Dan Foreman-Mackey cb4d97aa1f Move jex.ffi to jax.ffi.

2024-12-29 13:06:19 +00:00

..

control_flow

[Jax][Pallas][Mosaic] Implement platform dependent diag, with branch selection driven by constant prop in mosaic lowering.

2024-12-22 00:50:51 -08:00

__init__.py

Change JAX's copyright attribution from "Google LLC" to "The JAX Authors.".

2022-09-22 12:27:19 -07:00

ann.py

Clean up and fix primal type to tangent type mapping

2024-09-18 13:43:54 -07:00

convolution.py

Clone RandomAlgorithm into lax.py, instead of using the version from XLA.

2024-10-11 11:03:15 -07:00

eigh.py

Partial rollback of https://github.com/google/jax/pull/23353 as discussed in https://github.com/google/jax/pull/23353#issuecomment-2326604708

2024-09-03 09:49:22 -07:00

fft.py

Clone RandomAlgorithm into lax.py, instead of using the version from XLA.

2024-10-11 11:03:15 -07:00

lax.py

Merge pull request #25648 from hawkinsp:warnings3

2024-12-20 13:41:43 -08:00

linalg.py

Move jex.ffi to jax.ffi.

2024-12-29 13:06:19 +00:00

other.py

Simplify and consolidate dot algorithm control in lax.

2024-10-07 13:21:34 -07:00

parallel.py

Clarify documentation for output_offsets operand of ragged_all_to_all.

2024-12-20 07:52:11 -08:00

qdwh.py

Partial rollback of https://github.com/google/jax/pull/23353 as discussed in https://github.com/google/jax/pull/23353#issuecomment-2326604708

2024-09-03 09:49:22 -07:00

slicing.py

Cleanup: replace lax._abstractify with core.get_aval

2024-12-12 14:08:17 -08:00

special.py

Use MLIR generated convenience functions athing(...) instead of writing AThingOp(...).result.

2023-11-17 11:47:14 -08:00

stack.py

Partial rollback of https://github.com/google/jax/pull/23353 as discussed in https://github.com/google/jax/pull/23353#issuecomment-2326604708

2024-09-03 09:49:22 -07:00

svd.py

Partial rollback of https://github.com/google/jax/pull/23353 as discussed in https://github.com/google/jax/pull/23353#issuecomment-2326604708

2024-09-03 09:49:22 -07:00

utils.py

[sharding_in_types] Check out_avals with mesh context too. This is because users can pass their own shardings to functions like einsum, reshape, broadcast`, etc

2024-12-18 14:42:40 -08:00

windowed_reductions.py

Cleanup: replace lax._abstractify with core.get_aval

2024-12-12 14:08:17 -08:00