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

Fix RemoveIdentityOps not correctly handling ops following fork-nodes #87

Merged
merged 114 commits into from
Sep 12, 2024
Merged

Fix RemoveIdentityOps not correctly handling ops following fork-nodes #87

merged 114 commits into from
Sep 12, 2024

Commits on Aug 30, 2023

  1. [Range] use monotonic range calc for reshapes

    @maltanar
    maltanar committed Aug 30, 2023
    Configuration menu
    Copy the full SHA
    6250172 View commit details
    Browse the repository at this point in the history

Commits on Oct 22, 2023

  1. QuantLSTM ONNX representation

    @shashwat1198
    shashwat1198 committed Oct 22, 2023
    Configuration menu
    Copy the full SHA
    64d95ac View commit details
    Browse the repository at this point in the history

  2. Clean QuantLSTM

    @shashwat1198
    shashwat1198 committed Oct 22, 2023
    Configuration menu
    Copy the full SHA
    e2e0737 View commit details
    Browse the repository at this point in the history

Commits on Oct 24, 2023

  1. [Transform] introduce ExtractQuantScaleZeroPt and simple test

    @maltanar
    maltanar committed Oct 24, 2023
    Configuration menu
    Copy the full SHA
    95279e4 View commit details
    Browse the repository at this point in the history

  2. [Test] add more testcases for ExtractQuantScaleZeroPt

    @maltanar
    maltanar committed Oct 24, 2023
    Configuration menu
    Copy the full SHA
    9e0a49a View commit details
    Browse the repository at this point in the history

Commits on Nov 6, 2023

  1. Update 0_how_to_work_with_onnx.ipynb 

    Change in the import and built-in functions.
    @Harsh9650
    Harsh9650 authored Nov 6, 2023
    Configuration menu
    Copy the full SHA
    51b6717 View commit details
    Browse the repository at this point in the history

Commits on Nov 14, 2023

  1. Add option to find_upstream to keep nodes visited even if not found

    @iksnagreb
    iksnagreb committed Nov 14, 2023
    Configuration menu
    Copy the full SHA
    94321b7 View commit details
    Browse the repository at this point in the history

  2. Add cleanup transformation sorting inputs of commutative operations

    @iksnagreb
    iksnagreb committed Nov 14, 2023
    Configuration menu
    Copy the full SHA
    7719a3e View commit details
    Browse the repository at this point in the history

  3. Address some linting issues

    @iksnagreb
    iksnagreb committed Nov 14, 2023
    Configuration menu
    Copy the full SHA
    c0f5b46 View commit details
    Browse the repository at this point in the history

Commits on Nov 15, 2023

  1. Fix RemoveIdentityOps not correctly handling ops following fork-nodes

    @iksnagreb
    iksnagreb committed Nov 15, 2023
    Configuration menu
    Copy the full SHA
    8902694 View commit details
    Browse the repository at this point in the history

  2. Change error message to address some linting issue

    @iksnagreb
    iksnagreb committed Nov 15, 2023
    Configuration menu
    Copy the full SHA
    c7b3590 View commit details
    Browse the repository at this point in the history

Commits on Nov 20, 2023

  1. Introduce 3D layout annotations defaulting to NWC 

    This is required for supporting scaled dot-product attention, which has
3 dimensional tensors with batch/head, sequence and embedding dimension.
    @iksnagreb
    iksnagreb committed Nov 20, 2023
    Configuration menu
    Copy the full SHA
    ebd9193 View commit details
    Browse the repository at this point in the history

Commits on Nov 21, 2023

  1. update in the matmul mac calculation

    Harish committed Nov 21, 2023
    Configuration menu
    Copy the full SHA
    6f8efa2 View commit details
    Browse the repository at this point in the history

  2. [Hotfix] round scaled-up internal result for QuantAvgPool2d

    @maltanar
    maltanar authored Nov 21, 2023
    Configuration menu
    Copy the full SHA
    cdc2747 View commit details
    Browse the repository at this point in the history

