Optimization the Initial Weights of Artificial Neural Networks via Genetic Algorithm Applied to Hip Bone Fracture Prediction

Table 1

Standard parameter set for training.

Parameter	Value
Transfer function of the hidden neurons	Tan sigmoid
Transfer function of the output neurons	Log sigmoid
Training function	Trainscg
Maximal fail	1
Encoding	Real (decimal)
Chromosome length	71
Population size	30
Weight initialization routine	Rand
Initial range	−1 ~ 1
Fitness function	Mean square error
Selection operation	Roulette whe5el
Crossover	BLX − 0.5
Mutation	Non-uniform
Elitist	2
Stopping criterion	100 iterations