lululxvi / deepxde

A library for scientific machine learning and physics-informed learning
https://deepxde.readthedocs.io
GNU Lesser General Public License v2.1
2.64k stars 741 forks source link

fpinn相关 #1014

Open FloraGong opened 1 year ago

FloraGong commented 1 year ago

咨询您如果周期边界或者非0边界如何处理fpinn中的分数阶拉普拉斯算子离散。

lululxvi commented 1 year ago

I don't know.

wangxiao2019 commented 1 year ago

Hello! Can you integrate the Fractional and FractionalTime classes from fpde into the model? This is because when computing the general form of spatial or temporal fractional partial differential equations, there's an issue of 'FractionalTime' object has no attribute 'losses_train'.

lululxvi commented 1 year ago

I don't understand your question. Can you be more specific?

wangxiao2019 commented 1 year ago

I don't understand your question. Can you be more specific? "For example, I want to solve a spatial two-dimensional fractional-order and time one-dimensional integer-order equation as follows: $\frac{\partial u(x, y, t)}{\partial t}+(-\Delta)^{\alpha / 2} u(x, y, t)=f(x, y, t)$ For this, I set its geometric domain as: geom = dde.geometry.Disk([0, 0], 1) timedomain = dde.geometry.TimeDomain(0, 1) geomtime = dde.geometry.GeometryXTime(geom, timedomain)

Then, when I call the dde.data.TimeFPDE class, data = dde.data.TimeFPDE( geomtime, fpde, alpha, [bc, ic], [20, 100], num_domain=50, num_boundary=50, solution=func, meshtype="dynamic", num_test=1000 ) I get an error.

How should I set up the code for such two-dimensional or three-dimensional fractional-order equations with time dependence?"

lululxvi commented 1 year ago

It is strange. Do other demo examples work?