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.

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

Ad/remove api label #8

Open
wants to merge 357 commits into
base: develop
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
357 commits
Select commit Hold shift + click to select a range
d94c564
[WC] Fix ratio_defining_params (#2653)
l-bat May 14, 2024
551285d
Bump version of black to 24.4.2 (#2678)
AlexanderDokuchaev May 15, 2024
66b381c
Remove openvino-dev dependency (#2675)
kshpv May 15, 2024
707075a
[TF] Fix dependency conflict (#2677)
AlexanderDokuchaev May 15, 2024
69ebebe
Added scale_estimation flag to IR metadata. (#2680)
andreyanufr May 15, 2024
9cfc7b4
[quantization] Fix config compatible check (#2681)
UsingtcNower May 16, 2024
24f0ee7
[Torch][PTQ] Resnet18 example updated to showcase comrpession save/lo…
daniil-lyakhov May 20, 2024
aaa407d
Add more functions for experimental tensor (#2687)
AlexanderDokuchaev May 22, 2024
3d11e8a
Add log message for when no matches were found for AWQ algorithm (#2691)
nikita-savelyevv May 22, 2024
6a974ad
Bump torch to 2.3.0 (#2690)
AlexanderDokuchaev May 23, 2024
7b01c26
[PT] Increase threshold for e2e reference (#2693)
AlexanderDokuchaev May 23, 2024
1e4b763
[OpenVINO] Fix bug when Add node is not 0-index child (#2692)
kshpv May 23, 2024
ce29e75
Andreyan/awq extention (#2676)
andreyanufr May 23, 2024
d6c0850
[Torch] Do not wrap not tensor inputs (#2649)
daniil-lyakhov May 23, 2024
47187e5
[OV] Compress bf16 weights (#2686)
alexsu52 May 23, 2024
4d7fd56
Revert "Andreyan/awq extention (#2676)" (#2695)
ljaljushkin May 24, 2024
936533e
Remove compress_to_fp16=False from examples (#2682)
truhinnm May 24, 2024
dc6dd74
[OpenVINO] Add 2024.2 OV references (#2697)
kshpv May 24, 2024
1468a9b
Fix NNCF compatibility with torch.compile (#2665)
nikita-savelyevv May 24, 2024
b8b3e3b
Remove dependency from typing_extensions (#2696)
AlexanderDokuchaev May 24, 2024
2369078
Update BKC for PyTroch (#2700)
AlexanderDokuchaev May 27, 2024
eb377e6
Fix SQ algorithm ports usage (#2672)
KodiaqQ May 28, 2024
14f628d
Removal of installation extras in setup.py (#2694)
AlexanderDokuchaev May 28, 2024
3d3b32b
[OpenVINO] Remove duplicated reference graphs (#2699)
kshpv May 28, 2024
1f8e26d
[OpenVINO] Yolov8 example: Remove unnecessary extra input (#2701)
kshpv May 28, 2024
90a7f0d
Fix test examples (#2703)
AlexanderDokuchaev May 28, 2024
13e39ad
Remove OMZ model_tools dependency (#2702)
kshpv May 31, 2024
c0c61f1
[Documentation] QAT after PTQ docs update (#2664)
daniil-lyakhov Jun 3, 2024
1b5c9d8
[Docs] Fix url link to quantization docs (#2715)
daniil-lyakhov Jun 3, 2024
77a8867
[OV] GPTQ (#2684)
alexsu52 Jun 4, 2024
cd64227
Accuracy results for scale estimation algorithm in docs. (#2706)
andreyanufr Jun 4, 2024
99b5b85
Removal install package with extra from tests (#2717)
AlexanderDokuchaev Jun 4, 2024
131cb36
bandit happy (#2722)
alexsu52 Jun 6, 2024
99d6d95
Update metric for tinyllama_data_aware_gptq_backend_OV (#2725)
alexsu52 Jun 7, 2024
65c9515
Fix for import matplotlib (#2724)
alexsu52 Jun 7, 2024
da18bf3
[CI] Set actions references to specific commit hash, provide explicit…
AlexanderDokuchaev Jun 10, 2024
3fa9207
Update calibrate tool (#2721)
andrey-churkin Jun 10, 2024
14b24a6
[CI] fix GA workflows (#2726)
AlexanderDokuchaev Jun 10, 2024
1eec305
Fix torch operator unpatching (#2719)
nikita-savelyevv Jun 11, 2024
289fefc
[CI] Fail pipeline on improvement metric (#2728)
AlexanderDokuchaev Jun 12, 2024
9af800d
Tensor revision (#2689)
AlexanderDokuchaev Jun 12, 2024
f1dc76a
Update doc links (#2730)
l-bat Jun 12, 2024
17cc066
[CI] Add coverage for pt and tf jobs (#2729)
AlexanderDokuchaev Jun 12, 2024
d06b174
[PT] Remove replace_custom_modules function (#2732)
AlexanderDokuchaev Jun 12, 2024
b780041
Implement __rpow__ for Tensor (#2731)
AlexanderDokuchaev Jun 13, 2024
9200a22
Release notes v2.11 (#2710)
KodiaqQ Jun 13, 2024
85b3263
Represent symmetrically quantized weights in signed data type (#2434)
