ode

Hi I would like to propose a better implementation for 'test_indices':

We can remove the unneeded np.array casting:

Cleaner/New:
test_indices = list(set(range(len(texts))) - set(train_indices))

Old:
test_indices = np.array(list(set(range(len(texts))) - set(train_indices)))

The basics of DiscreteSystem exists now, but the little details around it need to be finished.

there need more results of predictions for a system of PDE and compare with reliable results.

http://eqworld.ipmnet.ru/en/solutions/syspde.htm

We really should be testing each and every feature of Latexify and currently we are not.

Test generation is made easy by a macro that we supply:

using Latexify
@Latexify.generate_test latexify("x/y")

generates a test and puts it in your clipboard to be pasted:

@test latexify("x/y") == 
raw"$\frac{x}{y}$"

One just have to make sure that the test does ac

Hi,

I am pretty new to neurodiffeq, thank you very much for the excellent library.

I am interested in the way, and the computational speed, of computing partial derivatives w.r.t. the inputs.

Take forward ODE (1D, 1 unknown variable) solver for example, the input is x, a batch of coordinates, and the output of the neural network is y, the approximated solution of the PDE at these coo

Currently, values for the llh, gradient, the computed trajectories of states and observables, or the sensitivities are checked in unit tests.
Unfortunately, some bugs, such as incorrect Jacobians or switched minus signs in the Newton solver, will not necessarily affect those quantities. However, they will substantially impact solver performance, by causing way too many steps to be taken. Hence, w