Condition takes only the following keyword BUG

[LoveFrootLoops]

Upon running the Tutorial 1 Code (copy-pasted), an error message appeared instantly:

Traceback (most recent call last):
File "C:\Users\PycharmProjects\pythonProject\venv\lib\site-packages\IPython\core\interactiveshell.py", line 3508, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "", line 1, in
runfile('C:\Users\PycharmProjects\pythonProject\test.py', wdir='C:\Users\PycharmProjects\pythonProject')
File "C:\Program Files\JetBrains\PyCharm 2023.1.1\plugins\python\helpers\pydev_pydev_bundle\pydev_umd.py", line 198, in runfile
pydev_imports.execfile(filename, global_vars, local_vars) # execute the script
File "C:\Program Files\JetBrains\PyCharm 2023.1.1\plugins\python\helpers\pydev_pydev_imps_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "C:\Users\PycharmProjects\pythonProject\test.py", line 82, in
class SimpleODE(SpatialProblem):
File "C:\Users\PycharmProjects\pythonProject\test.py", line 110, in SimpleODE
'x0': Condition(Span({'x': 0.}), initial_condition),
File "C:\Users\PycharmProjects\pythonProject\venv\lib\site-packages\pina\condition.py", line 69, in init
raise ValueError('Condition takes only the following keyword arguments: {input_points, output_points, location, function, data_weight}.')
ValueError: Condition takes only the following keyword arguments: {input_points, output_points, location, function, data_weight}.

[ndem0]

Dear @LoveFrootLoops, thanks for reporting!
Yes, tutorials need to be updated to the current version of the package (we changed some methods in the last commits). Hope to fix it asap, I'll notify here when we'll finish!

[dario-coscia]

@ndem0 I am on it to change the tutorial

[dario-coscia]

👋🏻 @LoveFrootLoops Thank you for the report.

Recently we changed the class Condition, which now takes only the following keyword arguments: input_points, output_points, location, function, data_weight. Unfortunately the documentation is not updated on the website yet, as we do it automatically once a month. Anyway you can look at the the class docstrings for reference

Now the tutorials should all work, hope it helps.

[dario-coscia]

@ndem0 If for you it is ok, it can be merged

[LoveFrootLoops]

Thanks for the fast response! @dario-coscia Is it still possible to include multiple functions in a condition, such as "function=[func1, func2]"?

[dario-coscia]

Thanks for the fast response! @dario-coscia Is it still possible to include multiple functions in a condition, such as "function=[func1, func2]"?

No, it is not possible. The kwarg function must be a callable function. Also I am wondering, why do you need it? Generally, in a PINN learning approach you want to minimize the residual obtained in a specific location between the function of interest and the expected output. Yes, the function can be scalar/ vectorial and multivariate, but it is always one.

Tell me if it helps. For reference I suggest you Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next, which is a nice and concise introduction to PINNs and Neural Operators.

[LoveFrootLoops]

Thanks for the fast response! @dario-coscia Is it still possible to include multiple functions in a condition, such as "function=[func1, func2]"?

No, it is not possible. The kwarg function must be a callable function. Also I am wondering, why do you need it? Generally, in a PINN learning approach you want to minimize the residual obtained in a specific location between the function of interest and the expected output. Yes, the function can be scalar/ vectorial and multivariate, but it is always one.

Tell me if it helps. For reference I suggest you Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next, which is a nice and concise introduction to PINNs and Neural Operators.

Good morning ;)
If I have multiple cost functions for the same domain/location I would now have to define multiple conditions. That’s why I asked if it possible to define multiple functions for the same domain without repetition.

[dario-coscia]

@LoveFrootLoops can you send me the snippet of code you are using? I don't understand why you should minimize the same function on the same location.

[LoveFrootLoops]

Let’s assume I have two PDEs which I want to solve for the same function “u” ont he same domain Omega, let’s call them PDE1 and PDE2. Then I would have to do the following:

conditions = {
'Omega1' Condition(location=Span({'x': [0, 1], 'y': [0, 1]}), function=PDE1),
'Omega2' Condition(location=Span({'x': [0, 1], 'y': [0, 1]}), function=PDE2)
}

instead of:

conditions = {
'Omega' Condition(location=Span({'x': [0, 1], 'y': [0, 1]}), function=[PDE1,PDE2])
}

[dario-coscia]

I dont really understand. I have never seen that in the same domain more conditions are applied. Can you send me the reference of equations you are dealing with?

If it's something useful we can think to make a pull request to make it in the next release.

[LoveFrootLoops]

Although my demonstration involved PDEs, I want to clarify that my point applies to a broader range of scenarios. For instance, restrictions such as the positivity of 'u' over the domain Omega can be effectively addressed with this approach. However, at present, I must define a new condition for each constraint, even if it pertains to the same domain.

Additionally, if I understand correctly, PINA allows for minimizing 'u' given specific input and output data. However, how about handle input and output data that correspond to derivatives of the function 'u.'
In mechanics, there is another approach to consider where multiple functions correspond to each other through a mapping but need to solve different PDEs over the same domain Omega. For instance, the functions "strain," "stress," and "displacement" may need to be determined. Stresses are subject to the PDE 'div(stress)=f,' while strains must satisfy the PDE 'grad(displacement)=strain.' These two functions, strains and stresses, are related through a mapping 'C(strain)=stress,' which may be quite complex.

In this scenario, three conditions would be required over the same domain. It's worth noting that this approach has applications in various fields, including engineering and physics.

[dario-coscia]

Although my demonstration involved PDEs, I want to clarify that my point applies to a broader range of scenarios. For instance, restrictions such as the positivity of 'u' over the domain Omega can be effectively addressed with this approach. However, at present, I must define a new condition for each constraint, even if it pertains to the same domain.

Additionally, if I understand correctly, PINA allows for minimizing 'u' given specific input and output data. However, how about handle input and output data that correspond to derivatives of the function 'u.'

In mechanics, there is another approach to consider where multiple functions correspond to each other through a mapping but need to solve different PDEs over the same domain Omega. For instance, the functions "strain," "stress," and "displacement" may need to be determined. Stresses are subject to the PDE 'div(stress)=f,' while strains must satisfy the PDE 'grad(displacement)=strain.' These two functions, strains and stresses, are related through a mapping 'C(strain)=stress,' which may be quite complex.

In this scenario, three conditions would be required over the same domain. It's worth noting that this approach has applications in various fields, including engineering and physics.

Thank you for the reply. Indeed you are right, it might be useful to insert this possibility. I have opened a new issue #92 for inserting this possibility. Tell me if you are interested working on it.

I will close this issue once the tutorials are merged by @ndem0. To follow the discussion please use #92