l-bat Jun 13, 2024
b1cc78e
GPTQ documentation (#2735)
alexsu52 Jun 13, 2024
14df477
Excluded comprasion operations from quantizable operations (#2737)
alexsu52 Jun 14, 2024
ae935c2
Added 24.3 references (#2738)
KodiaqQ Jun 15, 2024
0c22c2c
Andreyan/awq extention (#2708)
andreyanufr Jun 17, 2024
5384965
Ignored scope matching working with multiple NNCFGraphs (#2723)
kshpv Jun 18, 2024
5c9a417
[Tests] Remove raising error when graph is not existed (#2744)
kshpv Jun 18, 2024
45f4eee
Bump OV/NNCF version (#2740)
KodiaqQ Jun 18, 2024
25399bd
[CI] install missed requirements for torch-cuda action (#2748)
AlexanderDokuchaev Jun 18, 2024
58b0c64
Update main.py (#2742)
Lars-Codes Jun 19, 2024
af341c9
xfail for test_torch_compile in windows (#2751)
AlexanderDokuchaev Jun 19, 2024
f4bd077
Tests revision (#2749)
KodiaqQ Jun 20, 2024
6a1e216
Update WC metric values (#2754)
l-bat Jun 21, 2024
306f394
[OV] Parallel tests by scope (#2758)
AlexanderDokuchaev Jun 21, 2024
4ccb32e
Bump version of scikit-learn (#2756)
AlexanderDokuchaev Jun 24, 2024
2ef2578
[PTQ] MinMax make empty cache != initialized cache (#2746)
kshpv Jun 24, 2024
408c67d
Release Tensor (#2739)
AlexanderDokuchaev Jun 24, 2024
a4971f1
Fix do_dequantization func (#2757)
l-bat Jun 25, 2024
2871993
Fix default value for ignored scope in calibrate.py (#2759)
kshpv Jun 25, 2024
4924b7e
[PT] Support scaled_dot_product_attention (#2761)
AlexanderDokuchaev Jun 25, 2024
81e9e32
[CI] Add paths-ignore like in pre-commit (#2762)
AlexanderDokuchaev Jun 25, 2024
2ba43bd
E2M1 compression. (#2745)
andreyanufr Jun 28, 2024
beb9c04
Update Fast-/BC algorithms (#2747)
KodiaqQ Jun 28, 2024
a25e56d
Support TF 2.15 (#2609)
andrey-churkin Jul 1, 2024
082841c
[PT] Quantization of addmm function (#2713)
AlexanderDokuchaev Jul 1, 2024
e8ea252
Remove reference data for TF 2.8 from the tests (#2770)
andrey-churkin Jul 1, 2024
f5ad4ea
Revert "Update Fast-/BC algorithms (#2747)" (#2771)
KodiaqQ Jul 1, 2024
1f07622
Extending functions by lstsq, svd, eye (#2774)
ljaljushkin Jul 2, 2024
aeaaf19
Adjust NNCF to numpy 2.0 api (#2772)
AlexanderDokuchaev Jul 3, 2024
a644a1e
Update Fast-/BC with MatMul/Gemm support (#2776)
KodiaqQ Jul 3, 2024
fd0e33c
E8M0 scale for E2M1 weights. (#2767)
andreyanufr Jul 3, 2024
d256991
[OV] Clean test environment (#2779)
AlexanderDokuchaev Jul 4, 2024
2552f1b
1) Fixed bug with clamp range in scale estimation. (#2781)
andreyanufr Jul 5, 2024
28d99a0
Make NNCF common utils code pass mypy checks (#2780)
anzr299 Jul 5, 2024
0ff601c
Compile model only for CPU (#2786)
KodiaqQ Jul 5, 2024
f45bcad
Fix for llm examples with openvino-nightly (#2782)
ljaljushkin Jul 5, 2024
f5ce945
Update wc_reference_data.yaml (#2791)
andreyanufr Jul 6, 2024
9fcba3e
Ignore ReleaseNotes.md by link checker (#2793)
AlexanderDokuchaev Jul 8, 2024
26fd1fd
Add assignment operators and matmul (#2792)
AlexanderDokuchaev Jul 8, 2024
f02e5ef
Updated docs with MXFP4 (e2m1, e8m0) information. (#2797)
andreyanufr Jul 8, 2024
5b0438d
[OpenVINO] Fix PRelu type name (#2795)
kshpv Jul 9, 2024
42ae1f8
[TEST] Turn off dynamic quantization in the lm_weight_compression (#2…
alexsu52 Jul 10, 2024
e9ae8f5
Add a linalg.pinv wrapper to common tensor (#2798)
ljaljushkin Jul 10, 2024
73b044a
Remove unused code (#2789)
andrey-churkin Jul 10, 2024
35f8f1a
Reduce test_examples logs (#2800)
KodiaqQ Jul 10, 2024
40233c0
Fix Readme.md links (#2768)
KodiaqQ Jul 11, 2024
d033c6a
Fix test (#2806)
andrey-churkin Jul 12, 2024
6926cf1
[PTQ][OV] BF16 support (#2307)
KodiaqQ Jul 12, 2024
232b435
Lower weight compression memory footprint by sorting weights accordin…
nikita-savelyevv Jul 12, 2024
021c7e5
Support per-channel mode in AWQ, GPTQ and Scale Estimation algos (#2785)
l-bat Jul 12, 2024
1f49375
Introduced versioning of reference for conformance tests (#2805)
ljaljushkin Jul 12, 2024
bc959d1
[OpenVINO] Add a new HW pattern (#2804)
kshpv Jul 15, 2024
cae4012
added references for 2024.4 nightly (#2810)
ljaljushkin Jul 15, 2024
2ffaabb
Set upper limit for pydot version 3.0.0 (#2814)
