Skip to content

Pull requests: google/aqt

New pull request New
53 Open 549 Closed
53 Open 549 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Support pl.BlockSpec that holds block_shape with None.
#630 opened May 24, 2024 by copybara-service bot Loading…
Add from_qtensor which transform QTensor into PallasQTensor.
#629 opened May 24, 2024 by copybara-service bot Loading…
Move PallasQTensor materialization code as a method. Call materialization before dequant. This simplifies user interface.
#628 opened May 24, 2024 by copybara-service bot Loading…
cast scale to dequant_type when restoring scale to original shape in pallas.
#627 opened May 24, 2024 by copybara-service bot Loading…
injectable transpose_fn for lhs_scale_transpose_to_output and rhs_scale_transpose_to_output.
#626 opened May 23, 2024 by copybara-service bot Loading…
Allow tiling of dot_general with a tile size
#625 opened May 23, 2024 by copybara-service bot Loading…
[Pallas Support] Add dot_general for pallas.
#624 opened May 22, 2024 by copybara-service bot Loading…
Put numerics and numerics-related logic into QTensor
#623 opened May 22, 2024 by copybara-service bot Loading…
[Pallas support] Add quant_blockwisely function, which quantizes tensor according to the BlockSpec of the inputs.
#596 opened Apr 24, 2024 by copybara-service bot Loading…
Migrate sparsity dependencies to new open source library.
#593 opened Apr 22, 2024 by copybara-service bot Loading…
Regression test on binary pytree stability.
#589 opened Apr 18, 2024 by copybara-service bot Loading…
Rename QTensor to QArray
#587 opened Apr 17, 2024 by copybara-service bot Loading…
Adding new base directory for the open source sparsity library.
#585 opened Apr 15, 2024 by copybara-service bot Loading…
Put numerics and numerics-related logic into QTensor
#583 opened Apr 15, 2024 by copybara-service bot Loading…
Merge sparsity libraries.
#574 opened Apr 8, 2024 by copybara-service bot Loading…
Remove qvalue_astype on make_conv_general_dilated to accelerate of conv
#571 opened Apr 8, 2024 by copybara-service bot Loading…
Adding sparsity functionality for computing channelwise masks.
#561 opened Apr 3, 2024 by copybara-service bot Loading…
Update Gemax to add an option to use the QTensor-saving Freezer.
#553 opened Mar 27, 2024 by copybara-service bot Loading…
This CL adds ridge regression as a method for post-processing the scales, reducing quantization error.
#546 opened Mar 22, 2024 by copybara-service bot Loading…
Improve dtype handling.
#485 opened Feb 16, 2024 by copybara-service bot Loading…
internal
#453 opened Feb 7, 2024 by copybara-service bot Loading…
internal.
#440 opened Feb 6, 2024 by copybara-service bot Loading…
Change the default value of dequant_dtype in QTensor from jnp.float32 to None to avoid silent upcast in the future. Then on the call-side of other libraries, use jnp.float32 for initialization.
#435 opened Feb 2, 2024 by copybara-service bot Loading…
Add support for sharding across multiple contracting axis in local AQT.
#432 opened Feb 1, 2024 by copybara-service bot Loading…
Remove uses of deprecated jax.lax.tie_in
#415 opened Jan 29, 2024 by copybara-service bot Loading…
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.