Heat capacity @ 25K - meV/atom - Cv_25

ARFS Top features

ARFS selected descriptors

---

Correlation analysis

Distance correlation

Dependency graphs

Feature learnability

---

Model performance

5-Fold CV Metrics overview

RF - MATMINER

	train_rmse	test_rmse	train_errors	test_errors	train_r2	test_r2
mean	0.00254	0.00684	0.00168	0.00466	0.992	0.94062
min	0.0024	0.006	0.0016	0.0043	0.9917	0.924
max	0.0026	0.0073	0.0017	0.0051	0.9923	0.9561
std	8e-05	0.000440908	4e-05	0.000272764	0.0002	0.0106304

RF - MATMINER+LOBSTER

	train_rmse	test_rmse	train_errors	test_errors	train_r2	test_r2
mean	0.0025	0.0068	0.00168	0.00462	0.99204	0.94168
min	0.0025	0.0059	0.0016	0.0042	0.9918	0.93
max	0.0025	0.0072	0.0017	0.005	0.9922	0.9577
std	0	0.000460435	4e-05	0.000271293	0.000162481	0.00922115

MODNet - MATMINER

	train_rmse	test_rmse	train_errors	test_errors	train_r2	test_r2
mean	0.00438	0.00552	0.00298	0.00376	0.9757	0.96168
min	0.0039	0.005	0.0027	0.0036	0.9709	0.9525
max	0.0048	0.0058	0.0033	0.0039	0.9813	0.9671
std	0.000305941	0.000278568	0.000203961	0.00010198	0.00358664	0.00508268

MODNet - MATMINER+LOBSTER

	train_rmse	test_rmse	train_errors	test_errors	train_r2	test_r2
mean	0.00414	0.00564	0.00292	0.00386	0.97822	0.95962
min	0.0038	0.0049	0.0026	0.0034	0.9665	0.9501
max	0.0052	0.0065	0.0037	0.0044	0.9823	0.9705
std	0.000535164	0.000535164	0.000396989	0.00032	0.00589827	0.00840795

Corrected resampled t-test on 10-fold CV

Summary

	t_stat	p_value	significance_stars	d_av	rel_improvement	percent_folds_improved
RF	1.23233	0.124525		0.195034	1.80666	80
MODNet	-1.26104	0.880499		-0.411092	-2.9137	20

---

Model Explainer

PFI

SHAP

---

Misc

ARFS n-iter convergence checks

MAE/ fold from 10-fold CV

Alternative visual summary of input data for t-test