ENH: Low memory adjoint #145
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit is an upgrade of the adjoint model allowing time step checkpoints to lower the maximum memory usage. By default, in the forward sweep of the adjoint, at each time, n arrays of shape (nrow, ncol) where pushed. This arrays were poped at the end of the forward sweep which result in some case (large domain and time steps) to a too high memory peak. This commit adds checkpoints in the forward sweep to reduce the memory peak and pass the variables that must be checkpointed at each time (i.e. fluxes and states) to a vector of active cell instead of a grid of number of rows and columns. This upgrade should be pretty efficient for large domain and time step calibration
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This commit is an upgrade of the adjoint model allowing time step checkpoints to lower the maximum memory usage. By default, in the forward sweep of the adjoint, at each time, n arrays of shape (nrow, ncol) where pushed. This arrays were poped at the end of the forward sweep which result in some case (large domain and time steps) to a too high memory peak. This commit adds checkpoints in the forward sweep to reduce the memory peak and pass the variables that must be checkpointed at each time (i.e. fluxes and states) to a vector of active cell instead of a grid of number of rows and columns.
This upgrade should be pretty efficient for large domain and time step calibration