AlexanderDokuchaev Jul 16, 2024
d113c2a
Fix pytest warnings (#2813)
AlexanderDokuchaev Jul 16, 2024
380d2bb
Use a graph to find the output port id for the activation quantizatio…
andrey-churkin Jul 18, 2024
2a83751
[OpenVINO] Add HW patterns (#2822)
kshpv Jul 18, 2024
43d3300
Revert changes (#2827)
andrey-churkin Jul 19, 2024
49e9820
Post-training Activation Pruning algorithm (#2683)
yujiepan-work Jul 19, 2024
e70befd
Increase approx abs (#2826)
AlexanderDokuchaev Jul 22, 2024
8744c20
[CI] Correction files for conformance tests (#2824)
AlexanderDokuchaev Jul 22, 2024
204bc9f
Reduce test_examples logs (#2828)
alexsu52 Jul 22, 2024
b328e4d
Fix shared_memory issue with cast (#2834)
KodiaqQ Jul 22, 2024
60000e4
[PTQ] Remove logs from autogenerated ignored node names (#2815)
kshpv Jul 23, 2024
5f5562d
Fix for model transformer command (#2823)
KodiaqQ Jul 23, 2024
f981d53
Bump NNCF version (#2811)
KodiaqQ Jul 23, 2024
0e1c83a
[Torch] Use metatypes instead of types in pattern matching (#2787)
daniil-lyakhov Jul 23, 2024
f1df8c9
Fix incorrect scales in SmoothQuant algo (#2830)
l-bat Jul 23, 2024
33846e4
Bump laber action to v5 (#2836)
AlexanderDokuchaev Jul 25, 2024
33bbf6e
Update Usage.md (#2842)
anzr299 Jul 25, 2024
d94b93b
[Experimental] TorchFX PTQ backend (#2764)
daniil-lyakhov Jul 26, 2024
d1e4ec5
Scale for covering full range of compressed weights. (#2790)
andreyanufr Jul 30, 2024
25eff06
[TorchFX] Graph/metatypes tests implementation (#2796)
anzr299 Jul 30, 2024
2a0a93e
[TorchFX]: Test Fast Bias Correction algorithm (#2812)
anzr299 Jul 30, 2024
f60b805
Skip inception_v3_imagenet_rb_sparsity_int8 test from Tensorflow back…
alexsu52 Jul 31, 2024
ef2aa5a
Add junit report to torch cpu/gpu makefile tasks (#2857)
ksilligan Jul 31, 2024
02c9b53
[Torch Fx] Statistics Aggregator test (#2856)
rk119 Jul 31, 2024
31f4e2b
Update OpenVINO version (#2851) (#2859)
KodiaqQ Jul 31, 2024
80c612e
Update ReleaseNotes.md (#2837) (#2850)
KodiaqQ Jul 31, 2024
655f296
[TorchFX] Test implementation of YoloV8 is removed (#2858)
daniil-lyakhov Jul 31, 2024
f5aa90f
[OpenVINO] Add Slice metatype (#2847)
kshpv Jul 31, 2024
e2183b9
[PT] bump torch 2.4.0 (#2852)
AlexanderDokuchaev Aug 2, 2024
587374f
[PT] Update reference for squeezenet1_1_imagenet_int8_per_tensor (#28…
AlexanderDokuchaev Aug 2, 2024
63fcb15
[Torch Fx] Add test for quantized model graph (#2860)
anzr299 Aug 5, 2024
ef49c75
[Solver] Check all output branches are quantized before merging quant…
daniil-lyakhov Aug 5, 2024
d8a37a7
Updated weight compression metrics regarding to OpenVINO 2024.3 (#2868)
alexsu52 Aug 5, 2024
697ba35
Optimization of the "calculate_signed_scale" function (#2855)
alexsu52 Aug 6, 2024
6bf0220
[Torch Fx] Remove Yolov8 and Ultralytics from Torch Fx Model Test (#2…
anzr299 Aug 6, 2024
fa072ab
Update commands.py (#2871)
anzr299 Aug 6, 2024
b108455
Updated metrics for llm example tests regarding to OpenVINO 2024.3 (#…
alexsu52 Aug 6, 2024
eb91af2
[TorchFX] Conformance test init (#2841)
daniil-lyakhov Aug 7, 2024
3695da7
Limit ONNX version in examples (#2876)
KodiaqQ Aug 9, 2024
01b748f
[GA] pull_request (#2881)
AlexanderDokuchaev Aug 9, 2024
8634037
Fixed unified FakeConverts duplication (#2877)
KodiaqQ Aug 9, 2024
27296b4
[TorchFX] Pre-hook insertion support (#2861)
daniil-lyakhov Aug 9, 2024
a7d575e
[TorchFX][FBC] Constant linear layers support (#2866)
daniil-lyakhov Aug 13, 2024
1e315f6
Fix pre-commit (#2884)
daniil-lyakhov Aug 13, 2024
d02c613
[GA] Fix labeler action (#2885)
AlexanderDokuchaev Aug 13, 2024
d5b9a0b
Increased tolerance for unstable tests (#2883)
ljaljushkin Aug 13, 2024
9a0b5d2
Memory Monitor tool (#2801)
nikita-savelyevv Aug 13, 2024
7744ebf
[TorchFX] SmoothQuant algorithm implementation (#2875)
daniil-lyakhov Aug 16, 2024
60fe68a
Add PyPi publishing file (#2825)
kblaszczak-intel Aug 19, 2024
bdf8d27
Add xfail for conformance test (#2890)
AlexanderDokuchaev Aug 19, 2024
b882992
[CI][PTQ] benchmark (#2819)
AlexanderDokuchaev Aug 19, 2024
3bda8f2
