Machine learning model construction device, numerical control, machine learning model construction method, and non-transitory computer readable medium encoded with a machine learning model construction program

1. A machine learning model construction device comprising:

a learning circuit including hardware configured to construct a neural network model according to a setting value including at least a flag representing a value of “1” for a case in which a perceptron among a plurality of perceptrons in the learning circuit is not used, and a value of “0” for a case in which a perceptron among the plurality of perceptrons in the learning circuit is used; and

a control circuit including hardware configured to calculate the setting value as a value for constructing a predetermined neural network model in the learning circuit,

wherein the control circuit constructs a neural network model in the learning circuit by setting an output of a multiplier in each perceptron included in the learning circuit to any one of a value arrived at by multiplying a weighting value by an input of the multiplier, a value input to the multiplier, and a value of zero, according to the setting value, and setting a sum total of outputs of multipliers in each perceptron as the output of each perceptron.

2. A machine learning model construction device according to claim 1 , wherein the control circuit receives input of a parameter value, defines a neural network model corresponding to the received parameter value as the predetermined neural network model, and calculates the setting value as the value for constructing the predetermined neural network model in the learning circuit.

3. A machine learning model construction device according to claim 2 , wherein the neural network further includes a layer number and a dimension number, and the received parameter value is a value designating the layer number of the neural network model and the dimension number of each layer.

4. A machine learning model construction device according to claim 2 , wherein the control circuit reads the parameter value from a setting file.

5. A machine learning model construction device according to claim 2 , wherein the control circuit draws a connection relationship of perceptrons according to the parameter value, and a graphical user interface for performing adjustment of the parameter value while a user references the connection relationship of the perceptrons, on a display.

6. A machine learning model construction device according to claim 1 , wherein the control circuit allows a user to select a neural network model prepared in advance, defines the neural network model selected as the predetermined neural network model, and calculates the setting value as the value for constructing the predetermined neural network model in the learning circuit.

7. A numerical control for controlling a machine tool, the numerical control comprising the machine learning model construction device according to claim 1 ,

wherein the control circuit receives input of a parameter value for constructing a neural network model that is related to control of the machine tool in the learning circuit, sets a neural network model corresponding to the received parameter value as the predetermined neural network model, and calculates the setting value as the value for constructing the predetermined neural network model in the learning circuit.

8. A non-transitory computer-readable recording medium encoded with a machine learning model construction program for causing a computer to function as a control circuit of a machine learning model construction device which includes a learning circuit realized by hardware capable of constructing a neural network model according to a setting value including at least a flag representing a value of “1” for a case in which a perceptron among a plurality of perceptrons in the learning circuit is not used, and a value of “0” for a case in which a perceptron among a plurality of perceptrons in the learning circuit is used, the machine learning model construction program causing the control circuit to perform steps comprising:

a step of controlling to calculate the setting value as a value for constructing a predetermined neural network model in the learning circuit, and

a step of constructing a neural network model in the learning circuit, by setting an output of a multiplier in each perceptron included in the learning circuit, to any one of a value arrived at by multiplying a weighting value by an input of the multiplier, a value input to the multiplier, and a value of zero, according to the setting value, and setting a sum total of outputs of multipliers in each perceptron as the output of each perceptron.

9. A method of constructing a machine learning model performed by a machine learning model construction device including a learning circuit realized by hardware capable of constructing a neural network model according to a setting value including a flag representing a value of “1” for a case in which a perceptron among a plurality of perceptrons in the learning circuit is not used, and a value of “0” for a case in which a perceptron among the plurality of perceptrons in the learning circuit is used, the method comprising:

a step of controlling to calculate the setting value as a value for constructing a predetermined neural network model in the learning circuit, and

a step of constructing a neural network model in the learning circuit, by setting an output of a multiplier in each perceptron included in the learning circuit, to any one of a value produced by multiplying a weighting value by an input of the multiplier, a value input to the multiplier, and a value of zero, according to the setting value, and setting a sum total of outputs of multipliers in each perceptron as the output of each perceptron.