3kGAT
MSE : 0.342 +/- 0.117
distance-regression-on-chili-3kPMLP
MSE : 0.359 +/- 0.017
distance-regression-on-chili-3kGraphUNet
MSE : 0.055 +/- 0.001
distance-regression-on-chili-3kGraphSAGE
MSE : 0.055 +/- 0.002
distance-regression-on-chili-3kEdgeCNN
MSE : 0.015 +/- 0.001
distance-regression-on-chili-3kMean
MSE : 0.265
distance-regression-on-chili-3kGIN
MSE : 0.464 +/- 0.005
position-regression-on-chili-100kGAT
Positional MAE: 16.336 +/- 0.000
position-regression-on-chili-100kGraphSAGE
Positional MAE: 16.337 +/- 0.000
position-regression-on-chili-100kGraphUNet
Positional MAE: 14.824 +/- 0.315
position-regression-on-chili-100kGIN
Positional MAE: 16.336 +/- 0.000
position-regression-on-chili-100kEdgeCNN
Positional MAE: 16.336 +/- 0.000
position-regression-on-chili-100kGCN
Positional MAE: 16.336 +/- 0.000
position-regression-on-chili-100kMean
Positional MAE: 16.336
position-regression-on-chili-100kPMLP
Positional MAE: 16.336 +/- 0.000
position-regression-on-chili-3kGCN
Positional MAE: 16.575 +/- 0.000
position-regression-on-chili-3kGraphSAGE
Positional MAE: 16.575 +/- 0.000
position-regression-on-chili-3kGIN
Positional MAE: 16.575 +/- 0.000
position-regression-on-chili-3kMean
Positional MAE: 16.575
position-regression-on-chili-3kGAT
Positional MAE: 16.575 +/- 0.000
position-regression-on-chili-3kGraphUNet
Positional MAE: 14.765 +/- 0.395
position-regression-on-chili-3kEdgeCNN
Positional MAE: 16.575 +/- 0.000
position-regression-on-chili-3kPMLP
Positional MAE: 16.575 +/- 0.000
saxs-regression-on-chili-100kEdgeCNN
MSE : 0.007 +/- 0.009
saxs-regression-on-chili-100kGIN
MSE : 0.009 +/- 0.000
saxs-regression-on-chili-100kGCN
MSE : 0.010 +/- 0.000
saxs-regression-on-chili-100kGraphSAGE
MSE : 0.011 +/- 0.002
saxs-regression-on-chili-100kMean
MSE : 0.038
saxs-regression-on-chili-100kPMLP
MSE : 0.003 +/- 0.000
saxs-regression-on-chili-100kGAT
MSE : 0.009 +/- 0.000
saxs-regression-on-chili-100kGraphUNet
MSE : 0.009 +/- 0.000
saxs-regression-on-chili-3kGIN
MSE : 0.008 +/- 0.000
saxs-regression-on-chili-3kGAT
MSE : 0.008 +/- 0.000
saxs-regression-on-chili-3kGraphUNet
MSE : 0.008 +/- 0.000
saxs-regression-on-chili-3kPMLP
MSE : 0.022 +/- 0.025
saxs-regression-on-chili-3kEdgeCNN
MSE : 0.006 +/- 0.004
saxs-regression-on-chili-3kGCN
MSE : 0.008 +/- 0.000
saxs-regression-on-chili-3kMean
MSE : 0.037
saxs-regression-on-chili-3kGraphSAGE
MSE : 0.008 +/- 0.001
space-group-classification-on-chili-100kPMLP
F1-score (Weighted): 0.047 +/- 0.012
space-group-classification-on-chili-100kMost Frequent Class
F1-score (Weighted): 0.010
space-group-classification-on-chili-100kGraphSAGE
F1-score (Weighted): 0.044 +/- 0.002
space-group-classification-on-chili-100kRandom
F1-score (Weighted): 0.002 +/- 0.001
space-group-classification-on-chili-100kGAT
F1-score (Weighted): 0.044 +/- 0.001
space-group-classification-on-chili-100kEdgeCNN
F1-score (Weighted): 0.158 +/- 0.035
space-group-classification-on-chili-100kGCN
F1-score (Weighted): 0.043 +/- 0.001
