mindstudio-probe 1.1.0__py3-none-any.whl → 1.1.1__py3-none-any.whl

mindstudio_probe-1.1.0.dist-info/RECORD

@@ -1,287 +0,0 @@
-msprobe/README.md,sha256=PVuqD2Uyf_FMKdVLlu8_vOtRPw6SK_uNdRA7uIdBbEo,5296
-msprobe/__init__.py,sha256=q2S8b3VTSn1wTMrkf8Gm79OOq9EerMIK6Y26gFdu_0s,688
-msprobe/config.json,sha256=B91ept9Wzif4_N4ZoMg3U0X7lf2c8sX8u_1b0zY-PR8,1089
-msprobe/msprobe.py,sha256=lJR_VokSvNOSMy9uQEFPJ1DrkYkVu4WAkDLeg41HSsk,5069
-msprobe/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/core/common_config.py,sha256=r3wRMWfVyGpuqGYKHitTcAnkU-osSCa0-yog4slimHU,4245
-msprobe/core/advisor/advisor.py,sha256=6rj5tJGBU-pD6fT1FQMpU8Bw5KlEWaqS-vAWOlvVzAI,5641
-msprobe/core/advisor/advisor_const.py,sha256=thEEVYKPimv9WGxxaC4vBXQ-Coq6yBlZ394T_b49Ibw,2624
-msprobe/core/advisor/advisor_result.py,sha256=yM78n6NXLP1bjTsGeSddwb1bWQwtg1ZqjnPHmTzVxHE,2359
-msprobe/core/common/const.py,sha256=XAlMqH5s_8ptqIGh-l3wV68iOWWBuR0DIfFHoqgzdyA,11834
-msprobe/core/common/exceptions.py,sha256=CR78r3u_xUjSeF7oFpQAYZcI0LxjbFcuqrhpTVHAH9E,2883
-msprobe/core/common/file_utils.py,sha256=witWNngvKztXjZFKCAD1pNMcX9JDlRfhi7Iw3Ew5Ids,19597
-msprobe/core/common/inplace_op_checker.py,sha256=t3qxnnvLmxpvV6TYq1jvyy5lslJpICFmhe5YwqtWDTU,1240
-msprobe/core/common/inplace_ops.yaml,sha256=CqE0wY6VFNvjRdmHHAR5l0BTSm_HHw6Ui0gL068pkOA,3094
-msprobe/core/common/log.py,sha256=QFFess5TDso5HLjlMpuLtjFqcEeEHlfiDXKMoTn_5R4,2582
-msprobe/core/common/utils.py,sha256=Nhe4VkypbKWB6xzQsoMuqoj96173qpUxnfrAAaqdIhc,14702
-msprobe/core/compare/acc_compare.py,sha256=7ccTPRCl8eRoaGbv3YTiTk8EGpZmZmElLXkK8JRnZRI,24318
-msprobe/core/compare/check.py,sha256=7Xkt9XUKU2_QSPqCbH3ogIrXSOySPp-lsrU-pdqKSqM,7495
-msprobe/core/compare/compare_cli.py,sha256=JoB0EnAid2hm17S5xBYi69p54OBDcE9XxU3reGgu4TQ,3344
-msprobe/core/compare/highlight.py,sha256=xlXUEh7hLjtlLOzCUO1123j-0yeDdaEcPkErKe-UvKI,11727
-msprobe/core/compare/multiprocessing_compute.py,sha256=XBmlQ5udZIuZFcVQwGQGxqHefx1roZ8SsyWzJWljY34,6892
-msprobe/core/compare/npy_compare.py,sha256=qepwUlt80PTy_8kOSxm62-1mjx0HQiRUUO_jTmx2p_w,12623
-msprobe/core/compare/utils.py,sha256=DrS1gVeqkXTrsuGsY9R1oEWMyqwm9NxVEjIRsxAhufw,20873
-msprobe/core/data_dump/data_collector.py,sha256=RHv2uahILgZbr030QiaPJTt04rYtM7A5PFafUz97buA,6307
