TABLE II.

Performance of predicting smoking status and BMI from the GGMP microbiome data (genus level).

Smoking Status	ROC-AUC	F1-Score	Time (s)
DeepEn-Phy (nParameter=125,811)	0.7043	0.6839	1,433
Vanilla MLP (nParameter=133,433)	0.6520	0.6346	352
PopPhy-CNN (nParameter=151,618)	0.6163	0.5972	567
Random Forest (nTree=500)	0.6478	0.5409	294
Gradient Boosting (nTree=100)	0.6813	0.6076	11
Logistic LASSO (L1 λ=0.001)	0.6654	0.6439	12
Logistic Ridge (L2 λ=0.001)	0.6422	0.6425	12
BMI	MSE	RMSE	Time (s)
DeepEn-Phy (nParameter=40,038)	12.6812	3.5611	747
Vanilla MLP (nParameter=50,883)	16.6410	4.0793	192
PopPhy-CNN	–	–	–
Random Forest (nTree=500)	12.9072	3.5927	992
Gradient Boosting (nTree=100)	12.9317	3.5961	9
Linear LASSO (L1 λ=0.005)	14.8221	3.8500	10
Linear Ridge (L2 λ=0.005)	28.3400	5.3235	4