[GA] Remove unreached step (#2893)
AlexanderDokuchaev Aug 20, 2024
a8c5ea4
Replace the exception with a warning (#2901)
andrey-churkin Aug 22, 2024
e2b9898
Add early stopping for memory monitoring (#2897)
nikita-savelyevv Aug 22, 2024
6e05cf9
[GA] Add `Set api label` action (#2900)
AlexanderDokuchaev Aug 22, 2024
7c19980
Revert Slice as agnostic (#2905)
KodiaqQ Aug 22, 2024
3e61532
[GA] Fix set api label action (#2908)
AlexanderDokuchaev Aug 23, 2024
865798c
[GA] Fix path (#2909)
AlexanderDokuchaev Aug 23, 2024
c57df43
[TorchFX] Deptwise convolution support (#2896)
daniil-lyakhov Aug 23, 2024
1104f1b
[GA] fix script set api label (#2910)
AlexanderDokuchaev Aug 23, 2024
c500822
Add weighted progress tracking for weight compression (#2892)
nikita-savelyevv Aug 26, 2024
3a9304c
Fix track context manager (#2915)
nikita-savelyevv Aug 26, 2024
ec25a29
[TorchFX] Fix depthwise + transpose conv weights quantization (#2912)
daniil-lyakhov Aug 27, 2024
417c2a1
Weight compression via Lora Correction Algorithm (#2816)
ljaljushkin Aug 28, 2024
23ca5e8
Raise error threshold (#2921)
nikita-savelyevv Aug 28, 2024
b4e216d
Move test_templates to cross_fw directory (#2887)
AlexanderDokuchaev Aug 28, 2024
cf06bc1
Compatibility table updated (#2925)
alexsu52 Aug 28, 2024
cc7b3d2
Added statefull GPTNeox for PTQ conformance tests (#2923)
ljaljushkin Aug 28, 2024
caf851a
Remove Yolov7 deadlink from Readme.md (#2927)
anzr299 Aug 29, 2024
6be2fd5
Preserve the constant subgraph (#2902)
andrey-churkin Aug 29, 2024
2a52f86
Fix random.sample for python312 (#2914)
AlexanderDokuchaev Aug 29, 2024
0e90dac
[GA] Fix remove api label (#2911)
AlexanderDokuchaev Aug 29, 2024
74cc471
Update readme and release notes (#2930)
MaximProshin Aug 30, 2024
ff967e8
Make NNCF common tensor statistics code pass mypy checks (#2865)
rk119 Aug 30, 2024
2088962
Bump NNCF version (#2933)
KodiaqQ Aug 30, 2024
eb61347
Fix torch compile applied to model forward (#2932)
nikita-savelyevv Aug 30, 2024
aa37119
[TorchFX] Bias correction implementation (#2882)
daniil-lyakhov Aug 30, 2024
3454626
Added link to HF compressed models (#2939)
MaximProshin Sep 2, 2024
03f65a7
[Quantization] Add quantizer filter pass to Solver (#2903)
kshpv Sep 2, 2024
25ab806
[GA] Removal code coverage (#2886)
AlexanderDokuchaev Sep 3, 2024
3effa14
[ONNX] Extend se block ignored pattern (#2926)
kshpv Sep 3, 2024
03efd62
Support the scale estimation inside GPTQ (#2784)
alexsu52 Sep 4, 2024
ca41c70
[Docs] smooth_quant_alpha parameter (#2945)
alexsu52 Sep 5, 2024
d7bb1ca
Extend NNCF logging by the device information where quantization was …
andrey-churkin Sep 5, 2024
f41df7b
[TF] Update reference after PR 2903 (#2946)
kshpv Sep 5, 2024
b99ffde
[TorchFX][MinMax] Use `get_input_edge_by_port_id` instead of `get_pre…
anzr299 Sep 5, 2024
187dda2
Accuracy/Footprint trade-off with different int4 compression algorith…
ljaljushkin Sep 5, 2024
6f1b2dd
Improve weight compression time and memory for BF16 OV models (#2949)
nikita-savelyevv Sep 5, 2024
67d6386
Avoid squeeze for seq_len equal to one. (#2821)
andreyanufr Sep 9, 2024
a75b47a
Added metrics for 2024.5 nightly (#2954)
ljaljushkin Sep 9, 2024
20ecaed
[TESTS] add ignored scope for mobielenet_v3 torchvision (#2947)
kshpv Sep 9, 2024
387aa30
Fix segmentation fault (#2955)
andrey-churkin Sep 10, 2024
b9dc0b6
Fix file comparison on windows (#2960)
AlexanderDokuchaev Sep 11, 2024
c0e24ec
[GA] Ignore files with tests that are not run in precommit actions (#…
AlexanderDokuchaev Sep 11, 2024
4401d44
Added similarity for weight compression with 2024.4 (#2958)
ljaljushkin Sep 11, 2024
6a8c366
Type mismatch in onnx quantize_with_accuracy_control validation_fn (#…
andrey-churkin Sep 12, 2024
c25f699
Alignment behavior of get_ignored_node_names_from_ignored_scope (#2966)
AlexanderDokuchaev Sep 12, 2024
d3d5a28
Increase threshold for test_memory_values_isolated (#2967)
nikita-savelyevv Sep 12, 2024
ae25915
data option for test examples (#2950)
alexsu52 Sep 13, 2024
67ddcf8
Fix version of datasets for conformance test (#2968)
