Skip to main content

Advertisement

Table 1 Results from the algorithm used to determine the best ANN architecture

From: A method using artificial neural networks to morphologically assess mouse blastocyst quality

Error from training 12.07%
Error from validation 30.00%
Error from test 25.00%
Error from all data 18.37%
Mean Square Error (MSE) 0.023206
Training Data 60%
Validation Data 20%
Test Data 20%
Number of neurons in the first layer 18
Number of neurons in the second layer 13
Function in the first layer purelin
Function in the second layer logsig
Training function trainscg