From: A graphical heuristic for reduction and partitioning of large datasets for scalable supervised training
Fraction of # data points
# features
\(n_c\)
Training time (seconds)
Prediction accuracy (%)
Tensorflow
GSH (/speed-up)
GSH
0.2
10,621
96
2603.82
1201.11 /2.2
98.54
98.45
0.5
15,613
240
11,851.98
1664.38 /7.1
99.13