AlexanderDokuchaev Sep 16, 2024
c966784
Updated Usage.md for GPTQ and Scale Estimation (#2971)
alexsu52 Sep 16, 2024
2af8d5c
Remove unused weight_modifiable_node_names property (#2965)
andrey-churkin Sep 16, 2024
42101d5
[GA] python310 (#2969)
AlexanderDokuchaev Sep 17, 2024
a39e673
[TESTS] Revert ignored scope for conformance (#2975)
kshpv Sep 18, 2024
eec9102
Update ReleaseNotes.md (#2940) (#2974)
KodiaqQ Sep 19, 2024
631c3e4
Bump OV version (#2973) (#2977)
KodiaqQ Sep 19, 2024
393c2dd
Move tests.shared to tests.cross_fw.shared (#2961)
AlexanderDokuchaev Sep 19, 2024
d9b3f38
[Torch Fx] PTQ Quantizer config test (#2937)
rk119 Sep 23, 2024
05f37f5
NF4 per-channel support for AWQ and Scale Estimation (#2898)
ljaljushkin Sep 23, 2024
2efda7d
Add GroupNorm to the OV list of ignored ops for ModelType.TRANSFORMER…
nikita-savelyevv Sep 24, 2024
b840385
[Torch FX] PTQ MinMax parameters test (#2989)
rk119 Sep 25, 2024
f274cf5
[OV] Add adaptive poolings metatypes (#2988)
kshpv Sep 25, 2024
c93676d
[TorchFX] Weights Compression Support (#2891)
anzr299 Sep 26, 2024
da5217a
[FX/PT/ONNX][MinMax] Weights nodes/ constant matmuls collection fix (…
daniil-lyakhov Sep 27, 2024
ad11ee1
[TorchFX][Conformance test] Migration to `torch.export.export` (#2990)
daniil-lyakhov Sep 30, 2024
b94e1df
Fix metric for tinyllama_scale_estimation_per_channel (#2983)
l-bat Sep 30, 2024
2c8b70c
[TorchFX] Graph tests extended with edge shapes (#2918)
daniil-lyakhov Sep 30, 2024
ce1fb51
NNCF helpers (#2979)
KodiaqQ Oct 1, 2024
b52bf6c
Merge wc_reference_data_2024.4.yaml and wc_reference_data.yaml (#2997)
alexsu52 Oct 7, 2024
3b2b8c3
Synthetic data updates (#2999)
KodiaqQ Oct 7, 2024
174bd03
Support backup precision option for WC (#2978)
l-bat Oct 7, 2024
63669db
[TorchFX] PTQ MinMax algorithm test (#3000)
rk119 Oct 7, 2024
4dbe89b
Add deprecation message for nncf.torch.create_compressed_model (#2986)
AlexanderDokuchaev Oct 9, 2024
d8c13a2
[PTQ] Add option for merging quantizers strategy in advanced paramete…
kshpv Oct 14, 2024
989f60b
[PT] TorchFunctionMode (#2998)
AlexanderDokuchaev Oct 14, 2024
8ef38ec
[experimental][collectors] Redundant code is removed (#3006)
daniil-lyakhov Oct 14, 2024
e81a1a3
pyproject.toml (#2981)
AlexanderDokuchaev Oct 15, 2024
4bbcefb
[GA] test examples (#3001)
AlexanderDokuchaev Oct 15, 2024
30f579b
Data aware compression statistics enhancement (#3003)
nikita-savelyevv Oct 15, 2024
03f2422
PTMeanReducer is removed (#3016)
daniil-lyakhov Oct 15, 2024
cb0fe0d
[GA] fail status for example workflow (#3015)
AlexanderDokuchaev Oct 16, 2024
17f799e
[NNCF][BC] Add test case with depthwise/transpose convolutions to tem…
rk119 Oct 16, 2024
73eb8b3
Visualisation of weight compression results (#3009)
ljaljushkin Oct 16, 2024
5619afb
[TorchFX] Model transformer tests (#2920)
daniil-lyakhov Oct 16, 2024
e6a4752
Update aggregator.py (#2995)
zina-cs Oct 17, 2024
14158ce
Data-aware compression fixes (#3019)
nikita-savelyevv Oct 17, 2024
e9d2415
[BC] Fix collected inputs (#3021)
KodiaqQ Oct 17, 2024
dfc5d78
Move `activations_to_wc_statistics` to SE class (#3022)
nikita-savelyevv Oct 17, 2024
4e388d2
Drop supports python3.8 and add support numpy2 (#2919)
AlexanderDokuchaev Oct 18, 2024
7385c41
compress_weights telemetry update (#3011)
KodiaqQ Oct 18, 2024
7c94b23
[Torch FX] Post Quantize Weights Compression (#2984)
anzr299 Oct 21, 2024
d33ce2f
Remove warning from examples (#3023)
l-bat Oct 21, 2024
7263096
[TorchFX] Rensen18 example (#2913)
daniil-lyakhov Oct 22, 2024
efdceca
set pythonpath for pytest (#3026)
AlexanderDokuchaev Oct 22, 2024
d758013
name
AlexanderDokuchaev Oct 23, 2024
d393ef6
-
AlexanderDokuchaev Oct 23, 2024
df9d774
test
AlexanderDokuchaev Oct 23, 2024
fff1a1c
t
AlexanderDokuchaev Oct 23, 2024
151cca8
t
AlexanderDokuchaev Oct 23, 2024
edfc666
t
AlexanderDokuchaev Oct 23, 2024
5e41af1
t
AlexanderDokuchaev Oct 23, 2024
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
12 changes: 0 additions & 12 deletions .dockerignore

