See https://github.com/llvm/torch-mlir/issues/647#issuecomment-1114328241

This commit adds lowering of `aten.avg_pool2d` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-l...

This commit fixes the `ConstantPad2dStaticModule` test case by adding
the lowering of `aten.pad`...

This patch updates the `torch_mlir::convertTensorToMlirElementsAttr()`
method to enable the cre...

BART, BigBird and GPT2 are not being serialized and hence removed.
Also, changed the script to o...

1. This commit adds lowering of "while-like" prim loop to scf.while
operation.
2. Adds lowering ...

Uses the latest snapshot tags and adds the releases to same asset
directories so it can be run ...

Fix the inplace update tensor issue we had
where the torchscript execution would update the inpu...

This reverts commit 78f5747568f4ae9e95a83f680dc066a185ddff92.

This reverts commit 51dd4625927329f528b549faec90ff0be2d93392.

This reverts commit 1f102cc400a72fa0926eb79f657e347af7c2d142.

This reverts commit 564734b2d7573e66b6b26ac81d35e0df01952102.

This commit decomposes `aten.index_put.hacked_twin` op into
`valsem.aten.index_put_impl` op.

Si...

This commit adds more test cases `aten::index_put` op.
This commit also fixes formatting issues ...

This commit adds lowering of `aten.ceil.float` op.
This commit also fixes formatting for the fil...

Compiling torch-mlir against a source version of PyTorch or an official
wheel compiled with the...

This commit adds lowering of `aten.ne.float_int` op.

Signed-Off By: Vivek Khandelwal <vivek@nod...

This commit adds lowering of `aten.ge.float` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-lab...

This commit adds lowering of `aten.div.float` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-la...

This commit adds lowering of `aten.ge.float_int` op.

Signed-Off By: Vivek Khandelwal <vivek@nod...

I wasn't able to find exactly what frontend situation created it, but
`torch.jit.trace` will som...

The updated LLVM code includes a patch to create bfloat16 array
attributes, thus enabling a dif...

Prior to this patch, the result type for several tensor operations could
only be float32, float...

* Update README.md

* Create development.md

Add a separate development.md file

This commit adds e2e heavydep tests for the torch benchmarks.

Signed-Off By: Vivek Khandelwal <...

Fix filename changed missed in Code Review.

env vars seems to be lost in manylinux docker.
Use a version file like IREE does.

* Add to eager tests to xfail while they are fixed.

Also XFAIL ConstantPad2dStaticModule_basi...

- Add bazel GitHub actions.

Set the environment variable and export it since it doesn't seem to get passed down.

only upload torch*.whl to unblock OSX build failures during upload. We have to move to svenstaro...

Example to demonstrate the extraction of forward as well as
backward graph via Functorch's AOT m...

Prior to this patch, the top-level README did not include the line for
running the Python regres...

Update artifact directory to ./build_tools/python_deploy/wheelhouse/*.whl

Comparing index type instead of integer types solves the problem.

remove now deprecated inputs to build and test

Sets up auto-pinning of latest torch-nightly

use sudo for mac OS

fix build naming

Update pinned pytorch version. Will submit a follow on PR to bump.
Also update artifacts directory

* Add oneshot release snapshot for test/ondemand

Add some build scripts to test new release f...

Added the dynamic registration of return function to the execution
engine. This makes sure that ...

This avoids issues where PyTorch version drift has made things
incompatible.

One caveat is that...

This makes it much easier to convert models and hides all the
ClassAnnotator complexity.

This a...

Since they run in distinct jobs, using the same ccache would
cause one job to overwrite the cach...

That way, downstreams don't have to duplicate this list.

Also, remove "external config" feature...

This commit adds lowering of `max_pool2d_with_indices` op.

Signed-Off By: Vivek Khandelwal <viv...

* shape: add shape transfer function for aten.neg

Prior to this patch, the list of shape tran...

This commit adds lowering of `aten::max_pool2d_with_indices_backward` op.

This commit also fixe...

Builds whl package for OSX. Need to validate smoke tests next

NB: `shouldnt_normalize2` and `shouldnt_normalize3` currently XPASS i.e., args *will* successful...

RefineTypes doesn't handle shape refinement anymore.

torch.layout information has been added.

This PR adds rules for building the compiler part with bazel, a followup PRs will build the pyth...

The lowering of `aten.ne.Scalar` op has been added to
the linalg backend.

This commit adds the following support to the op `nll_loss_backward`:
- `input` tensor can be r...

Follows up on #623 for out-of-tree builds of torch-mlir, which
added building `torch-mir-dialec...

This commit adds support for multi-dimensional tensors as input to the
`_index_put_impl_` op. T...

Signed-off-by: Anup Gangwar <[email protected]>

Co-authored-by: Anup Gangwar <anup.gangwar...

The issue was in the canonicalizer for torch.aten.ge.int -- in cases
where the operands were swa...

This should make the releases sort properly when `pip`'s
`-f`/`--find-links` argument is used.

This commit decomposes different variants of `aten.where.*` op into
`aten.where.Self` op. It cov...

This commit decomposes `aten.new_empty` op into `aten.empty.memory_format` op.

This commit also...

A recent PyTorch commit made ConstantPad2d call a helper function with a
`Union[int, float]` typ...

As per the docs on:
https://github.com/eregon/publish-release

> Note that the release must *not...

I am investigating the breakage.

Also, fix "externals" rename in setup.py and some cases where ...

The reified code to compute the shape of torch.aten.constant_pad_nd
uses negative indices when ...

The logic in the rewriting phase had a bug in case of a read-only op
coming before mutation ops....

That was back from an earlier stage in the project when e2e was a big
deal because we didn't hav...

- update diagram to use the name "Eager Mode" instead of
`torch.dispatch`, which wasn't a very...