-msprobe/core/data_dump/json_writer.py,sha256=2VyTiqtkECKP6FskdM0Gk_Z_F3OfU3o7VlGa3hKCk_A,4523
-msprobe/core/data_dump/scope.py,sha256=vI9rbuqPJBHgNIK6s03rcvEh6mdl5prlApPcEg65qkM,6603
-msprobe/core/data_dump/data_processor/base.py,sha256=2q3oePnN-kA0_w4j6gOsnW8ooxacHLENZS5sG_WcEDo,12843
-msprobe/core/data_dump/data_processor/factory.py,sha256=3v1AInU7DarnVYQUPu4qZngigxvhYLggh22EWASEfJo,3924
-msprobe/core/data_dump/data_processor/mindspore_processor.py,sha256=HquCKRMKCFFP2jDoE1Yb0b6fBThEvEZtA_dwV3r5p3o,8312
-msprobe/core/data_dump/data_processor/pytorch_processor.py,sha256=PiNERrLwzMmV_uAZmEmp3kP9UFL54OTJ2Jo-vVmH2-4,16629
-msprobe/core/grad_probe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/core/grad_probe/constant.py,sha256=-X2I3SCCEIXMakPwYPu1I7V4AdwFGRVvzR9oSOn83F0,1721
-msprobe/core/grad_probe/grad_compare.py,sha256=f16wuKKC3kvft31COdafa3H34yict1dKEI50oGdjfwc,7561
-msprobe/core/grad_probe/utils.py,sha256=UNQ29c87Ss6aPq4UjX6O7WR5Brsx6Kqu4b1frBky5qw,2506
-msprobe/docs/01.installation.md,sha256=ZDinC1749CtFL2fHIcMrhHeDHCdzgcZQqS4muP2TBTE,3723
-msprobe/docs/02.config_introduction.md,sha256=kJJcnki2R8S3V_V15-bFplH3lqgMIVuUObE6FS5IJr4,21104
-msprobe/docs/03.config_examples.md,sha256=xIp97s5J46jVGJcp58bPvZfFBe5SsBOA2fTCO2aNqWg,3881
-msprobe/docs/04.acl_config_examples.md,sha256=ltVE0XTbiThAde3u9KPovup5ZPF2qLjS69cqF2jTtro,2369
-msprobe/docs/05.data_dump_PyTorch.md,sha256=XogA3EbC0bLSg8PSkizM3LVeLyiS-5IJVyVOfYmME9Y,13436
-msprobe/docs/06.data_dump_MindSpore.md,sha256=ntODIjWjdqo5HMEP0I4DGsDNr_BgyBnW7auAipV9Frs,13784
-msprobe/docs/07.accuracy_checker_PyTorch.md,sha256=jk6hFPvmOWeroJA0N7sy8V27Y--VEyPQw-rssYPNMXw,28220
-msprobe/docs/08.accuracy_checker_online_PyTorch.md,sha256=i4kX9hDja4NUr9hglb0RkCjqXjVmWaXZk4SoEQCJn5w,11828
-msprobe/docs/09.accuracy_checker_MindSpore.md,sha256=I1RE8DEkRwjKbPjRY1rgrryMU5kxbJI9EPk5jw6d44U,5177
-msprobe/docs/10.accuracy_compare_PyTorch.md,sha256=h7p_wI4qAObMotzfctXlvlFLNCVC0ssaprxI1bdtmOY,20549
-msprobe/docs/11.accuracy_compare_MindSpore.md,sha256=Obr4suf01leNxAHzV052mW5xoWw0nFcKLmrhUyrM_qE,17415
-msprobe/docs/12.overflow_check_PyTorch.md,sha256=SkuTYAAzynawgFUoBjX4npNRDodZ04yaorqqFtwk_Is,3638
-msprobe/docs/13.overflow_check_MindSpore.md,sha256=efo_nfa5ywyO2v8gVL9YpQukj2KpcO8zHZDWCfOwJq8,2471
-msprobe/docs/14.data_parse_PyTorch.md,sha256=ERT5L3igrm1HrPsNbFMFbEtA5jvJ9HukWPP2CCuROdY,18749