space-group-classification-on-chili-100kGIN
F1-score (Weighted): 0.043 +/- 0.000
space-group-classification-on-chili-100kGraphUNet
F1-score (Weighted): 0.043 +/- 0.000
space-group-classification-on-chili-3kGIN
F1-score (Weighted): 0.125 +/- 0.026
space-group-classification-on-chili-3kPMLP
F1-score (Weighted): 0.135 +/- 0.006
space-group-classification-on-chili-3kMost Frequent Class
F1-score (Weighted): 0.108
space-group-classification-on-chili-3kGraphUNet
F1-score (Weighted): 0.095 +/- 0.036
space-group-classification-on-chili-3kGAT
F1-score (Weighted): 0.113 +/- 0.013
space-group-classification-on-chili-3kEdgeCNN
F1-score (Weighted): 0.733 +/- 0.207
space-group-classification-on-chili-3kGCN
F1-score (Weighted): 0.099 +/- 0.019
space-group-classification-on-chili-3kRandom
F1-score (Weighted): 0.009 +/- 0.008
space-group-classification-on-chili-3kGraphSAGE
F1-score (Weighted): 0.151 +/- 0.045
x-ray-pdf-regression-on-chili-100kPMLP
MSE : 0.013 +/- 0.000
x-ray-pdf-regression-on-chili-100kGCN
MSE : 0.014 +/- 0.000
x-ray-pdf-regression-on-chili-100kGraphSAGE
MSE : 0.037 +/- 0.026
x-ray-pdf-regression-on-chili-100kGAT
MSE : 0.013 +/- 0.000
x-ray-pdf-regression-on-chili-100kGraphUNet
MSE : 0.013 +/- 0.000
x-ray-pdf-regression-on-chili-100kGIN
MSE : 0.013 +/- 0.000
x-ray-pdf-regression-on-chili-100kMean
MSE : 0.007
x-ray-pdf-regression-on-chili-100kEdgeCNN
MSE : 0.012 +/- 0.000
x-ray-pdf-regression-on-chili-3kGAT
MSE : 0.029 +/- 0.030
x-ray-pdf-regression-on-chili-3kGCN
MSE : 0.012 +/- 0.000
x-ray-pdf-regression-on-chili-3kMean
MSE : 0.008
x-ray-pdf-regression-on-chili-3kPMLP
MSE : 0.012 +/- 0.000
x-ray-pdf-regression-on-chili-3kGraphUNet
MSE : 0.012 +/- 0.000
x-ray-pdf-regression-on-chili-3kEdgeCNN
MSE : 0.011 +/- 0.000
x-ray-pdf-regression-on-chili-3kGraphSAGE
MSE : 0.012 +/- 0.000
xrd-regression-on-chili-100kMean
MSE : 0.021
xrd-regression-on-chili-100kEdgeCNN
MSE : 0.006 +/- 0.000
xrd-regression-on-chili-100kGraphSAGE
MSE : 0.018 +/- 0.014
xrd-regression-on-chili-100kGIN
MSE : 0.009 +/- 0.000
xrd-regression-on-chili-100kGraphUNet
MSE : 0.009 +/- 0.000
xrd-regression-on-chili-100kPMLP
MSE : 0.008 +/- 0.001
xrd-regression-on-chili-100kGCN
MSE : 0.009 +/- 0.000
xrd-regression-on-chili-100kGAT
MSE : 0.108 +/- 0.172
xrd-regression-on-chili-3kGraphUNet
MSE : 0.010 +/- 0.000
xrd-regression-on-chili-3kEdgeCNN
MSE : 0.008 +/- 0.001
xrd-regression-on-chili-3kMean
MSE : 0.017
xrd-regression-on-chili-3kGraphSAGE
MSE : 0.010 +/- 0.000
xrd-regression-on-chili-3kGAT
MSE : 0.010 +/- 0.000
xrd-regression-on-chili-3kPMLP
MSE : 0.010 +/- 0.000
xrd-regression-on-chili-3kGCN
MSE : 0.010 +/- 0.000

Build AI with AI

From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.

AI Co-coding
Ready-to-use GPUs
Best Pricing
Get Started

Hyper Newsletters

Subscribe to our latest updates
We will deliver the latest updates of the week to your inbox at nine o'clock every Monday morning
Powered by MailChimp
CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning | Papers | HyperAI