  3. Merge pull request fastmachinelearning#89 from iksnagreb/feature/3d_d… 

    …ata_layouts

Introduce 3D layout annotations defaulting to NWC
    @maltanar
    maltanar authored Nov 21, 2023
    Configuration menu
    Copy the full SHA
    62223f3 View commit details
    Browse the repository at this point in the history

Commits on Jan 12, 2024

  1. Merge branch 'main' into feature/tensor_stats

    @maltanar
    maltanar committed Jan 12, 2024
    Configuration menu
    Copy the full SHA
    6298df5 View commit details
    Browse the repository at this point in the history

  2. Merge main into feature/tensor_stats + deprecate propagate_range

    @maltanar
    maltanar committed Jan 12, 2024
    Configuration menu
    Copy the full SHA
    e8de415 View commit details
    Browse the repository at this point in the history

Commits on Jan 14, 2024

  1. Transformation pass to introduce quantnodes

    Harish committed Jan 14, 2024
    Configuration menu
    Copy the full SHA
    538a935 View commit details
    Browse the repository at this point in the history

Commits on Jan 30, 2024

  1. [Range] support multi-output ops, add Split

    @maltanar
    maltanar committed Jan 30, 2024
    Configuration menu
    Copy the full SHA
    c752dfb View commit details
    Browse the repository at this point in the history

  2. [Extract] support extracting ConvTranspose bias

    @maltanar
    maltanar committed Jan 30, 2024
    Configuration menu
    Copy the full SHA
    251030b View commit details
    Browse the repository at this point in the history

  3. [Range] support dense ConvTranspose for analysis

    @maltanar
    maltanar committed Jan 30, 2024
    Configuration menu
    Copy the full SHA
    f5f17f3 View commit details
    Browse the repository at this point in the history

  4. [Range] support Python notation for input ranges ([ch0_min, ch1_min,.… 

    ….],[ch0_max, ch1_max,..])
    @maltanar
    maltanar committed Jan 30, 2024
    Configuration menu
    Copy the full SHA
    f209c9d View commit details
    Browse the repository at this point in the history

  5. [Cleanup] allow overriding full input tensor size

    @maltanar
    maltanar committed Jan 30, 2024
    Configuration menu
    Copy the full SHA
    96906cb View commit details
    Browse the repository at this point in the history

Commits on Jan 31, 2024

  1. [Extract] use derived output dtype for conv bias extraction

    @maltanar
    maltanar committed Jan 31, 2024
    Configuration menu
    Copy the full SHA
    11850c1 View commit details
    Browse the repository at this point in the history

  2. [Range] also work with np.float16

    @maltanar
    maltanar committed Jan 31, 2024
    Configuration menu
    Copy the full SHA
    184bfd7 View commit details
    Browse the repository at this point in the history

  3. [Cleanup] support both str and tuple format for input shape override

    @maltanar
    maltanar committed Jan 31, 2024
    Configuration menu
    Copy the full SHA
    2d6322b View commit details
    Browse the repository at this point in the history

Commits on Feb 5, 2024

  1. use newer tf2onnx

    @maltanar
    maltanar authored Feb 5, 2024
    Configuration menu
    Copy the full SHA
    1de24c5 View commit details
    Browse the repository at this point in the history

  2. [Test] add validated/ to ONNX model zoo URLs

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    fd3b319 View commit details
    Browse the repository at this point in the history

  3. Merge pull request fastmachinelearning#96 from fastmachinelearning/fi… 

    …x/tf2onnx-1.16.1

Try newer tf2onnx version to get CI back in green
    @maltanar
    maltanar authored Feb 5, 2024
    Configuration menu
    Copy the full SHA
    813128f View commit details
    Browse the repository at this point in the history

  4. Merge branch 'main' into feature/tensor_stats

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    3a33770 View commit details
    Browse the repository at this point in the history

  5. Merge pull request fastmachinelearning#97 from fastmachinelearning/fe… 

    …ature/tensor_stats

Range analysis improvements and better input shape override
    @maltanar
    maltanar authored Feb 5, 2024
    Configuration menu
    Copy the full SHA
    3fd9386 View commit details
    Browse the repository at this point in the history

