Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Missing eq_param_index when loading BundleSolution #211

Open
ptflores1 opened this issue Dec 30, 2023 · 4 comments · May be fixed by #212
Open

Missing eq_param_index when loading BundleSolution #211

ptflores1 opened this issue Dec 30, 2023 · 4 comments · May be fixed by #212

Comments

@ptflores1
Copy link

When calling BundleSolution1D.load() the new instance is initialized without an eq_param_index argument. Since the _diff_eqs_wrapper function uses the local variable eq_param_index there is no way of recovering the original value and the functionality is lost.
@ptflores1
Copy link
Author

The code in

neurodiffeq/neurodiffeq/solvers_utils.py

Lines 519 to 533 in 1b6d8c5

elif load_dict["type_name"] == "BundleSolver1D" or load_dict["parent_type_name"] == "BundleSolver1D":
t_min = load_dict['solver'].r_min[0]
t_max = load_dict['solver'].r_max[0]
solver = cls(ode_system=de_system,
conditions=cond,
metrics=load_dict['metrics'],
nets=nets,
optimizer=optimizer,
train_generator=train_generator,
valid_generator=valid_generator,
t_min=t_min,
t_max=t_max,
theta_min=tuple(load_dict['solver'].r_min[1:]),
theta_max=tuple(load_dict['solver'].r_max[1:]))

can be modified to

elif load_dict["type_name"] == "BundleSolver1D" or load_dict["parent_type_name"] == "BundleSolver1D": 
     t_min = load_dict['solver'].r_min[0] 
     t_max = load_dict['solver'].r_max[0] 
  
     solver = cls(ode_system=de_system, 
                  conditions=cond, 
                  metrics=load_dict['metrics'], 
                  nets=nets, 
                  optimizer=optimizer, 
                  train_generator=train_generator, 
                  valid_generator=valid_generator, 
                  t_min=t_min, 
                  t_max=t_max, 
                  theta_min=tuple(load_dict['solver'].r_min[1:]), 
                  theta_max=tuple(load_dict['solver'].r_max[1:]),
                  eq_param_index=(index - len(cond) - 1 for index in load_dict['solver'].eq_param_index)) # new line
)

@ptflores1
Copy link
Author

@shuheng-liu would that be okay for a PR?

@sathvikbhagavan sathvikbhagavan linked a pull request Dec 31, 2023 that will close this issue
Open
@sathvikbhagavan
Copy link
Collaborator

sathvikbhagavan commented Dec 31, 2023 •
edited
Loading

Thanks for the report! I don't think we need to recompute as it is stored in https://github.com/NeuroDiffGym/neurodiffeq/blob/master/neurodiffeq/solvers.py#L1354

It should be fixed in #212

@ptflores1
Copy link
Author

Great! Thanks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

refactor: load eq_param_index of BundleSolvers correctly NeuroDiffGym/neurodiffeq
2 participants
@ptflores1 @sathvikbhagavan