summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Pretty: fix printing of EMaybeTom Smeding8 days
|
* Compile: Fix formatting of generated codeTom Smeding8 days
|
* Fix refcount bug with indexing for nested arraysTom Smeding8 days
|
* test: regresion-accum1 (still failing)Tom Smeding8 days
|
* test: Respect shape lower bounds betterTom Smeding8 days
|
* Compile: Fix debugPrintAST formattingTom Smeding8 days
|
* test-framework: More matchable paths (begin/end anchors)Tom Smeding8 days
|
* Compile: Decrement accumval only when added, not movedTom Smeding9 days
|
* Compile: remove unused functionTom Smeding9 days
|
* Compile: Better alloc & refc debug facilitiesTom Smeding9 days
|
* test: Some more Compile tests (still passing, but code still broken)Tom Smeding9 days
|
* test: type R = TScal TF64Tom Smeding9 days
|
* test: Start of a list of compile testsTom Smeding9 days
|
* Fix various issues in Compile (still broken)Tom Smeding9 days
|
* Language: support Maybe and Accum termsTom Smeding9 days
|
* Compile.Exec: also remove tmpdir when compilation failsTom Smeding9 days
|
* Rules: small fixesTom Smeding9 days
|
* D2[Array] now has a Maybe instead of zero-size for zeroTom Smeding9 days
| | | | Remaining problem: 'add' in Compile doesn't use the D2 stuff
* Update rules.texTom Smeding9 days
|
* Compile: More checksTom Smeding12 days
|
* Compile: More debugging machineryTom Smeding13 days
|
* test-framework: --hedgehog-replayTom Smeding13 days
|
* Compile: Don't toList the Bag in 'scope'Tom Smeding14 days
|
* Compile.Exec: Better diagnostics on GCC errorTom Smeding14 days
|
* Compile: First compilation of fold1iTom Smeding14 days
|
* Compile: Decrement array elements when freeing arrayTom Smeding14 days
|
* Compile: Add some forgotten refcount decrementsTom Smeding2025-03-21
|
* Compile: compileAssign helper functionTom Smeding2025-03-21
|
* Commutativity marker on fold1iTom Smeding2025-03-20
|
* test: Compile final gradient function (WIP)Tom Smeding2025-03-17
| | | | This doesn't work yet because Compile doesn't yet support EFold1Inner
* test fwd: test with simplified expressionTom Smeding2025-03-17
|
* Compile EAccumTom Smeding2025-03-17
|
* simplifyOneHotTermTom Smeding2025-03-16
|
* Complete accumulator revamp!Tom Smeding2025-03-15
|
* test-framework: Allow filtering tests by substringTom Smeding2025-03-15
|
* Much process with accumulator revampTom Smeding2025-03-14
|
* WIP revamp accumulator projection type reprTom Smeding2025-03-14
| | | | | | | I stopped working on this because I realised that having sparse products (and coproducts, prehaps) everywhere is a very bad idea in general, and that we need to fix that first before really being able to do anything else productive with performance.
* test: Split adTestGen into one function per test caseTom Smeding2025-03-11
| | | | This improves (compactifies) hedgehog output
* test-framework: Even more compact output with testGroupCollapseTom Smeding2025-03-11
|
* test-framework: Improve outputTom Smeding2025-03-10
|
* More compact test outputTom Smeding2025-03-10
|
* test: Also test pre-simplified termTom Smeding2025-03-09
|
* test: Run gradientByForward with compiled DN funTom Smeding2025-03-09
|
* test: Make GMM inputs a little bit smallerTom Smeding2025-03-09
|
* Just allow non-array accumulator typesTom Smeding2025-03-09
| | | | Not sure why I didn't allow this previously
* Clean up code organisation a littleTom Smeding2025-03-09
|
* idana: CleanupTom Smeding2025-03-07
|
* pretty: Print x value for (!)Tom Smeding2025-03-07
|
* idana: Track array shapesTom Smeding2025-03-07
|
* pprintExprTom Smeding2025-03-07
|
generated by cgit v1.2.3-70-g09d2 (git 2.48.1) at 2025-04-04 23:43:11 +0200