Showing Predictor "NACD (L1) "

This predictor was successfully trained!

Meta

Notes:

from V Baracos

[Trained to optimize Log Likelihood, with L1 regularization ]

Publicly Accessible: yes

Viewable by Administrators: yes

Actions:EditDelete

Hide

Dataset Show


Predictor Settings / Retrain Show

Cross-validation results
5-Fold Cross-Validation Statistics*
Measure PSSP Predictor (median) PSSP Predictor (mean) K-M Predictor
Concordance Index0.76 ± 0.010.76 ± 0.010.5 ± 0.0
Hinged L1 Loss9.39 ± 0.3513.28 ± 0.05
Uncensored L1 loss10.55 ± 0.414.73 ± 0.3812.18 ± 0.05
Marginal L1 Loss15.96 ± 0.5716.94 ± 0.4921.74 ± 0.18
Marginal L2 LossN / A
Hinged L1 Log-Loss0.56 ± 0.020.65 ± 0.010.79 ± 0.01
Uncensored L1 Log-Loss0.75 ± 0.020.96 ± 0.020.97 ± 0.01
Marginal L1 Log-Loss0.7 ± 0.030.76 ± 0.020.99 ± 0.01
Log-Likelihood Loss2.57 ± 0.01
D-calibration χ2 statistics11.64 ± 3.70.97 ± 0.12
D-calibration χ2 p-value0.85 ± 0.151.0 ± 0.0
*Mean value ± standard deviation.


Download Predictions (CSV)

Individual Predictions

D-Calibration Histogram

Kaplan Meier Visualization

Show Feature Weights

Examine Classification Accuracy

Statistics accuracy, specificity, sensitivity (1-calibration) for
classifier with Cutoff:   Day