  6. Merge branch 'import_edits' of https://github.com/Harsh9650/qonnx int… 

    …o Harsh9650-import_edits
    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    7d5c3a5 View commit details
    Browse the repository at this point in the history

  7. fix typo in notebooks README

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    3b0f11e View commit details
    Browse the repository at this point in the history

  8. [Notebook] more fixes to 0_how_to_work_with_onnx.ipynb

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    baff921 View commit details
    Browse the repository at this point in the history

  9. Merge pull request fastmachinelearning#81 from Harsh9650/import_edits 

    Update 0_how_to_work_with_onnx.ipynb
    @maltanar
    maltanar authored Feb 5, 2024
    Configuration menu
    Copy the full SHA
    e3cba58 View commit details
    Browse the repository at this point in the history

  10. Merge branch 'main' into transformation_introducing_quantnode

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    038f21e View commit details
    Browse the repository at this point in the history

  11. [Test] override_batchsize -> override_inpsize for cleanup

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    2feab84 View commit details
    Browse the repository at this point in the history

  12. [GraphQnt] some cleanup and renaming

    @maltanar
    maltanar committed Feb 5, 2024
    Configuration menu
    Copy the full SHA
    3e132fe View commit details
    Browse the repository at this point in the history

Commits on Feb 6, 2024

  1. [DataType] add preliminary support for FLOAT16

    @maltanar
    maltanar authored Feb 6, 2024
    Configuration menu
    Copy the full SHA
    99841c1 View commit details
    Browse the repository at this point in the history

Commits on Feb 8, 2024

  1. Revised version for QuantizeGraph

    Harish committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    42df7a0 View commit details
    Browse the repository at this point in the history

Commits on Feb 13, 2024

  1. Merge pull request fastmachinelearning#90 from fastmachinelearning/ma… 

    …tmul_mac_update

update in the matmul mac calculation
    @maltanar
    maltanar authored Feb 13, 2024
    Configuration menu
    Copy the full SHA
    39442cb View commit details
    Browse the repository at this point in the history

Commits on Feb 15, 2024

  1. inference cost breakdown

    Harish committed Feb 15, 2024
    Configuration menu
    Copy the full SHA
    7608e7c View commit details
    Browse the repository at this point in the history

  2. Update inference_cost.py

    @Harsh9650
    Harsh9650 authored Feb 15, 2024
    Configuration menu
    Copy the full SHA
    d120742 View commit details
    Browse the repository at this point in the history

  3. revised version of quantize_graph

    Harish committed Feb 15, 2024
    Configuration menu
    Copy the full SHA
    04619a3 View commit details
    Browse the repository at this point in the history

Commits on Feb 21, 2024

  1. Update inference_cost.py

    @Harsh9650
    Harsh9650 authored Feb 21, 2024
    Configuration menu
    Copy the full SHA
    be9a9f8 View commit details
    Browse the repository at this point in the history

Commits on Feb 23, 2024

  1. Merge pull request fastmachinelearning#80 from fastmachinelearning/fe… 

    …ature/extract_quant_scale_zeropt

Extract non-trivial Quant scale and zeropoint
    @maltanar
    maltanar authored Feb 23, 2024
    Configuration menu
    Copy the full SHA
    4743ae2 View commit details
    Browse the repository at this point in the history

  2. Merge branch 'feature/find_upstream_keep_visited' of https://github.c… 

    …om/iksnagreb/qonnx into iksnagreb-feature/find_upstream_keep_visited
    @maltanar
    maltanar committed Feb 23, 2024
    Configuration menu
    Copy the full SHA
    1815bb4 View commit details
    Browse the repository at this point in the history

  3. fix linting problems

    @maltanar
    maltanar committed Feb 23, 2024
    Configuration menu
    Copy the full SHA
    fad667f View commit details
    Browse the repository at this point in the history

  4. Merge branch 'main' into iksnagreb-feature/find_upstream_keep_visited

    @maltanar
    maltanar committed Feb 23, 2024
    Configuration menu
    Copy the full SHA
    b3186cb View commit details
    Browse the repository at this point in the history

