Draft
Conversation
* Add nanodrr package * Add jaxtyping dependency * Update pose sampler * Replace DiffDRR components in trainer * Monkey-patch loss function * Update load subject function * Fix devices and dtypes * Start deprecating MVC * Remove RigidTransforms in majority of functions
* Port loss functions and SE(3) utilities * Remove unused args * Fix single-subject loading
* Switch to local roma fork * Get compile working on the training step * Remove unused args * Update pre-commit hooks * Fix imports
* Update roma * Fuse the optimizer * Make XrayTransforms compilable * Skip cudagraphs (for now) * Make augmentations compilable * Set up bf16 autocasting * Update roma dependency
* Simplify load function * Wrap metrics in a NamedTuple * Mask loss from poor samples
* Update packages * Increase num_samples per image * Remove old args * Fix bf16 dtypes * Update * Optimize queue loader * Update packages * Add optional flags for compile/bf16 * Simplify DRR plotting * Fix module name * Fix flags * Fix logged images
* Fix nanodrr imports * Refactor initialization to a submodule * Move network, scheduler, and loss to a submodule * Make submodule for data ops * Make submodule for data ops * Go full tensor * Remove unused augmentation * Fix loss mean
* Update uv.lock * Add badges to docs
Upgrade rendering backend to nanodrr
* Load affine from numpy * Simplify imports
* Construct subject iterator entirely in submodule * Reorder trainer functions from first to last * Simplify rendering function * Add to-center coordinate frame * Fix isocenter type * Add some type annotations * Recenter poses at origin before dgeo * Rename submodule datum -> data * Update packages * Revert changes to xframe * Remove ipynb checkpoints * Simplify autocast * Make sure to change dtype of subject * Add setuptools for compile
* Make a class for intrinsics * Update intrinsics resampling * Add typehints * Implement a draft of new Register class * Add a dataclass to hold registration results * Add docstrings * Deprecate the `registrar` module
* Rename kwarg * Move Intrinsics to separate file * Deprecate useless module * Deprecate animations * Deprecate dcm2nii
* Add docstrings * Deprecate configs * Sync * Add cyclopts commands * Ignore line length in the CLI module * Add test for drift in the API/CLI * Add tests.yml
* Save fresh copies of the iterations * Add animate function * Sync
* Prevent line-length linting on config files * Reformat configs
* Move parse_scales to base class * Better data transfer * Save the datapaths * Separate data and modules in logging * Update animation to new loggers * Add logging * Change to __call__ method * Remove Subject from save * Fix circular imports * Remove bloated types module
* Update packages * Add savepath arg to CLI * Switch CLI to __call__ * Change savepath to be a folder * Fix GIF savepath
* Vary mu_bone every iteration * Annotate outputs * Enable high-precision matmuls * Annotate output * Update dependency path
* Fix autocomplete in subclasses * Allow user to set different patience per scale * Clean up signature manipulation * Fix single patience parameter * Fix environment * Add test to check for CLI drift * Fix drift * Suppress unnecessary warning
* Make SO(3) utils pure torch * Remove diffdrr
* Update packages * Deprecate RandomHUToMu
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Change the rendering backend to
nanodrr, our performance-oriented reimplementation ofDiffDRR.Closes #31
Closes #94
Closes #117
Closes #118