Helmholtz energy @ 25K - meV/atom - H_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 |
1.77708 |
4.75148 |
1.07046 |
2.91236 |
0.99592 |
0.97048 |
min |
1.6821 |
4.0754 |
1.0454 |
2.6023 |
0.9953 |
0.9562 |
max |
1.8965 |
6.4472 |
1.091 |
3.429 |
0.9964 |
0.9778 |
std |
0.0749397 |
0.892747 |
0.016584 |
0.286805 |
0.000354401 |
0.0080086 |
RF - MATMINER+LOBSTER
train_rmse |
test_rmse |
train_errors |
test_errors |
train_r2 |
test_r2 |
|
|---|---|---|---|---|---|---|
mean |
1.77896 |
4.74592 |
1.07348 |
2.93166 |
0.99592 |
0.97042 |
min |
1.6681 |
4.0322 |
1.051 |
2.5816 |
0.9954 |
0.954 |
max |
1.8796 |
6.6022 |
1.0921 |
3.5257 |
0.9965 |
0.9787 |
std |
0.0835833 |
0.983811 |
0.0175756 |
0.325418 |
0.000365513 |
0.00923589 |
MODNet - MATMINER
train_rmse |
test_rmse |
train_errors |
test_errors |
train_r2 |
test_r2 |
|
|---|---|---|---|---|---|---|
mean |
1.00462 |
2.79296 |
0.55522 |
1.50746 |
0.99866 |
0.98928 |
min |
0.8252 |
1.9187 |
0.4934 |
1.2858 |
0.9982 |
0.98 |
max |
1.1584 |
3.7937 |
0.5995 |
1.786 |
0.9991 |
0.9946 |
std |
0.10715 |
0.722653 |
0.0345719 |
0.193421 |
0.000287054 |
0.00538494 |
MODNet - MATMINER+LOBSTER
train_rmse |
test_rmse |
train_errors |
test_errors |
train_r2 |
test_r2 |
|
|---|---|---|---|---|---|---|
mean |
0.9519 |
2.69644 |
0.54838 |
1.50492 |
0.99884 |
0.99 |
min |
0.8868 |
1.877 |
0.538 |
1.2041 |
0.9985 |
0.9816 |
max |
1.0502 |
3.6385 |
0.5663 |
1.7663 |
0.999 |
0.995 |
std |
0.0596633 |
0.671187 |
0.0105446 |
0.207836 |
0.000185472 |
0.00491976 |
Corrected resampled t-test on 10-fold CV#
Summary
t_stat |
p_value |
significance_stars |
d_av |
rel_improvement |
percent_folds_improved |
|
|---|---|---|---|---|---|---|
RF |
0.288931 |
0.38959 |
0.0228384 |
0.291326 |
50 |
|
MODNet |
0.404648 |
0.347592 |
0.053225 |
1.19647 |
40 |
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