  5. Merge pull request fastmachinelearning#83 from iksnagreb/feature/find… 

    …_upstream_keep_visited

Add option to find_upstream to keep nodes visited even if not found
    @maltanar
    maltanar authored Feb 23, 2024
    Configuration menu
    Copy the full SHA
    fe4aa37 View commit details
    Browse the repository at this point in the history

  6. Merge branch 'feature/sort_commutative_inputs' of https://github.com/… 

    …iksnagreb/qonnx into iksnagreb-feature/sort_commutative_inputs
    @maltanar
    maltanar committed Feb 23, 2024
    Configuration menu
    Copy the full SHA
    8bba76b View commit details
    Browse the repository at this point in the history

  7. [Wrapper] explicitly return None for name/index finder functions

    @maltanar
    maltanar committed Feb 23, 2024
    Configuration menu
    Copy the full SHA
    5e2d0b8 View commit details
    Browse the repository at this point in the history

  8. Merge pull request fastmachinelearning#94 from fastmachinelearning/tr… 

    …ansformation_introducing_quantnode

Transformation pass to introduce quant nodes
    @maltanar
    maltanar authored Feb 23, 2024
    Configuration menu
    Copy the full SHA
    fd61cfe View commit details
    Browse the repository at this point in the history

Commits on Feb 27, 2024

  1. Feat (transformation): adding ResizeConvolutionToDeconvolution transf… 

    …ormation
    @i-colbert
    i-colbert committed Feb 27, 2024
    Configuration menu
    Copy the full SHA
    565d978 View commit details
    Browse the repository at this point in the history

  2. Feat (transformation): restructure onnx data for upsampling tests

    @i-colbert
    i-colbert committed Feb 27, 2024
    Configuration menu
    Copy the full SHA
    cfc8dda View commit details
    Browse the repository at this point in the history

  3. Feat (tests): adding resize to deconv unit tests

    @i-colbert
    i-colbert committed Feb 27, 2024
    Configuration menu
    Copy the full SHA
    057c263 View commit details
    Browse the repository at this point in the history

  4. Fix (transformation): fixing warnings to exit on unsupported condition

    @i-colbert
    i-colbert committed Feb 27, 2024
    Configuration menu
    Copy the full SHA
    e199a46 View commit details
    Browse the repository at this point in the history

  5. Pre-commit fixes

    @i-colbert
    i-colbert committed Feb 27, 2024
    Configuration menu
    Copy the full SHA
    19c73fe View commit details
    Browse the repository at this point in the history

Commits on Mar 1, 2024

  1. package installations added

    @shashwat1198
    shashwat1198 committed Mar 1, 2024
    Configuration menu
    Copy the full SHA
    59a7ca0 View commit details
    Browse the repository at this point in the history

Commits on Mar 7, 2024

  1. Fix (tests): removing generated onnx graph

    @i-colbert
    i-colbert committed Mar 7, 2024
    Configuration menu
    Copy the full SHA
    30b838b View commit details
    Browse the repository at this point in the history

  2. Fix (util): centralizing auto pad utility

    @i-colbert
    i-colbert committed Mar 7, 2024
    Configuration menu
    Copy the full SHA
    8d66e89 View commit details
    Browse the repository at this point in the history

  3. Fix: adding check and warning for dimension mismatch

    @i-colbert
    i-colbert committed Mar 7, 2024
    Configuration menu
    Copy the full SHA
    bf1840d View commit details
    Browse the repository at this point in the history

  4. Fix: handling case where bias is generated from another node

    @i-colbert
    i-colbert committed Mar 7, 2024
    Configuration menu
    Copy the full SHA
    fd834aa View commit details
    Browse the repository at this point in the history

  5. Pre-commit fixes

    @i-colbert
    i-colbert committed Mar 7, 2024
    Configuration menu
    Copy the full SHA
    29e6843 View commit details
    Browse the repository at this point in the history

Commits on Mar 10, 2024

  1. Fix: handling scalar vs. array scaling factor

    @i-colbert
    i-colbert committed Mar 10, 2024
    Configuration menu
    Copy the full SHA
    13906da View commit details
    Browse the repository at this point in the history

