Fetal birthweight prediction with measured data by a temporal machine learning method

Table 3 Summary of parameter values in each model

Models	Parameters	Values	Parameters Meaning
SVM	kernel	linear	kernel function
	C	1.0	regularization parameter
	Cache_size	200	specify the size of the kernel cache
	tol	0.001	tolerance for stopping criterion
	gamma	scale	kernel coefficient
BPNN	kernel initializer	uniform	kernel initializer function
	activation1	relu	activation of hidden layer
	activation2	sigmoid	activation of output layer
	optimizer	Adam	training optimization algorithm
	epochs	200	number of times shown to the network
	batch size	128	batch size
Linear-R	fit_intercept	True	whether to calculate the intercept for this model
	normalize	False	whether to standardize the data
	copy_X	True	If True, X will be copied
CNN	lr	0.01	learning rate
	epochs	100	number of times shown to the network
	optimizer	Adam	training optimization algorithm
	Conv1_in_channels	1	number of channels in the input image
	Conv1_outchannels	10	number of channels produced by the convolution
	Conv1_kernel_size	1	size of the convolving kernel
	Conv1_strid	2	stride of the convolution
	Conv2_in_channels	10	number of channels in the input image
	Conv2_outchannels	20	number of channels produced by the convolution
	Conv2_kernel_size	1	size of the convolving kernel
	Conv2_strid	2	stride of the convolution
RF	n estimators	200	the number of trees in the forest
	Min_samples_leaf	1	the minimum number of samples required to be at a leaf node
	Min_samples_split	2	the minimum number of samples required to split an internal node
	max depth	None	the maximum depth of the tree.
	max features	7	the number of features to consider when looking for the best split

ISSN: 1472-6947