How to use the torchdiffeq._impl.misc._convert_to_tensor function in torchdiffeq

def __init__(
        self, func, y0, rtol, atol, first_step=None, safety=0.9, ifactor=10.0, dfactor=0.2, max_num_steps=2**31 - 1,
        **unused_kwargs
    ):
        _handle_unused_kwargs(self, unused_kwargs)
        del unused_kwargs

        self.func = func
        self.y0 = y0
        self.rtol = rtol if _is_iterable(rtol) else [rtol] * len(y0)
        self.atol = atol if _is_iterable(atol) else [atol] * len(y0)
        self.first_step = first_step
        self.safety = _convert_to_tensor(safety, dtype=torch.float64, device=y0[0].device)
        self.ifactor = _convert_to_tensor(ifactor, dtype=torch.float64, device=y0[0].device)
        self.dfactor = _convert_to_tensor(dfactor, dtype=torch.float64, device=y0[0].device)
        self.max_num_steps = _convert_to_tensor(max_num_steps, dtype=torch.int32, device=y0[0].device)

#                   Update RK State                    #
        ########################################################
        dt_next = _optimal_step_size(
            dt, mean_sq_error_ratio, safety=self.safety, ifactor=self.ifactor, dfactor=self.dfactor, order=5)
        if not (dt_next<0.02): #not (dt_next<0.02 or dt_next>0.1):
            y_next = y1 if accept_step else y0
            f_next = f1 if accept_step else f0
            t_next = t0 + dt if accept_step else t0
            interp_coeff = _interp_fit_dopri5(y0, y1, k, dt) if accept_step else interp_coeff
        else:
            if dt_next<0.02:
                print("warning the step of dopri5 {} is too small, set to 0.01".format(dt_next))
                dt_next = _convert_to_tensor(0.01, dtype=torch.float64, device=y0[0].device)
            if dt_next>0.1:
                print("warning the step of dopri5 {} is too big, set to 0.1".format(dt_next))
                dt_next = _convert_to_tensor(0.1, dtype=torch.float64, device=y0[0].device)
            y_next = y1
            f_next = f1
            t_next = t0 + dt
            interp_coeff = _interp_fit_dopri5(y0, y1, k, dt)
        rk_state = _RungeKuttaState(y_next, f_next, t0, t_next, dt_next, interp_coeff)
        return rk_state

accept_step = (torch.tensor(mean_sq_error_ratio) <= 1).all()

        ########################################################
        #                   Update RK State                    #
        ########################################################
        dt_next = _optimal_step_size(
            dt, mean_sq_error_ratio, safety=self.safety, ifactor=self.ifactor, dfactor=self.dfactor, order=5)
        if not (dt_next<0.02): #not (dt_next<0.02 or dt_next>0.1):
            y_next = y1 if accept_step else y0
            f_next = f1 if accept_step else f0
            t_next = t0 + dt if accept_step else t0
            interp_coeff = _interp_fit_dopri5(y0, y1, k, dt) if accept_step else interp_coeff
        else:
            if dt_next<0.02:
                print("warning the step of dopri5 {} is too small, set to 0.01".format(dt_next))
                dt_next = _convert_to_tensor(0.01, dtype=torch.float64, device=y0[0].device)
            if dt_next>0.1:
                print("warning the step of dopri5 {} is too big, set to 0.1".format(dt_next))
                dt_next = _convert_to_tensor(0.1, dtype=torch.float64, device=y0[0].device)
            y_next = y1
            f_next = f1
            t_next = t0 + dt
            interp_coeff = _interp_fit_dopri5(y0, y1, k, dt)
        rk_state = _RungeKuttaState(y_next, f_next, t0, t_next, dt_next, interp_coeff)
        return rk_state

def __init__(
        self, func, y0, rtol, atol, first_step=None, safety=0.9, ifactor=10.0, dfactor=0.2, max_num_steps=2**31 - 1,
        **unused_kwargs
    ):
        _handle_unused_kwargs(self, unused_kwargs)
        del unused_kwargs

