microsoft
/
nn-Meter
зеркало из
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

Граф коммитов

  • 0f00a22a2d
         [Fix bug] Modify profiling arguments (#86) Jiahang Xu 2022-10-25 16:41:35 +0800
  • 29b7f9a073
         Update customized workspace create doc (#84) Jiahang Xu 2022-10-25 16:40:35 +0800
  • d33c66422f
         [Fix bug] torch fusion rule test (#85) Jiahang Xu 2022-10-25 16:07:52 +0800
  • 4baeb61dac add note of aarch64 jiahangxu 2022-10-25 16:06:42 +0800
  • 204be7d4ab sort model configs by #param jiahangxu 2022-10-25 15:37:48 +0800
  • aee661ef25 refactor get_flops_params and latency_metrics jiahangxu 2022-10-25 15:19:38 +0800
  • d62656914a Merge branch 'dev/profiling-features' of github.com:microsoft/nn-Meter into dev/profiling-features jiahangxu 2022-10-25 14:23:57 +0800
  • f2ff909b92 add comment code jiahangxu 2022-10-25 14:22:33 +0800
  • c68b71b283 update common_layers jiahangxu 2022-10-25 14:16:41 +0800
  • fe3e3eda4a update build lut code jiahangxu 2022-10-25 14:15:16 +0800
  • 8259610f16 unnessary commit jiahangxu 2022-10-25 02:11:23 -0400
  • 059f14681b modify profiling args jiahangxu 2022-10-25 01:01:46 -0400
  • d97f32934e update customized workspace create doc jiahangxu 2022-09-26 05:15:05 -0400
  • 842f170356 Merge branch 'dev/quantize-predictor' of https://github.com/microsoft/nn-Meter into dev/quantize-predictor jiahangxu 2022-09-19 15:44:05 +0800
  • 296e8e3576 update from m20 server (including swish) jiahangxu 2022-09-19 15:43:51 +0800
  • a700e6d293
         Fix shape in PyTorch Fusion Rule Test (#81) Jiahang Xu 2022-09-13 18:38:57 +0800
  • b1fff23876 add logger in fusion rule test jiahangxu 2022-09-13 08:53:52 +0000
  • 3bf5a68b78 fix bugs of torch fc op jiahangxu 2022-09-13 08:50:11 +0000
  • 39460e1c46 fix bug in torch fusion rules jiahangxu 2022-09-13 03:21:49 +0000
  • 7ac63167f4
         Handle model profiling timeout (#76) Jiahang Xu 2022-09-09 19:05:08 +0800
  • 066088d20b use time_threshold param to replace default 300s jiahangxu 2022-09-09 10:40:42 +0000
  • 681f4ae86e
         change the backend profile argument (#78) Jiahang Xu 2022-09-09 18:08:53 +0800
  • 28af7704b5
         Update onnx version (#80) Jiahang Xu 2022-09-09 18:08:33 +0800
  • da8f02e3ae Merge branch 'dev/onnx-version' into dev/dataset-generator dev/dataset-generator jiahangxu 2022-09-08 14:44:04 +0000
  • de18d21cfe Merge branch 'dev/onnx-version' into dev/handle-timeout jiahangxu 2022-09-08 14:29:44 +0000
  • e22975ed23 Merge branch 'dev/onnx-version' into dev/backend-inputshape jiahangxu 2022-09-08 14:26:07 +0000
  • 758a989278 Merge branch 'dev/onnx-version' of github.com:microsoft/nn-Meter into dev/onnx-version jiahangxu 2022-09-08 09:47:30 +0000
  • d14aa35cf1 update onnx onnxsim version jiahangxu 2022-09-08 09:45:22 +0000
  • 34cd260ca1 debug pipeline error 3 jiahangxu 2022-09-08 09:42:11 +0000
  • 03e0ab6d10 debug pipeline error 2 jiahangxu 2022-09-08 08:51:23 +0000
  • 66e2f93855
         Update image link (#79) Zhiqiang Wang 2022-09-08 16:47:27 +0800
  • c440f730ea debug pipeline error jiahangxu 2022-09-08 08:40:52 +0000