-msprobe/docs/15.free_benchmarking_PyTorch.md,sha256=jRWFWmd1ZbEKQFg0rEMBZkFgtphiEtbup_MpUvO203A,9772
-msprobe/docs/16.free_benchmarking_MindSpore.md,sha256=nu5XxkEL_fd3mcHZUbDnkxfNuv8WQm0A8Q6mQlGRtsU,7964
-msprobe/docs/17.grad_probe.md,sha256=l7JT4Hcz7tTsckJNUArq3horDAiyUBrkf72SPa5g6QI,9417
-msprobe/docs/18.online_dispatch.md,sha256=KuQtWygBMNpc6B-nX2-z25ZBJUGXSXIlwwkcA_sM63A,4126
-msprobe/docs/FAQ.md,sha256=nz_PSl1fmaXcoONuhSfBWsWU1_KiAJ1ks1Dp3EIDjL0,9416
-msprobe/docs/S02.report_free_benchmarking_validation_performance_baseline.md,sha256=4NkHEa8VN3ebeUgrw9T-7rlL-qL_xB63w81O5m29D6s,10799
-msprobe/docs/img/BLOOM-7B_1.png,sha256=Ul9XNHCX52PYufGaZ5mPfwTNJxwUO46KOKp7JxvDeAk,3534
-msprobe/docs/img/BLOOM-7B_2.png,sha256=HKefltWpBM2UkVHFy9rB238g6W5w-8CpsueW6Zf52HM,5251
-msprobe/docs/img/BLOOM-7B_3.png,sha256=gtL8dXIoqoJwk2Nd-KKChVM47IBdEY6KVYVRejnwKt0,5095
-msprobe/docs/img/BLOOM-7B_4.png,sha256=1yuQ2Dh7ak-VZt_kLprXVHtvfUxaxfsgrKdYzudHR94,4727
-msprobe/docs/img/GPT-3_1.png,sha256=EVm59Bvd3cX01RpZBqDyMVECjsiySTNy5RR9YyJoCTw,7898
-msprobe/docs/img/GPT-3_2.png,sha256=4wNkqagB28IJi808pnvUB72E03b2OeWdAASq5ggyl8A,6716
-msprobe/docs/img/GPT-3_3.png,sha256=6-fbdYtgcYU1xe9yOg6tAnKQ8SfZ6-x6rPVNSRxGP8I,8732
-msprobe/docs/img/GPT-3_4.png,sha256=QYpb7ZJaOAM_wJ5qdQQon0TKu7apVRL7ijovl18Uvzg,6609
-msprobe/docs/img/GPT-3_5.png,sha256=8eiCyqLmiaH6xwqpyTe3XTcesbGQUw8gApDI7ONBBkA,4676
-msprobe/docs/img/GPT-3_6.png,sha256=4IZvCeYHgK7gB87dgf6HgTaQ5uBK0s2sZXLblsDIVtQ,7658
-msprobe/docs/img/GPT-3_7.png,sha256=dil9jwfsYxtFupUu_aTCnKjuuEk3X_UMnqWJS0NzWCA,8139
-msprobe/docs/img/GPT-3_8.png,sha256=RNXIQgrCObpYtoKTkdTxr66AkndY_k2mGbksNgIqccQ,6084
-msprobe/docs/img/YOLOV5S_1.png,sha256=kCI017bFT7X3RrBPLMESTrnHZcRJYF-Ko_oJpjcsFWw,7019
-msprobe/docs/img/YOLOV5S_2.png,sha256=NH4vy35cGDl3qjlcUhDTWNpGEKcuboQ_YILxWQpvx40,5796
-msprobe/docs/img/accuracy_checking_details.png,sha256=dAedq5tbuzM3uEKPZzoanFSkzc0ep3HSW0r_3iTWMvE,39062
-msprobe/docs/img/accuracy_checking_result.png,sha256=6dVDcp3hl-zFIOgLigbRrMdgeBDiMfKWxupL8Hugvxo,6022
-msprobe/docs/img/api_precision_compare_details.png,sha256=QkN8uldQ6j1Bvd4WxtucZXUMr--AoBUjSCJ01hTkpFw,25476
-msprobe/docs/img/api_precision_compare_result.png,sha256=Kp9bIK1rlw3zDyX4Oqw1CVPohTtJBig0RYK2iIhum9g,4538