Commits on Mar 11, 2024

  1. Fix: adding quant test cases and fixing bugs

    @i-colbert
    i-colbert committed Mar 11, 2024
    Configuration menu
    Copy the full SHA
    b7eebaa View commit details
    Browse the repository at this point in the history

Commits on Mar 12, 2024

  1. Merge pull request fastmachinelearning#102 from i-colbert/icolbert/nn… 

    …-resize-to-deconv

Feat: Adding ResizeConvolutionToDeconvolution transformation
    @maltanar
    maltanar authored Mar 12, 2024
    Configuration menu
    Copy the full SHA
    c5bd87f View commit details
    Browse the repository at this point in the history

Commits on Apr 25, 2024

  1. Add unit test for SortCommutativeInputsInitializerLast transformation

    @iksnagreb
    iksnagreb committed Apr 25, 2024
    Configuration menu
    Copy the full SHA
    7ebbeac View commit details
    Browse the repository at this point in the history

  2. Address some linting issues

    @iksnagreb
    iksnagreb committed Apr 25, 2024
    Configuration menu
    Copy the full SHA
    38df9fb View commit details
    Browse the repository at this point in the history

Commits on May 11, 2024

  1. Remove some commented debug code

    @timkpaine
    timkpaine committed May 11, 2024
    Configuration menu
    Copy the full SHA
    57d0d9d View commit details
    Browse the repository at this point in the history

Commits on May 21, 2024

  1. fix linting

    @maltanar
    maltanar committed May 21, 2024
    Configuration menu
    Copy the full SHA
    1dfda07 View commit details
    Browse the repository at this point in the history

  2. [InfCost] per-node norm mac/param counts, always floats for json

    @maltanar
    maltanar committed May 21, 2024
    Configuration menu
    Copy the full SHA
    2cc6d52 View commit details
    Browse the repository at this point in the history

  3. update README

    @maltanar
    maltanar committed May 21, 2024
    Configuration menu
    Copy the full SHA
    4dd2000 View commit details
    Browse the repository at this point in the history

  4. Merge branch 'main' into inference_cost_breakdown

    @maltanar
    maltanar committed May 21, 2024
    Configuration menu
    Copy the full SHA
    0ca12ce View commit details
    Browse the repository at this point in the history

  5. [Test] fix changes return style for inference cost

    @maltanar
    maltanar committed May 21, 2024
    Configuration menu
    Copy the full SHA
    a4e7e35 View commit details
    Browse the repository at this point in the history

  6. Merge pull request fastmachinelearning#101 from fastmachinelearning/i… 

    …nference_cost_breakdown

inference cost breakdown
    @maltanar
    maltanar authored May 21, 2024
    Configuration menu
    Copy the full SHA
    db969e6 View commit details
    Browse the repository at this point in the history

Commits on Jun 17, 2024

  1. fix notebook 0

    @makoeppel
    makoeppel committed Jun 17, 2024
    Configuration menu
    Copy the full SHA
    f9db218 View commit details
    Browse the repository at this point in the history

  2. refactor LowerConvsToMatMul class, increase rtol in test_conv_lowerin… 

    …g_convmnist()
    @makoeppel
    makoeppel committed Jun 17, 2024
    Configuration menu
    Copy the full SHA
    1b2774c View commit details
    Browse the repository at this point in the history

Commits on Jul 19, 2024

  1. avoiding mp.Pool in case of using only 1 worker for easier pdb debugging

    @lstasytis
    lstasytis committed Jul 19, 2024
    Configuration menu
    Copy the full SHA
    17bd6d0 View commit details
    Browse the repository at this point in the history

Commits on Jul 30, 2024

  1. strings attribute support added to CustomOp

    mdaniowi committed Jul 30, 2024
    Configuration menu
    Copy the full SHA
    5414416 View commit details
    Browse the repository at this point in the history

Commits on Aug 1, 2024

  1. strings attr test added to test_attr.py

    mdaniowi committed Aug 1, 2024
    Configuration menu
    Copy the full SHA
    ba5c41f View commit details
    Browse the repository at this point in the history