This file was deleted.

2 changes: 2 additions & 0 deletions .github/ISSUE_TEMPLATE/good_first_issue.yml
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,8 @@ body:
discussions, guides.
value: |
- [Contribution guide - start here!](https://github.com/openvinotoolkit/nncf/blob/develop/CONTRIBUTING.md)
- [Intel DevHub Discord channel](https://discord.gg/7pVRxUwdWG) - engage in discussions, ask questions and talk to OpenVINO developers
- [How to link your Pull Request to an issue](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#manually-linking-a-pull-request-to-an-issue-using-the-pull-request-sidebar)
validations:
required: true

Expand Down
92 changes: 62 additions & 30 deletions .github/action_configs/labeler.yml
Original file line number Diff line number Diff line change
@@ -1,49 +1,81 @@
# See help here: https://github.com/marketplace/actions/labeler

dependencies:
- 'requirements.txt'
- '**/setup.py'
- any:
- changed-files:
- any-glob-to-any-file:
- 'requirements.txt'
- '**/setup.py'

NNCF PT:
- 'examples/torch/**/*!(.md)'
- 'examples/post_training_quantization/torch/**/*!(.md)'
- 'nncf/torch/**/*!(.md)'
- 'tests/torch/**/*!(.md)'
- 'nncf/quantization/**/torch_backend.py'
- any:
- changed-files:
- any-glob-to-any-file:
- 'examples/torch/**/*!(.md)'
- 'examples/post_training_quantization/torch/**/*!(.md)'
- 'nncf/torch/**/*!(.md)'
- 'tests/torch/**/*!(.md)'
- 'nncf/quantization/**/torch_backend.py'

NNCF TF:
- 'examples/tensorflow/**/*!(.md)'
- 'examples/post_training_quantization/tensorflow/**/*!(.md)'
- 'nncf/tensorflow/**/*!(.md)'
- 'tests/tensorflow/**/*!(.md)'
- 'nncf/quantization/**/tf_backend.py'
- any:
- changed-files:
- any-glob-to-any-file:
- 'examples/tensorflow/**/*!(.md)'
- 'examples/post_training_quantization/tensorflow/**/*!(.md)'
- 'nncf/tensorflow/**/*!(.md)'
- 'tests/tensorflow/**/*!(.md)'
- 'nncf/quantization/**/tf_backend.py'