-msprobe/docs/img/auto_analyze_log.png,sha256=3Jd9bl0BBrYxs5fFlcKYqeIjFbPlYegVmw4GGUe6s1o,65144
-msprobe/docs/img/compare_result_pkl.png,sha256=pMCd38zNjQ1LzcA_86DPlNA_jAfKJ0QYcSBDmzVA0Jw,36133
-msprobe/docs/img/compare_result_pkl_md5.png.png,sha256=ys9key_66mHsKST53SmGdzhYyA4dJtB2PVrGcqQkld0,20464
-msprobe/docs/img/cpu_info.png,sha256=8n9pjTDB8_Q8_i1mo4CMNqb2JF96UGTBCPWLgPo_XBU,33797
-msprobe/docs/img/free_benchmark.png,sha256=TjEImRJHHyt0-0R1fFWdhx1y4Lx8PM3wYgrF2XInkS0,72372
-msprobe/docs/img/free_benchmark_framework.png,sha256=hfYfF7Au7sgHaQeduovUdfL-nhWguIGeZZDaUHO3AfI,259617
-msprobe/docs/img/grad_probe_image-1.png,sha256=CfCd5eWGxfajKaEIWz2wSv5do1EeJRg9WEMLI7wUqVs,42344
-msprobe/docs/img/grad_probe_image-2.png,sha256=K7fC5F2ln5bYzqaXazF_adQhIdv4cjqPdo9qiQyc6uM,26563
-msprobe/docs/img/grad_probe_image-3.png,sha256=x-xvegKI7yQEl4lFzrKA9K0zRdT6m6kTdvs1epXaTjg,22581
-msprobe/docs/img/grad_probe_image-4.png,sha256=QX8iEZR4fiopqy2tFSPrWZo6AtYimYmpfy6rVVtXfk4,22779
-msprobe/docs/img/grad_probe_image.png,sha256=Uva-rNS275o4l1vHXCKAAAQmGOuzkKhjiFrCdm7iHVY,11977
-msprobe/docs/img/module_compare.png,sha256=8LjiAcZwLK-Me6YrkpWaAfOKFycStSapWdLsqZyS8oo,56655
-msprobe/docs/img/ms_dump.png,sha256=9tHk4gizGhkdycZRRiIRxBHJoPtCl9tsdMnl3PjVsk8,74578
-msprobe/docs/img/ms_layer.png,sha256=aqlWHCMoiNDksJE7m_ilpM0ifDLt8fxgh-MGNxpfNOU,277469
-msprobe/docs/img/pt_dump.png,sha256=c-YrWYiRERBj_x410-aq3-uMcxgII0ETVk-hT1KUr-4,70698
-msprobe/mindspore/__init__.py,sha256=H-r1h2V_yLn_gH4Jko7JPcJzceQImOnZ0KNhQ-okWRA,128
-msprobe/mindspore/cell_processor.py,sha256=-kQIecCEBkwjaEFG8OnS30JpYC72Z8yaqDuG1LC2XAU,1961
-msprobe/mindspore/ms_config.py,sha256=naAxopsLVz5iuXeO79d9nzBekY5PFZvvNCjSTKASNX8,5270
-msprobe/mindspore/runtime.py,sha256=8cdkkNgW37dVo_k5XgGUg_Juy31o86n6g2QIGuyHvAA,90
-msprobe/mindspore/service.py,sha256=0jzg17ABM_s23vrfHGQQCouM6ytsokG-OlOmEiFT8cE,13484
-msprobe/mindspore/task_handler_factory.py,sha256=x3cwKfuCl0ghZbHqL7a6PI1T54WnIh6Z7y00nvd2kCY,940
-msprobe/mindspore/api_accuracy_checker/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/api_accuracy_checker/api_accuracy_checker.py,sha256=kw5s32M_8VO-bOuchiKkhryfzGiUxjCheYElAYDsQws,13961
-msprobe/mindspore/api_accuracy_checker/api_info.py,sha256=4H_SSibhs6K3ZGtwnQxaxle4Rqg-qWIrV98I3vxvDwA,4194
