'Accelerated Failure Time' label issues

#1

Hi everyone,

I’ve been trying to run this tutorial for Accelerated Failure Time in survival analysis.

When I run the following example:

import numpy as np
import xgboost as xgb

# 4-by-2 Data matrix
X = np.array([[1, -1], [-1, 1], [0, 1], [1, 0]])
dtrain = xgb.DMatrix(X)

# Associate ranged labels with the data matrix.
# This example shows each kind of censored labels.
#                         uncensored    right     left  interval
y_lower_bound = np.array([      2.0,     3.0, -np.inf,     4.0])
y_upper_bound = np.array([      2.0, +np.inf,     4.0,     5.0])

dtrain.set_float_info('label_lower_bound', y_lower_bound)
dtrain.set_float_info('label_upper_bound', y_upper_bound)

I get the error “Unknown metainfo: label_lower_bound”. I’m aware that the AFL model is a new addition to XGBoost, I just want to know if the current version supports upper and lower labels in a DMatrix yet (since it’s used in the tutorial provided by XGBoost).

Any clarity will be much appreciated.

Carl

#2

AFT is brand new feature and is not part of 1.0. It will be part of 1.1 release.

#4

Thanks for the fast response. Guess I’ll have to wait for 1.1 then.

#5

You can install nightly builds to access latest features.

#6

Thanks, that sorted me out!