简单来说就是确定激活函数的坐标

from kan import *

model = KAN(width=[2,3,2,1])
x = torch.normal(0,1,size=(100,2))
model(x);
beta = 100
model.plot(beta=beta)
# [2,3,2,1] means 2 input nodes
# 3 neurons in the first hidden layer,
# 2 neurons in the second hidden layer,
# 1 output node

边或者激活函数的索引

Each activation function is indexed by where is the layer index, is the input neuron index, is the output neuron index. All of them starts from 0. For example, the one in the bottom left corner is (0, 0, 0). Let's try to make it symbolic and see it turns red.

每个激活函数都被(层的索引号，输入神经元索引号，输出索引号)确定

model.fix_symbolic(0,0,0,'sin')
model.plot(beta=beta)
model.unfix_symbolic(0,0,0)