NNCF ONNX:
- 'examples/onnx/**/*!(.md)'
- 'examples/post_training_quantization/onnx/**/*!(.md)'
- 'nncf/onnx/**/*!(.md)'
- 'tests/onnx/**/*!(.md)'
- 'nncf/quantization/**/onnx_backend.py'
- any:
- changed-files:
- any-glob-to-any-file:
- 'examples/onnx/**/*!(.md)'
- 'examples/post_training_quantization/onnx/**/*!(.md)'
- 'nncf/onnx/**/*!(.md)'
- 'tests/onnx/**/*!(.md)'
- 'nncf/quantization/**/onnx_backend.py'

NNCF OpenVINO:
- 'examples/openvino/**/*!(.md)'
- 'examples/post_training_quantization/openvino/**/*!(.md)'
- 'nncf/openvino/**/*!(.md)'
- 'tests/openvino/**/*!(.md)'
- 'nncf/quantization/**/openvino_backend.py'
- any:
- changed-files:
- any-glob-to-any-file:
- 'examples/openvino/**/*!(.md)'
- 'examples/post_training_quantization/openvino/**/*!(.md)'
- 'nncf/openvino/**/*!(.md)'
- 'tests/openvino/**/*!(.md)'
- 'nncf/quantization/**/openvino_backend.py'

NNCF PTQ:
- 'nncf/quantization/**/*!(.md)'
- 'tests/post_training/**/*!(.md)'
- any:
- changed-files:
- any-glob-to-any-file:
- 'nncf/quantization/**/*!(.md)'
- 'tests/post_training/**/*!(.md)'

documentation:
- '**/*.md'
- 'docs/**/*'
- any:
- changed-files:
- any-glob-to-any-file:
- '**/*.md'
- 'docs/**/*'

experimental:
- 'nncf/experimental/**/*!(.md)'
- any:
- changed-files:
- any-glob-to-any-file:
- 'nncf/experimental/**/*!(.md)'

NNCF Common:
- 'examples/common/**/*!(.md)'
- 'nncf/common/**/*!(.md)'
- 'tests/common/**/*!(.md)'
- any:
- changed-files:
- any-glob-to-any-file:
- 'examples/common/**/*!(.md)'
- 'nncf/common/**/*!(.md)'
- 'tests/common/**/*!(.md)'

release target:
- any:
- base-branch:
- '^release_v*'
59 changes: 24 additions & 35 deletions .github/workflows/api_changes_check.yml
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
name: API changes check
permissions: read-all

on:
pull_request_target:
pull_request:
branches:
- develop

Expand All @@ -23,11 +25,11 @@ jobs:
issues: write
steps:
- name: Download built HTML doc as artifact from previous step
uses: alehechka/download-tartifact@v2
uses: alehechka/download-tartifact@1195216b256562056097b175df17b167557f8681 # v2
with:
name: html_doc_artifact
- name: Checkout latest doc_pages branch tip
uses: actions/checkout@v3
uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29 # v4.1.6
with:
ref: doc_pages
path: previous_doc_state
Expand All @@ -39,37 +41,24 @@ jobs:
echo ${CHANGED_FILES}
CHANGED_FILES=$(echo $CHANGED_FILES | tr '\n' ' ')
echo "changed_files=${CHANGED_FILES}" >> $GITHUB_OUTPUT
- uses: actions/github-script@v6
if: ${{ !(contains(steps.diff.outputs.changed_files, 'differ')) && contains(github.event.pull_request.labels.*.name, 'API') }}
with:
github-token: ${{ secrets.ADD_LABELS_WITH_REST_API }}
script: |
github.rest.issues.removeLabel({
issue_number: context.issue.number,
owner: context.repo.owner,
repo: context.repo.repo,
name: "API"
})
- uses: actions/github-script@v6

- name: Add label
if: ${{ contains(steps.diff.outputs.changed_files, 'differ') }}
run: |
echo '{"pr_number": "${{ github.event.pull_request.number }}", "action": "add"}' > api_status.json

- name: Remove label
if: ${{ !(contains(steps.diff.outputs.changed_files, 'differ')) && contains(github.event.pull_request.labels.*.name, 'API') }}
run: |
echo '{"pr_number": "${{ github.event.pull_request.number }}", "action": "remove"}' > api_status.json

- name: No change label
if: ${{ !(contains(steps.diff.outputs.changed_files, 'differ')) && !contains(github.event.pull_request.labels.*.name, 'API') }}
run: |
echo '{"pr_number": "${{ github.event.pull_request.number }}", "action": "none"}' > api_status.json