-msprobe/mindspore/api_accuracy_checker/api_runner.py,sha256=GPdIWQKAYJbyn1ENyqdxz438QWz96nw4b6-dUAwUsCQ,7648
-msprobe/mindspore/api_accuracy_checker/base_compare_algorithm.py,sha256=SAZUf_J95S3KpvoyMXxcnfqh8Fsqfa8O6hwV97qWaJE,7782
-msprobe/mindspore/api_accuracy_checker/cmd_parser.py,sha256=GThZCtFsvBKtwFYvlcxv3R_jd_EmrhwSRVRDNBJZYxY,472
-msprobe/mindspore/api_accuracy_checker/compute_element.py,sha256=ElyQi5nHf3It3f_FiY3ikZnr6PPHei1SOetP7jEaOGY,11712
-msprobe/mindspore/api_accuracy_checker/main.py,sha256=5BTuAE84hsCq7KnX7u4ehzDsgDNuYN5PhnAjnWLBORs,370
-msprobe/mindspore/api_accuracy_checker/type_mapping.py,sha256=NZo4gu9QMapW0SF-wlOxzvmN-DJN_qge6AnCGZ6RX5s,2448
-msprobe/mindspore/api_accuracy_checker/utils.py,sha256=O5idt-0YLQixvFJLVv1NLEG51oNOKQ6DDFt2Ciox7Xo,3114
-msprobe/mindspore/common/const.py,sha256=a3Xl8ecNKGcAY93KlsjWZKwGqHW2lsNOQmYgGv2D_dA,3165
-msprobe/mindspore/common/log.py,sha256=DU645l6b0icOTQUxl0BztJFOQRVhbeZSTqbMaQAzMqE,1205
-msprobe/mindspore/common/utils.py,sha256=HUGtGzC0Wwtvf8nfa0WkjYSRB5ZeGJ5TEWof9gLfTJg,3222
-msprobe/mindspore/compare/distributed_compare.py,sha256=lhF4IXGWRbHnHQWdP38wvv8iNWbbDkWPyDoyWt-7HbQ,3288
-msprobe/mindspore/compare/layer_mapping.py,sha256=9PiC9o8-MuX00Q6N_Yzbi3eDbEc4ChmyoXUjwR6CyzE,5820
-msprobe/mindspore/compare/modify_mapping.py,sha256=LwQo5Hq2hXXAx3KJDOZjdFTdQH5GtRbdnPWOJTdQH-w,4767
-msprobe/mindspore/compare/ms_compare.py,sha256=_H28G42RKgFfoATEfu37-pgNeyGyAcLh8r7elKFpdUU,19044
-msprobe/mindspore/compare/ms_graph_compare.py,sha256=Jqbxa0k1WlLf-7TZ-mlCMwZ91E2wb0vsFO2tej9NeRg,18177
-msprobe/mindspore/compare/ms_to_pt_api.yaml,sha256=NGzy6_yIArM6V0zYsW3sg3KLMJe0sr9ljKiIcHujJus,13203
-msprobe/mindspore/debugger/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/debugger/debugger_config.py,sha256=HC_maN0t4rE19irZzIFNY54MED-WTePycDnhrgqLqVw,3337
-msprobe/mindspore/debugger/precision_debugger.py,sha256=BCDd5eSHQeXR0usRYMyT2L7eD3XEmteQdXj-0eh1iaM,5359
-msprobe/mindspore/dump/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/dump/dump_tool_factory.py,sha256=4yKhJQSUfUUPEdkx9GXu-M3IAhObYHXmYC7llyzvhH0,1851
-msprobe/mindspore/dump/jit_dump.py,sha256=iCgXvskdPpT3bkTOS24QrGHPOYRvf46m6O1ODemZyKE,3970
-msprobe/mindspore/dump/kernel_graph_dump.py,sha256=hTUkEHJyxtFSkqNtEVjZ2o2HQYmVAXx-81p6gyApKJE,3672
