Skip to content

Revise 0.15.0 HISTORY.md #162

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

Merged
shravanngoswamii merged 2 commits into from
May 19, 2026
+9 −2
Merged

Revise 0.15.0 HISTORY.md #162

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dbcedfb
Revise 0.15.0 HISTORY.md
shravanngoswamii May 19, 2026
0dc87d1
format
shravanngoswamii May 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 9 additions & 2 deletions HISTORY.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
## 0.15.0

New evaluator-preparation and AD interface: `prepare` binds a callable to a sample input (vector or `NamedTuple`); `value_and_gradient!!` / `value_and_jacobian!!` return value-and-derivative pairs from the resulting `Prepared` wrapper. The `!!` suffix signals the returned derivative may alias the cache — copy if you need to keep it.
Added a new evaluator-preparation interface for AD backends. `prepare` binds a callable to a sample input, either a vector or a `NamedTuple`, and returns a `Prepared` wrapper. `value_and_gradient!!` and `value_and_jacobian!!` then compute value-and-derivative pairs from that wrapper. The `!!` suffix means the returned derivative may alias backend cache storage, so copy it if you need to keep it across calls.

```julia
using ADTypes, Mooncake # or DifferentiationInterface + ForwardDiff
Expand All @@ -10,7 +10,14 @@ val, grad = value_and_gradient!!(prepared, [1.0, 2.0, 3.0])
# val == -7.0; grad == [-1.0, -2.0, -3.0]
```

Two new AD-backend extensions ship with it: `AbstractPPLDifferentiationInterfaceExt` (any DI backend) and `AbstractPPLMooncakeExt` (`AutoMooncake`, `AutoMooncakeForward`). `AbstractPPLTestExt` gains a conformance harness via `generate_testcases` / `run_testcases` (reserved groups: `:vector`, `:namedtuple`, `:edge`, `:cache_reuse`).
Added two AD-backend extensions:

- `AbstractPPLDifferentiationInterfaceExt`, supporting DifferentiationInterface backends.
- `AbstractPPLMooncakeExt`, supporting `AutoMooncake` and `AutoMooncakeForward`.

The new evaluator path supports scalar gradients, vector-output Jacobians, `NamedTuple` inputs, reusable AD caches, input-shape checks, and constant context arguments via `context::Tuple`.

`AbstractPPLTestExt` now provides a small AD conformance harness through `generate_testcases` and `run_testcases`, with reserved groups `:vector`, `:namedtuple`, `:edge`, and `:cache_reuse`.

See [`docs/src/evaluators.md`](docs/src/evaluators.md) for the full interface, the `check_dims` and `context::Tuple` options, the `NamedTuple` input path, and extension-author guidance.

Expand Down
Loading