Version 0.2 #312
Version 0.2 #312
Conversation
|
|
||
| class DomainEquationCondition(ConditionInterface): | ||
| """ | ||
| Condition for input/output data. |
There was a problem hiding this comment.
typos to fix, let's remember to do it later!
| from abc import ABCMeta | ||
|
|
||
|
|
||
| class Optimizer(metaclass=ABCMeta): # TODO improve interface |
There was a problem hiding this comment.
I really like both Optimizer and Scheduler classes! I think the hook method can be made an abstract method for the base classes. In this way all the classes that inherit from those need to have this function (which is fundamental to attach the network parameters to optimize).
The only thing that I am concerned about is the back compatibility with Lightning, as by default Lightning will use built-in functions inherited from Torch Optimizers and Schedulers. Maybe in the base classes Optimizer and Scheduler we should inherit from torch base classes of optimizer and scheduler as well?
| @@ -0,0 +1,10 @@ | |||
| __all__ = [ | |||
| 'Condition', | |||
There was a problem hiding this comment.
I agree to divide conditions in multiple classes but I do not think that the residual method should be inside conditions. I see conditions more like an object that encapsulates data, if we constrain also to what the model should do we can't cover all test cases. I think that the loss to minimize should be defined only inside the solver (physics-informed, data-driven whatever), in this way is user responsibility to know how to use the data
Maybe we can think more of something like PhysicsConditions, DataConditions, GraphConditions (like the division in the dataloading)... These conditions are needed to call loss_data, loss_phys, and loss_graph (which are defined inside the solver). This helps us to do two things:
- We know which functions need to be defined in the solver by looking at the conditions the problem has (we can raise errors before training)
- We don't need to cover all classes of possible ways residuals are defined, we only need to pass the data that are needed to compute it.
| return tmp | ||
| from .utils import check_consistency | ||
| check_consistency(labels, dict) | ||
|
|
There was a problem hiding this comment.
I don't understand completely this. Is this back-compatible? Do we need to specify for each tensor all labels for all columns or is it done automatically? What is the format?
dario-coscia
added
pr-to-review
* Tutorial Equation 12 * .rst and readme fix for tutorial12 * small fix * modified rst --------- Co-authored-by: Dario Coscia <[email protected]>
* Update label_tensor.py cpu/gpu * Update test_adaptive_refinment_callbacks.py * Update test_optimizer_callbacks.py
* gpinn/basepinn new classes, pinn restructure * codacy fix gpinn/basepinn/pinn * inverse problem fix * Causal PINN (#267) * fix GPU training in inverse problem (#283) * Create a `compute_residual` attribute for `PINNInterface` * Modify dataloading in solvers (#286) * Modify PINNInterface by removing _loss_phys, _loss_data * Adding in PINNInterface a variable to track the current condition during training * Modify GPINN,PINN,CausalPINN to match changes in PINNInterface * Competitive Pinn Addition (#288) * fixing after rebase/ fix loss * fixing final issues --------- Co-authored-by: Dario Coscia <[email protected]> * Modify min max formulation to max min for paper consistency * Adding SAPINN solver (#291) * rom solver * fix import --------- Co-authored-by: Dario Coscia <[email protected]> Co-authored-by: Anna Ivagnes <[email protected]> Co-authored-by: valc89 <[email protected]> Co-authored-by: Monthly Tag bot <[email protected]> Co-authored-by: Nicola Demo <[email protected]>
Co-authored-by: dario-coscia <[email protected]>
No description provided.