-msprobe/mindspore/dump/kernel_kbyk_dump.py,sha256=9yXl_7TvogO9AhEx7dgRegDFRgURdWSUmEWg94XPIPA,2952
-msprobe/mindspore/dump/hook_cell/api_registry.py,sha256=Oy2NNO1Dc_NvBaM7qsszEK2PZvLZ4ZP7BZ8PYE_LP1o,6276
-msprobe/mindspore/dump/hook_cell/hook_cell.py,sha256=XF9SeZHQeTYil85aG6UH0eaNGTaJgPTg1OElDoSU2XA,1983
-msprobe/mindspore/dump/hook_cell/primitive_hooks.py,sha256=8TXRMnlPxaA18riOgh-qKacm6btiruGEzlQ7P99p_5E,9759
-msprobe/mindspore/dump/hook_cell/support_wrap_ops.yaml,sha256=To0fOzxAb9gkqV-cpZ3yCfAUitlxzfC5OohWzKV2bz0,13148
-msprobe/mindspore/dump/hook_cell/wrap_api.py,sha256=1fuLtAHfC_FdXlKe_9DAFihyhv2srkhxw8LTt8nwM1U,4909
-msprobe/mindspore/free_benchmark/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/free_benchmark/api_pynative_self_check.py,sha256=IwF_kaf3J82TPhueRM3EmB64pgAhW1gcw9WkluRvxBQ,4485
-msprobe/mindspore/free_benchmark/self_check_tool_factory.py,sha256=93hI__EICoU0fy-PAjFlW47LrtEI8qDm2AL67YExJxo,1761
-msprobe/mindspore/free_benchmark/common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/free_benchmark/common/config.py,sha256=xK1hEV8pHH6upC8-0ciio-GgN1-Q_DAMSdv1a__fhTU,1008
-msprobe/mindspore/free_benchmark/common/handler_params.py,sha256=jzpMBxgPlnvSq7MliGdmPnTixVaUjrOZyXzIRQ_nfy4,1083
-msprobe/mindspore/free_benchmark/common/utils.py,sha256=PLVu0Pwmq4uIPSS6ZpeZMEqvVsAen0aY2Erz_USe618,2837
-msprobe/mindspore/free_benchmark/data/support_wrap_ops.yaml,sha256=lsGOW1R6w2_xZsZ4L012lmeXTUwa-BcjpvNske4U9jU,10579
-msprobe/mindspore/free_benchmark/decorator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/free_benchmark/decorator/dec_forward.py,sha256=LYNUwiCpGte-r_GecwCuF5IcTa1a6W2lWrgwfO4zL1E,2385
-msprobe/mindspore/free_benchmark/decorator/decorator_factory.py,sha256=IiPSDsGy_gpY3eRqC0cl3mrRJPtN5Cc6IK2rCe1liZQ,3586
-msprobe/mindspore/free_benchmark/handler/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/free_benchmark/handler/base_handler.py,sha256=ABlPCir7fvgHsyck7KMalbL4NYsX3Nm7VSgY9_kYXMM,4556
-msprobe/mindspore/free_benchmark/handler/check_handler.py,sha256=zANPnKwyeLV0snl4bDG3C3BY6I4hHR0PXMHjNZhROPE,2526
-msprobe/mindspore/free_benchmark/handler/fix_handler.py,sha256=i148P8E4sVCPdWDOmanT44o5C58zxGlUjMC5nlOJU5I,2000
-msprobe/mindspore/free_benchmark/handler/handler_factory.py,sha256=ThcxVntpqGdBYvvSLtH9nugIutgTKd3nKCyPX14SPhY,1394