  • 80ade29986 fix bug jiahangxu 2022-09-08 08:32:36 +0000
  • 3b7b89043d update onnx version jiahangxu 2022-09-08 08:27:57 +0000
  • 102274198a
         Update image link Zhiqiang Wang 2022-08-31 18:43:18 +0800
  • 87a5ae18f7 change the backend profile argument jiahangxu 2022-08-31 04:44:38 -0400
  • ab2df772c6 update implementation jiahangxu 2022-07-25 18:52:23 +0800
  • 768578a46d fix typo and refine error info jiahangxu 2022-07-25 05:07:59 -0400
  • 36dda6a666 support time thershold of 300s for one model jiahangxu 2022-07-25 05:00:20 -0400
  • f54dfbb24d Merge branch 'main' into dev/dataset-generator jiahangxu 2022-07-25 04:49:30 -0400
  • 6899bf9e86 Merge branch 'main' into dev/quantize-predictor jiahangxu 2022-07-25 04:46:15 -0400
  • 20fbd6606c Merge branch 'main' into dev/transformer jiahangxu 2022-07-25 04:45:11 -0400
  • c5102b34ac update mobilenetv3 config in examples jiahangxu 2022-07-25 04:41:23 -0400
  • b2391b165b add post processing code in trans lut jiahangxu 2022-07-25 04:26:25 -0400
  • 9ff4312caf support csv input in builder jiahangxu 2022-07-19 13:42:44 +0800
  • 1f06458d8a temp save nasbench201 jiahangxu 2022-07-13 02:36:23 -0400
  • f72962e01c update operator opname jiahangxu 2022-07-12 11:45:42 -0400
  • 2a7a35d5f7 update and test 12 model families jiahangxu 2022-07-12 11:20:54 -0400
  • a2c2bcbc9a add examples dev/block-latency jiahangxu 2022-07-06 11:57:54 -0400
  • e03075d9ab update case of not applicable jiahangxu 2022-07-06 11:54:47 -0400
  • 5fd682d19b Merge branch 'main' into dev/block-latency jiahangxu 2022-07-06 10:51:21 -0400
  • 2f91065853 update block-wise latency output jiahangxu 2022-07-06 07:43:20 -0400
  • 646aa8fb25 update trans lut for spacev1 and attnnas jiahangxu 2022-07-05 02:38:25 -0400
  • efbc887567 update op level test tf/torch jiahangxu 2022-07-03 20:52:08 -0400
  • 4c10c00271
         Update version info after v2.0 (#75) Jiahang Xu 2022-07-01 11:07:10 +0800
  • b8d433f736
         Bump tensorflow from 2.7.0 to 2.7.2 in /docs/requirements (#74) dependabot[bot] 2022-07-01 11:01:47 +0800
  • cdb80dc72b Merge branch 'main' into dev/refactor jiahangxu 2022-06-28 09:14:43 -0400
  • 8e8a5e27e1
         Bump tensorflow from 2.7.0 to 2.7.2 in /docs/requirements dependabot[bot] 2022-06-28 13:09:20 +0000
  • bdef60ebc0
         sync v2.0 changes to main branch (#73) Jiahang Xu 2022-06-28 21:08:32 +0800
  • d916856f92 update nn-meter version jiahangxu 2022-06-28 09:06:52 -0400
  • cd9bcb3ce2 complete v2.0 test v2.0 v2.0 jiahangxu 2022-06-27 05:17:20 -0400
  • e66d7ef3b9 fix bugs in v2.0 jiahangxu 2022-06-27 05:00:25 -0400
  • 0b5e0ce0bc fix little bugs in v2.0 jiahangxu 2022-06-26 23:34:56 -0400
  • 113ad7e2a5
         Refactor nn_modules from building tools (#72) Jiahang Xu 2022-06-27 09:31:20 +0800
  • f04a64b39d update swish predictor jiahangxu 2022-06-24 10:45:17 -0400
  • ff64c3029d
         Bump numpy from 1.21.0 to 1.22.0 in /docs/requirements (#71) dependabot[bot] 2022-06-23 23:22:34 +0800
  • 14f8238bf4 update and test tf blocks jiahangxu 2022-06-23 11:20:18 -0400
  • 3a7995debd update and test torch blocks jiahangxu 2022-06-23 07:23:08 -0400
