chad-fast - [no description]

	Commit message (Collapse)	Author	Age
*	Implement reshape	Tom Smeding	5 days
\|
*	Better fold D{1,2} primitives	Tom Smeding	7 days
\|
*	WIP fold: everything but Compile (slow, but should be sound)	Tom Smeding	11 days
\|
*	WIP fold: Implement D[fold1i]	Tom Smeding	12 days
\| \| \| \|	Still need to handle the new primitives in the rest of the library
*	More precise reconstructBindings interface	Tom Smeding	12 days
\|
*	No need for assertSubenvEmpty in D[build]	Tom Smeding	12 days
\|
*	Helper functions bpush and weakenBindingsE	Tom Smeding	12 days
\|
*	Complete occCountX	Tom Smeding	2025-10-08
\|
*	WIP pruneExpr in AST.Count	Tom Smeding	2025-07-23
\|
*	Don't introduce sparsity if zero is cheap	Tom Smeding	2025-06-18
\|
*	Don't destroy effects in sparse plus	Tom Smeding	2025-06-18
\|
*	Put smart accumulator redirection behind config flag	Tom Smeding	2025-06-18
\|
*	Give DeepZero to With	Tom Smeding	2025-06-18
\|
*	Tests pass, should check if output is sensible	Tom Smeding	2025-06-18
\|
*	CHAD.hs compiles	Tom Smeding	2025-06-16
\|
*	WIP	Tom Smeding	2025-06-09
\|
*	More WIP sparsity	Tom Smeding	2025-06-08
\|
*	WIP mixed static/dynamic sparsity	Tom Smeding	2025-06-06
\|
*	ERecompute	Tom Smeding	2025-05-25
\|
*	Reorder TLEither to after TEither	Tom Smeding	2025-04-29
\|
*	WIP revamp accumulators again: explicit monoid types	Tom Smeding	2025-04-27
\| \| \| \| \| \| \| \|	No more D2 in accumulators! Paving the way for configurable sparsity of products and arrays. The idea is to make separate monoid types for a "product cotangent" and an "array cotangent" that can be lowered to either a sparse monoid or a non-sparse monoid. Downsides of this approach: lots of API duplication.
*	Add integer modulo operator	Tom Smeding	2025-04-21
\|
*	Some progress with accumMap	Tom Smeding	2025-04-18
\|
*	Populate accumMap	Tom Smeding	2025-04-06
\|
*	Pass around an accumMap (but it's empty still)	Tom Smeding	2025-03-28
\|
*	D2[Array] now has a Maybe instead of zero-size for zero	Tom Smeding	2025-03-26
\| \| \| \|	Remaining problem: 'add' in Compile doesn't use the D2 stuff
*	Commutativity marker on fold1i	Tom Smeding	2025-03-20
\|
*	Complete accumulator revamp!	Tom Smeding	2025-03-15
\|
*	Just allow non-array accumulator types	Tom Smeding	2025-03-09
\| \| \| \|	Not sure why I didn't allow this previously
*	Clean up code organisation a little	Tom Smeding	2025-03-09
\|
*	Add ext field to remaining AST constructors	Tom Smeding	2025-01-27
\|
*	WIP UnMonoid (to be used for compiling to C)	Tom Smeding	2024-12-06
\|
*	Fix indentation	Tom Smeding	2024-12-06
\|
*	WIP accum top-level args	Tom Smeding	2024-11-26
\|
*	Prepare for introducing top-level args in accum mod	Tom Smeding	2024-11-23
\|
*	Use accum storage for Case too	Tom Smeding	2024-11-23
\|
*	drevScoped returns a data type, not CPS	Tom Smeding	2024-11-23
\|
*	Factor let storage-dependent scoping logic into separate function	Tom Smeding	2024-11-22
\|
*	Configuration for CHAD	Tom Smeding	2024-11-14
\|
*	Make EBuild derivative aware of zero cotangent arrays	Tom Smeding	2024-11-11
\|
*	Test GMM; it fails	Tom Smeding	2024-11-10
\|
*	Complete GMM implementation	Tom Smeding	2024-11-10
\|
*	Some more primitive operators	Tom Smeding	2024-11-09
\|
*	WIP maximum/minimum	Tom Smeding	2024-11-08
\|
*	Custom derivatives	Tom Smeding	2024-11-08
\|
*	Remove build1	Tom Smeding	2024-11-07
\|
*	Benchmark	Tom Smeding	2024-11-07
\|
*	Generate EOneHot in D[EIdx]	Tom Smeding	2024-11-05
\| \| \| \| \| \|	This generates a one-hot for the zero-dimensional inner array because indexing one level further to the actual element is too difficult. But this should simplify away fine.
*	WIP EOneHot	Tom Smeding	2024-11-04
\|
*	Store only on tape what's used in the dual	Tom Smeding	2024-10-29
\|