PINN (DeepXDE) for solving VDP equation

"""Backend supported: tensorflow.compat.v1, tensorflow, pytorch"""
import deepxde as dde
import numpy as np

def pde(x, y):
    dy_xx = dde.grad.hessian(y, x)
    dy_x=dde.grad.jacobian(y,x)
    return (
        dy_xx
        -1*(1-y**2)*dy_x
        +y
    )

timedomain = dde.geometry.TimeDomain(0, 10)

def boundary_l(x,on_initial):
    return on_initial and np.isclose(x[0],0)

ic1 = dde.icbc.IC(
    timedomain, lambda x:2,lambda _, on_initial: on_initial
)
#Define the initial condition, set the derivative of the network at 0
def bc_func2(inputs, outputs, X):
    return dde.grad.jacobian(outputs,inputs,i=0,j=None)
ic2=dde.icbc.OperatorBC(timedomain,bc_func2,boundary_l)


data = dde.data.TimePDE(
    timedomain, pde, [ic1, ic2], num_domain=1000, num_boundary=2, 
    #num_initial=160
)
net = dde.nn.FNN([1] + [60] * 3 + [1], "tanh", "Glorot normal")
model = dde.Model(data, net)

model.compile("adam", lr=1e-3)
model.train(epochs=10000)
model.compile("L-BFGS")
losshistory, train_state = model.train()

dde.saveplot(losshistory, train_state, issave=True, isplot=True)