Skip to main content

Table 8 Wheat datasets

From: Genome-enabled prediction using probabilistic neural network classifiers

  PNN15% (two classes) PNN15% (three classes) PNN30% (two classes) PNN30% (three classes)
Upper class
GY-1 0.270 (0.134) 0.288 (0.140) 0.499 (0.118) 0.475 (0.102)
GY-2 0.322 (0.118) 0.307 (0.111) 0.538 (0.117) 0.567 (0.116)
GY-3 0.310 (0.138) 0.268 (0.108) 0.452 (0.117) 0.453 (0.118)
GY-4 0.319 (0.121) 0.325 (0.118) 0.482 (0.112) 0.518 (0.115)
GY-5 0.333 (0.161) 0.326 (0.142) 0.545 (0.104) 0.550 (0.107)
GY-6 0.562 (0.159) 0.561 (0.157) 0.668 (0.087) 0.701 (0.083)
GY-7 0.263 (0.124) 0.290 (0.117) 0.503 (0.117) 0.529 (0.115)
Lower class
DTH-1 0.370 (0.114) 0.414 (0.124) 0.629 (0.091) 0.630 (0.091)
DTH-2 0.417 (0.134) 0.429 (0.132) 0.548 (0.116) 0.521 (0.104)
DTH-3 0.506 (0.158) 0.511 (0.151) 0.641 (0.093) 0.650 (0.095)
DTH-4 0.292 (0.090) 0.298 (0.089) 0.350 (0.094) 0.363 (0.097)
DTH-5 0.355 (0.158) 0.384 (0.164) 0.551 (0.091) 0.546 (0.087)
DTH-6 0.444 (0.087) 0.467 (0.094) 0.530 (0.119) 0.520 (0.112)
DTH-7 0.462 (0.116) 0.482 (0.113) 0.580 (0.122) 0.591 (0.115)
DTH-8 0.387 (0.104) 0.382 (0.103) 0.603 (0.089) 0.599 (0.095)
DTH-9 0.376 (0.138) 0.367 (0.112) 0.532 (0.105) 0.535 (0.107)
DTH-10 0.557 (0.112) 0.553 (0.114) 0.575 (0.117) 0.578 (0.124)
  1. Mean values of the area under the precision-recall curve AUCpr (standard deviation in parentheses) of 50 random partitions for the 15 and 30 % upper classes for grain yield (GY) in 7 environments (1–7) and for 15 and 30 % lower classes for days to heading (DTH) for classifier PNN with two and three classes. Numbers in bold are the highest AUCpr values