Skip to content
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

Jump to bottom

MatMulNBits + Add fusion #20587

Merged
merged 30 commits into from
May 16, 2024
Merged

MatMulNBits + Add fusion #20587

merged 30 commits into from
May 16, 2024

Commits on Apr 30, 2024

  1. save work

    @edgchen1
    edgchen1 committed Apr 30, 2024
    Configuration menu
    Copy the full SHA
    350f3e2 View commit details
    Browse the repository at this point in the history

Commits on May 1, 2024

  1. handle bias in fallback impl

    @edgchen1
    edgchen1 committed May 1, 2024
    Configuration menu
    Copy the full SHA
    752b7c9 View commit details
    Browse the repository at this point in the history

Commits on May 2, 2024

  1. reorder includes

    @edgchen1
    edgchen1 committed May 2, 2024
    Configuration menu
    Copy the full SHA
    4121ff4 View commit details
    Browse the repository at this point in the history

Commits on May 3, 2024

  1. save work - initial impl of optimizer

    @edgchen1
    edgchen1 committed May 3, 2024
    Configuration menu
    Copy the full SHA
    96fa18b View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'origin/main' into edgchen1/matmul_nbits… 

    …_bias
    @edgchen1
    edgchen1 committed May 3, 2024
    Configuration menu
    Copy the full SHA
    11c3b92 View commit details
    Browse the repository at this point in the history

Commits on May 6, 2024

  1. Add test, refine impl.

    @edgchen1
    edgchen1 committed May 6, 2024
    Configuration menu
    Copy the full SHA
    baab944 View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'origin/main' into edgchen1/matmul_nbits… 

    …_bias
    @edgchen1
    edgchen1 committed May 6, 2024
    Configuration menu
    Copy the full SHA
    1568758 View commit details
    Browse the repository at this point in the history

  3. fix fusion, update test

    @edgchen1
    edgchen1 committed May 6, 2024
    Configuration menu
    Copy the full SHA
    6402ea5 View commit details
    Browse the repository at this point in the history

  4. Enable bias in SQNBitGemm benchmark.

    @edgchen1
    edgchen1 committed May 6, 2024
    Configuration menu
    Copy the full SHA
    6a1feb2 View commit details
    Browse the repository at this point in the history

Commits on May 7, 2024

  1. clean up

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    a6c42e4 View commit details
    Browse the repository at this point in the history

  2. move test functions around to avoid unused function warnings

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    7fa290b View commit details
    Browse the repository at this point in the history

  3. add /bigobj for graph_tarnsform_test.cc

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    b5057eb View commit details
    Browse the repository at this point in the history

  4. lint

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    823c9db View commit details
    Browse the repository at this point in the history

  5. Add Neural Speed build to post-merge-jobs.yml.

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    7555f92 View commit details
    Browse the repository at this point in the history

  6. move graph_utils::IsSupportedProvider() in header so it is available … 

    …in minimal extended build
    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    10ba1db View commit details
    Browse the repository at this point in the history

  7. put MatMulNBitsBiasFusion test in contrib ops ifdef

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    471429c View commit details
    Browse the repository at this point in the history

  8. update ContribOperators.md

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    7f07d93 View commit details
    Browse the repository at this point in the history

  9. call RunWithConfig instead of Run

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    97a15a1 View commit details
    Browse the repository at this point in the history

  10. update OperatorKernels.md

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    905cbd2 View commit details
    Browse the repository at this point in the history

  11. add matmul_nbits_fusion files to extended minimal build source

    @edgchen1
    edgchen1 committed May 7, 2024
    Configuration menu
    Copy the full SHA
    0abc2cd View commit details
    Browse the repository at this point in the history

Commits on May 9, 2024

  1. Merge remote-tracking branch 'origin/main' into edgchen1/matmul_nbits… 

    …_bias
    @edgchen1
    edgchen1 committed May 9, 2024
    Configuration menu
    Copy the full SHA
    1a4b5a0 View commit details
    Browse the repository at this point in the history

  2. Make DML MatMulNBits input check accomodate new inputs.

    @edgchen1
    edgchen1 committed May 9, 2024
    Configuration menu
    Copy the full SHA
    50cb266 View commit details
    Browse the repository at this point in the history

  3. expect input arg counts of 0 for missing optional inputs

    @edgchen1
    edgchen1 committed May 9, 2024
    Configuration menu
    Copy the full SHA
    131141f View commit details
    Browse the repository at this point in the history

  4. Merge branch 'edgchen1/matmul_nbits_bias' of github.com:microsoft/onn… 

    …xruntime into edgchen1/matmul_nbits_bias
    @edgchen1
    edgchen1 committed May 9, 2024
    Configuration menu
    Copy the full SHA
    2417b80 View commit details
    Browse the repository at this point in the history

Commits on May 10, 2024

  1. add runtime optimization test for matmulnbits and add fusion

    @edgchen1
    edgchen1 committed May 10, 2024
    Configuration menu
    Copy the full SHA
    1d9c2f5 View commit details
    Browse the repository at this point in the history

  2. add test onnx file

    @edgchen1
    edgchen1 committed May 10, 2024
    Configuration menu
    Copy the full SHA
    afacb57 View commit details
    Browse the repository at this point in the history

  3. lint

    @edgchen1
    edgchen1 committed May 10, 2024
    Configuration menu
    Copy the full SHA
    55851e3 View commit details
    Browse the repository at this point in the history

Commits on May 15, 2024

  1. Add !ORT_NEURAL_SPEED ifdefs around adding MatMulNBitsFusion transfor… 

    …mer.
    @edgchen1
    edgchen1 committed May 15, 2024
    Configuration menu
    Copy the full SHA
    badcd0f View commit details
    Browse the repository at this point in the history

  2. add ifdef around test

    @edgchen1
    edgchen1 committed May 15, 2024
    Configuration menu
    Copy the full SHA
    b90c8ca View commit details
    Browse the repository at this point in the history

  3. Merge branch 'main' into edgchen1/matmul_nbits_bias

    @edgchen1
    edgchen1 committed May 15, 2024
    Configuration menu
    Copy the full SHA
    92ce6eb View commit details
    Browse the repository at this point in the history