Wenyin Wei

Since `@autojit` has been deprecated now, would you mind supplementing a little modification, change `@autojit` to `@jit`

Hello all, I am also disappointed at the performance of plotly.py because my dataset is not that small. plotly only utilized one of my CPUs to exectute its chart drawing....

Thank you for your prompt reply on the issue. I didn't put xtensor as a subpackage yet indeed, but installed it. let me try some more commands on linking xtensor.

Yes! I am also pretty confused. How can I meet the error even after I have cmaked 'xtensor' and installed it by building INSTALL target in visual studio. Let me...

I made sure that I did meet the trouble with the env: - Windows 10.0.19041.0, - CMake: 3.21.0 with option `XTENSOR_USE_OPENMP` ON, - xsimd: 7.6.0, xtensor: 0.23.10 - VS 2019...

The problem could be easily fixed by prepending namespace `xt::` before `to_array`

The `self.matrix()` suggestion still failed to match Christoffel symbols, cause generally a matrix is 2-order, how about `self.array()`?

The same argument also applies to the Levi-Civita symbol (not yet exist in einsteinpy), which are not tensors indeed but can `tensor_product` and `tensor_cross` with other tensors in the way...

I would collect more info and record it here, about how other symbol engines handle such stuff.

Could we add an attribute to the symbol definition like `phi = symbols('phi', period=2*sp.pi / periodic_2pi = True)` so that sympy knows the `2*pi*n` makes no difference to the resulting...