  • 2456f5a5b5 Merge branch 'dev/quantize-predictor' of github.com:microsoft/nn-Meter into dev/quantize-predictor jiahangxu 2022-06-23 12:42:57 +0800
  • 5b053e0074
         Bump numpy from 1.21.0 to 1.22.0 in /docs/requirements dependabot[bot] 2022-06-22 04:59:37 +0000
  • 9074549108 update pixel4 lut jiahangxu 2022-06-19 14:05:25 -0400
  • a0af614861 update onnx lut (add fusedMB conv) jiahangxu 2022-06-18 10:03:42 -0400
  • fb25d7c1d8 update nasbench201 jiahangxu 2022-06-17 22:02:23 -0400
  • 6d38e5d844 Merge branch 'main' into dev/dataset-generator jiahangxu 2022-06-17 13:56:31 -0400
  • e025aa1324 reset quantize operator.py and add swish in tf jiahangxu 2022-06-17 13:53:12 -0400
  • d8e446a03c Merge branch 'dev/transformer' of github.com:microsoft/nn-Meter into dev/transformer jiahangxu 2022-06-16 11:03:20 -0400
  • 95735c24b9 remove unused comments jiahangxu 2022-06-16 11:02:29 -0400
  • 6e16af4ca5 Merge branch 'dev/transformer' of github.com:microsoft/nn-Meter into dev/transformer jiahangxu 2022-06-14 13:40:33 -0400
  • 7d1737ad07 Merge branch 'dev/transformer' of github.com:microsoft/nn-Meter into dev/transformer jiahangxu 2022-06-14 13:40:18 -0400
  • 0d45eaec57 Merge branch 'main' into dev/transformer jiahangxu 2022-06-14 13:39:29 -0400
  • 1be23e09b9 Merge branch 'main' into dev/quantize-predictor jiahangxu 2022-06-14 13:37:22 -0400
  • 80fe22fbe5 fix bugs and change input of tranformer predictor jiahangxu 2022-06-14 07:46:07 -0400
  • 25a53d89f4
         Refine docs and readme.md, refactor nn_generator (#70) Jiahang Xu 2022-06-14 15:52:32 +0800
  • 0526693402 support lut for transformer jiahangxu 2022-06-14 03:01:23 -0400
  • e6efa36cbe update onnx_lut.json jiahangxu 2022-06-13 22:46:14 -0400
  • a16dd8a88d feature: support lut predictor jiahangxu 2022-06-13 10:53:28 -0400
  • 2b5031bdd8 temp save tensorflow blocks jiahangxu 2022-06-13 07:32:05 -0400
  • 0b0c384497 rename test_nn_generator jiahangxu 2022-06-13 07:06:51 -0400
  • 0310512b22 change modules to nn_modules 2 jiahangxu 2022-06-13 06:55:54 -0400
  • 1011e603bf change modules to nn_modules jiahangxu 2022-06-13 04:53:03 -0400
  • 8e4acd9829 Merge branch 'dev/setup-refactor' of github.com:microsoft/nn-Meter into dev/setup-refactor jiahangxu 2022-06-10 08:19:22 -0400
  • 77ae7d701c refine logging message jiahangxu 2022-06-10 08:00:29 -0400
  • 56289dcb7e fix bug of keepdims jiahangxu 2022-06-10 04:07:19 -0400
  • 2c6a2bc8e5 Merge branch 'dev/quantize-predictor' of github.com:microsoft/nn-Meter into dev/quantize-predictor jiahangxu 2022-06-09 14:41:17 +0800
  • 6310365fec update se op in both tf and torch jiahangxu 2022-06-08 07:03:05 -0400
  • 383a4bf644 fix conflict and fix import typo jiahangxu 2022-06-08 06:59:17 -0400
  • cd86b0b59e change the implementation of se in both tf and torch jiahangxu 2022-06-08 06:38:04 -0400
  • 3e8022d998 set keepdim=True in tf se op jiahangxu 2022-06-08 06:14:59 -0400
  • 17a5488cb9 add test data in build_predictor.py jiahangxu 2022-06-07 12:40:17 -0400
  • 995886559f Merge branch 'dev/quantize-predictor' of github.com:microsoft/nn-Meter into dev/quantize-predictor jiahangxu 2022-06-08 00:30:52 +0800