        self.func = func
        self.y0 = y0
        self.rtol = rtol if _is_iterable(rtol) else [rtol] * len(y0)
        self.atol = atol if _is_iterable(atol) else [atol] * len(y0)
        self.first_step = first_step
        self.safety = _convert_to_tensor(safety, dtype=torch.float64, device=y0[0].device)
        self.ifactor = _convert_to_tensor(ifactor, dtype=torch.float64, device=y0[0].device)
        self.dfactor = _convert_to_tensor(dfactor, dtype=torch.float64, device=y0[0].device)
        self.max_num_steps = _convert_to_tensor(max_num_steps, dtype=torch.int32, device=y0[0].device)

def before_integrate(self, t):
        f0 = self.func(t[0].type_as(self.y0[0]), self.y0)
        if self.first_step is None:
            first_step = _select_initial_step(self.func, t[0], self.y0, 4, self.rtol[0], self.atol[0], f0=f0).to(t)
        else:
            first_step = _convert_to_tensor(0.01, dtype=t.dtype, device=t.device)
        self.rk_state = _RungeKuttaState(self.y0, f0, t[0], t[0], first_step, interp_coeff=[self.y0] * 5)

def __init__(
        self, func, y0, rtol, atol, first_step=None, safety=0.9, ifactor=10.0, dfactor=0.2, max_num_steps=2**31 - 1,
        **unused_kwargs
    ):
        _handle_unused_kwargs(self, unused_kwargs)
        del unused_kwargs

        self.func = func
        self.y0 = y0
        self.rtol = rtol if _is_iterable(rtol) else [rtol] * len(y0)
        self.atol = atol if _is_iterable(atol) else [atol] * len(y0)
        self.first_step = first_step
        self.safety = _convert_to_tensor(safety, dtype=torch.float64, device=y0[0].device)
        self.ifactor = _convert_to_tensor(ifactor, dtype=torch.float64, device=y0[0].device)
        self.dfactor = _convert_to_tensor(dfactor, dtype=torch.float64, device=y0[0].device)
        self.max_num_steps = _convert_to_tensor(max_num_steps, dtype=torch.int32, device=y0[0].device)

def __init__(
        self, func, y0, rtol, atol, first_step=None, safety=0.9, ifactor=10.0, dfactor=0.2, max_num_steps=2**31 - 1,
        **unused_kwargs
    ):
        _handle_unused_kwargs(self, unused_kwargs)
        del unused_kwargs

        self.func = func
        self.y0 = y0
        self.rtol = rtol if _is_iterable(rtol) else [rtol] * len(y0)
        self.atol = atol if _is_iterable(atol) else [atol] * len(y0)
        self.first_step = first_step
        self.safety = _convert_to_tensor(safety, dtype=torch.float64, device=y0[0].device)
        self.ifactor = _convert_to_tensor(ifactor, dtype=torch.float64, device=y0[0].device)
        self.dfactor = _convert_to_tensor(dfactor, dtype=torch.float64, device=y0[0].device)
        self.max_num_steps = _convert_to_tensor(max_num_steps, dtype=torch.int32, device=y0[0].device)
        #self.n_step_record=[]

def __init__(
        self, func, y0, rtol, atol, first_step=None, safety=0.9, ifactor=10.0, dfactor=0.2, max_num_steps=2**31 - 1,
        **unused_kwargs
    ):
        _handle_unused_kwargs(self, unused_kwargs)
        del unused_kwargs

        self.func = func
        self.y0 = y0
        self.rtol = rtol if _is_iterable(rtol) else [rtol] * len(y0)
        self.atol = atol if _is_iterable(atol) else [atol] * len(y0)
        self.first_step = first_step
        self.safety = _convert_to_tensor(safety, dtype=torch.float64, device=y0[0].device)
        self.ifactor = _convert_to_tensor(ifactor, dtype=torch.float64, device=y0[0].device)
        self.dfactor = _convert_to_tensor(dfactor, dtype=torch.float64, device=y0[0].device)
        self.max_num_steps = _convert_to_tensor(max_num_steps, dtype=torch.int32, device=y0[0].device)
        #self.n_step_record=[]