Commits on Aug 9, 2024

  1. [README] updates to dev docs: pip upgrade, linting

    @maltanar
    maltanar committed Aug 9, 2024
    Configuration menu
    Copy the full SHA
    b0a6088 View commit details
    Browse the repository at this point in the history

  2. Merge pull request fastmachinelearning#130 from fastmachinelearning/f… 

    …eature/readme_dev_updates

Updates to dev docs: pip upgrade, linting
    @maltanar
    maltanar authored Aug 9, 2024
    Configuration menu
    Copy the full SHA
    326a525 View commit details
    Browse the repository at this point in the history

  3. ran pre-commit

    @lstasytis
    lstasytis committed Aug 9, 2024
    Configuration menu
    Copy the full SHA
    3835a37 View commit details
    Browse the repository at this point in the history

Commits on Aug 12, 2024

  1. pre-commit applied

    mdaniowi authored and mdaniowi committed Aug 12, 2024
    Configuration menu
    Copy the full SHA
    654bf15 View commit details
    Browse the repository at this point in the history

  2. Merge pull request fastmachinelearning#128 from mdanilow/feature/stri… 

    …ngs_attr

Feature/strings attr
    @maltanar
    maltanar authored Aug 12, 2024
    Configuration menu
    Copy the full SHA
    c4c16f7 View commit details
    Browse the repository at this point in the history

Commits on Aug 13, 2024

  1. Merge pull request fastmachinelearning#127 from lstasytis/fix/avoid-m… 

    …p-pool-with-one-worker

Avoiding mp.Pool in case of using only 1 worker for easier pdb debugging
    @maltanar
    maltanar authored Aug 13, 2024
    Configuration menu
    Copy the full SHA
    2281a77 View commit details
    Browse the repository at this point in the history

  2. Merge pull request fastmachinelearning#119 from makoeppel/fix_qonnx_i… 

    …mport_in_notebook_0

fix notebook 0
    @maltanar
    maltanar authored Aug 13, 2024
    Configuration menu
    Copy the full SHA
    a8ca195 View commit details
    Browse the repository at this point in the history

  3. Merge pull request fastmachinelearning#79 from shashwat1198/main 

    QuantLSTM (ONNX) QCDQ representation
    @maltanar
    maltanar authored Aug 13, 2024
    Configuration menu
    Copy the full SHA
    698c4bf View commit details
    Browse the repository at this point in the history

Commits on Aug 16, 2024

  1. Merge branch 'improve/test_conv_lowering_convmnist' of https://github… 

    ….com/makoeppel/qonnx into makoeppel-improve/test_conv_lowering_convmnist
    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    415634b View commit details
    Browse the repository at this point in the history

  2. run pre-commit on all files

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    75f8f8c View commit details
    Browse the repository at this point in the history

  3. Merge pull request fastmachinelearning#120 from makoeppel/improve/tes… 

    …t_conv_lowering_convmnist

Improve/test conv lowering convmnist
    @maltanar
    maltanar authored Aug 16, 2024
    Configuration menu
    Copy the full SHA
    84ad7ae View commit details
    Browse the repository at this point in the history

  4. [LowerConv] skip convs with non-initialized weights

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    8f66615 View commit details
    Browse the repository at this point in the history

  5. [Test] add (failing) quant weight conv testcase for lowering

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    a92093c View commit details
    Browse the repository at this point in the history

  6. [LowerConv] support lowering Conv with Quant node on weights

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    5e5bb55 View commit details
    Browse the repository at this point in the history

  7. [Test] extend quant weight conv testcase, now passing

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    c54f142 View commit details
    Browse the repository at this point in the history

  8. [LowerConv] support reshaping quant conv weight scales

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    8d1ee1d View commit details
    Browse the repository at this point in the history

  9. [Test] add MNv1 for quant conv lowering test

    @maltanar
    maltanar committed Aug 16, 2024
    Configuration menu
    Copy the full SHA
    55c2f6f View commit details
    Browse the repository at this point in the history