- name: Upload artifact
uses: actions/upload-artifact@v4
with:
github-token: ${{ secrets.ADD_LABELS_WITH_REST_API }}
script: |
github.rest.issues.addLabels({
issue_number: context.issue.number,
owner: context.repo.owner,
repo: context.repo.repo,
labels: ["API"]
})
- name: Add release label
uses: actions/github-script@v6
if: ${{ contains(github.event.pull_request.base.ref, 'release_v') }}
with:
github-token: ${{ secrets.ADD_LABELS_WITH_REST_API }}
script: |
github.rest.issues.addLabels({
issue_number: context.issue.number,
owner: context.repo.owner,
repo: context.repo.repo,
labels: ["release_target"]
})
name: api_status
path: api_status.json
24 changes: 24 additions & 0 deletions .github/workflows/assign_issue.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,24 @@
name: Take Issue
permissions: read-all

on:
issue_comment:
types:
- created
- edited

jobs:
take-issue:
name: Take issue
runs-on: ubuntu-latest
permissions:
issues: write
timeout-minutes: 10
steps:
- name: take an issue
uses: bdougie/take-action@1439165ac45a7461c2d89a59952cd7d941964b87 # v1.6.1
with:
message: Thank you for looking into this issue! Please let us know if you have any questions or require any help.
issueCurrentlyAssignedMessage: Thanks for being interested in this issue. It looks like this ticket is already assigned to a contributor. Please communicate with the assigned contributor to confirm the status of the issue.
trigger: .take
token: ${{ secrets.GITHUB_TOKEN }}
12 changes: 6 additions & 6 deletions .github/workflows/build_and_publish_doc.yml
Original file line number Diff line number Diff line change
@@ -1,4 +1,6 @@
name: Build complete docs and publish to GH Pages
permissions: read-all

on:
push:
branches:
Expand All @@ -11,8 +13,6 @@ concurrency:
group: ci-${{ github.head_ref }}-${{ github.workflow }}
cancel-in-progress: true

permissions:
contents: write
jobs:
call-build-html-doc:
uses: ./.github/workflows/build_html_doc.yml
Expand All @@ -25,18 +25,18 @@ jobs:
runs-on: ubuntu-20.04
steps:
- name: Checkout main repo # the github-pages-deploy-action seems to require this step
uses: actions/checkout@v3
uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29 # v4.1.6
- name: Download HTML doc build artifact
uses: alehechka/download-tartifact@v2
uses: alehechka/download-tartifact@1195216b256562056097b175df17b167557f8681 # v2
with:
name: html_doc_artifact
- name: Download schema doc build artifact
uses: alehechka/download-tartifact@v2
uses: alehechka/download-tartifact@1195216b256562056097b175df17b167557f8681 # v2
with:
name: schema_doc_artifact
path: html_build/html
- name: Publish built docs on Github Pages branch ${{ env.GH_PAGES_BRANCH }}
uses: JamesIves/github-pages-deploy-action@v4
uses: JamesIves/github-pages-deploy-action@5c6e9e9f3672ce8fd37b9856193d2a537941e66c # v4.6.1
with:
folder: html_build/html
token: ${{ secrets.PUSH_TO_GH_PAGES_BRANCH }}
Expand Down
10 changes: 8 additions & 2 deletions .github/workflows/build_html_doc.yml
Original file line number Diff line number Diff line change
@@ -1,4 +1,6 @@
name: HTML documentation build
permissions: read-all

on:
workflow_call:
inputs:
Expand All @@ -10,9 +12,13 @@ jobs:
runs-on: ubuntu-20.04
steps:
- name: Checkout
uses: actions/checkout@v3
uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29 # v4.1.6
with:
ref: ${{ inputs.ref }}
- uses: actions/setup-python@0a5c61591373683505ea898e09a3ea4f39ef2b9c # v5.0.0
with:
python-version: 3.10.14
cache: pip
- name: Install NNCF and doc requirements
run: |
pip install .
Expand All @@ -21,7 +27,7 @@ jobs:
run: |
sphinx-build -M html docs/api/source html_build
- name: Upload built HTMLs as job artifact
uses: alehechka/upload-tartifact@v2
uses: alehechka/upload-tartifact@a055d3a102b9ed9cfff1263bc713295047d0197e # v2
with:
name: html_doc_artifact
path: html_build/html
14 changes: 11 additions & 3 deletions .github/workflows/build_schema_page.yml
Original file line number Diff line number Diff line change
@@ -1,26 +1,34 @@
name: Config schema HTML build
permissions: read-all

on:
workflow_call:

jobs:
build-config-schema-html:
runs-on: ubuntu-20.04
steps:
- name: Checkout
uses: actions/checkout@v3
uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29 # v4.1.6

- uses: actions/setup-python@0a5c61591373683505ea898e09a3ea4f39ef2b9c # v5.0.0
with:
python-version: 3.10.14
cache: pip

- name: Install and Build
run: |
pip install json-schema-for-humans
pip install .
python -c 'import jstyleson; from nncf.config import NNCFConfig; jstyleson.dump(NNCFConfig.schema(), open("./schema.json", "w"), indent=2)'
mkdir -p schema
# In the current approach the line below will replace the schema/index.html generated by a
# In the current approach the line below will replace the schema/index.html generated by a
# phony schema/index.rst. We only need the phony schema/index.rst at the sphinx-build stage so that
# the sphinx can generate valid links in the TOC for the JSON schema across all pages.
generate-schema-doc --deprecated-from-description schema.json schema/index.html

- name: Upload result as artifact
uses: alehechka/upload-tartifact@v2
uses: alehechka/upload-tartifact@a055d3a102b9ed9cfff1263bc713295047d0197e # v2
with:
name: schema_doc_artifact
path: schema
Loading
Loading