Commits on Aug 18, 2024

  1. [Test] use np.isclose instead of equals for test condition

    @maltanar
    maltanar committed Aug 18, 2024
    Configuration menu
    Copy the full SHA
    a3451c5 View commit details
    Browse the repository at this point in the history

Commits on Aug 22, 2024

  1. [Util] break out test input generation function & allow seed setting

    @maltanar
    maltanar committed Aug 22, 2024
    Configuration menu
    Copy the full SHA
    100bfde View commit details
    Browse the repository at this point in the history

  2. [Lower] fix quant scale conversion, adjust seed 

    random input generated with seed=42 was causing a major difference
in Conv_13_out0 for no apparent reason (probably float / numerical
related)
    @maltanar
    maltanar committed Aug 22, 2024
    Configuration menu
    Copy the full SHA
    032681c View commit details
    Browse the repository at this point in the history

  3. Merge pull request fastmachinelearning#132 from fastmachinelearning/f… 

    …eature/convlower_qnt

Preserve weight quantizer while lowering convolutions
    @maltanar
    maltanar authored Aug 22, 2024
    Configuration menu
    Copy the full SHA
    bdf9405 View commit details
    Browse the repository at this point in the history

Commits on Aug 23, 2024

  1. Merge pull request fastmachinelearning#116 from timkpaine/tkp/clean 

    Remove some commented debug code
    @jmduarte
    jmduarte authored Aug 23, 2024
    Configuration menu
    Copy the full SHA
    c9ad9e5 View commit details
    Browse the repository at this point in the history

Commits on Sep 9, 2024

  1. [Util] add accumulator-aware quantized (A2Q) CIFAR-10 models

    @maltanar
    maltanar committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    3c870d6 View commit details
    Browse the repository at this point in the history

  2. [Test] correctly handle multi-channel input ranges in change_batchsize

    @maltanar
    maltanar committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    ee7464f View commit details
    Browse the repository at this point in the history

  3. [Util] handle per-channel ranges in get_random_input

    @maltanar
    maltanar committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    8694a6d View commit details
    Browse the repository at this point in the history

  4. Merge pull request fastmachinelearning#142 from fastmachinelearning/f… 

    …eature/test_a2q_nets

Add easy fetch of accumulator-aware quantized (A2Q) CIFAR-10 models for testing
    @maltanar
    maltanar authored Sep 9, 2024
    Configuration menu
    Copy the full SHA
    1f8938a View commit details
    Browse the repository at this point in the history

Commits on Sep 12, 2024

  1. Merge branch 'feature/sort_commutative_inputs' of https://github.com/… 

    …iksnagreb/qonnx into iksnagreb-feature/sort_commutative_inputs
    @maltanar
    maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    cc1b1f0 View commit details
    Browse the repository at this point in the history

  2. Merge pull request fastmachinelearning#85 from iksnagreb/feature/sort… 

    …_commutative_inputs

Add cleanup transformation sorting inputs of commutative operations
    @maltanar
    maltanar authored Sep 12, 2024
    Configuration menu
    Copy the full SHA
    e02f701 View commit details
    Browse the repository at this point in the history

  3. Merge branch 'fix/remove_identity_ops' of https://github.com/iksnagre… 

    …b/qonnx into iksnagreb-fix/remove_identity_ops
    @maltanar
    maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    93acc3a View commit details
    Browse the repository at this point in the history

  4. Added Identity node to the removal list

    @maltanar
    jvreca authored and maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    8bad7e7 View commit details
    Browse the repository at this point in the history

  5. [Test] add Identity op case to test_remove_identity_ops

    @maltanar
    maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    71ee780 View commit details
    Browse the repository at this point in the history

  6. [ModelWrapper] add top-level checks for fork/join checks

    @maltanar
    maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    0a4d5c5 View commit details
    Browse the repository at this point in the history

  7. [Test] add fork cases to RemoveIdentityOps test

    @maltanar
    maltanar committed Sep 12, 2024
    Configuration menu
    Copy the full SHA
    2d09341 View commit details
    Browse the repository at this point in the history