-msprobe/mindspore/free_benchmark/perturbation/add_noise.py,sha256=FSvBxnumdbBa5to4OhEwYqH8krUMl-CMIYgJfQbVAhU,2763
-msprobe/mindspore/free_benchmark/perturbation/base_perturbation.py,sha256=LPpiPuZoZURwJ6n_ANZlGLAwZlOZMJCV9lWHLz2pSQw,1270
-msprobe/mindspore/free_benchmark/perturbation/bit_noise.py,sha256=xcADRqz7QSV1IgGLlfqpwLcTjZ5dSH3Wo0RBijYagHI,3399
-msprobe/mindspore/free_benchmark/perturbation/exchange_value.py,sha256=TmDe7jMCBoakS3D0xiq1fGYVbA-bzwQTByMxvYsKtI0,2686
-msprobe/mindspore/free_benchmark/perturbation/improve_precision.py,sha256=qNJE4PiH3lG9KaMK3yfdbKf1E3-RAhQa_WRvhVFgPSo,2131
-msprobe/mindspore/free_benchmark/perturbation/no_change.py,sha256=fMwZxe0AgcHNNIRwPxuwMGb2-rCvveBEsfWu-LwucW8,1055
-msprobe/mindspore/free_benchmark/perturbation/perturbation_factory.py,sha256=ljxWcqJhbSfzPJBvxzRspVMw4yn7a4AKDqKMNHhZAH8,1934
-msprobe/mindspore/grad_probe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/grad_probe/global_context.py,sha256=4j-eNTszrzy9X-1ITDMWHaPTi3Xr47-qRWEBun1Hl4k,4013
-msprobe/mindspore/grad_probe/grad_analyzer.py,sha256=KARp8BRqX8KsgXiIhw9C1iriG-Ph1xppXLA-ZnXBUK0,9315
-msprobe/mindspore/grad_probe/grad_monitor.py,sha256=kM5eH_-pz1TbCaE87H4XkOe9ndDqBHEdaokmuW7mDs4,935
-msprobe/mindspore/grad_probe/grad_stat_csv.py,sha256=7_m_C4P7Yi-gAjXyLPW-d2cvRBR4v2CRzfg0hPMWMEM,3714
-msprobe/mindspore/grad_probe/hook.py,sha256=F4lFSkQEsJfRC2fNJXKZwwaH3agzXNRUXV1HmP-7zmU,4314
-msprobe/mindspore/grad_probe/utils.py,sha256=e7RAx4hyP1tciyHCtLCXDTbkuVZsWZxgltySEGo2Rog,1055
-msprobe/mindspore/overflow_check/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/mindspore/overflow_check/kernel_graph_overflow_check.py,sha256=aJbFygfpWJoxEgMpuaAwDsARaniyyqrhtHvM5TPU3NI,2831
-msprobe/mindspore/overflow_check/overflow_check_tool_factory.py,sha256=fHWkpwHZuYOylVX8DxEgIt3XLf4VlP78UWYxZDyq89M,1820
-msprobe/pytorch/__init__.py,sha256=dreYB-ntHrWrDG8F9VX6FKxXnBdXQZhxYxX_7wiP2vY,932
-msprobe/pytorch/function_factory.py,sha256=TS7qE0YpL-aD4T5gQon8pXRkb2Tx5TmNTatE6lH7Dic,3568
-msprobe/pytorch/module_processer.py,sha256=jz_yZvbeICnwGuEDTKnrHp7Gvk-tlNedp8frJW1czO4,6277
-msprobe/pytorch/parse.py,sha256=TmMxmRfIfvvHzKGoDT5sKUdnt2PonwwwOKg1H7NsbVo,711
-msprobe/pytorch/pt_config.py,sha256=hq2VOtHlgMzZBEGPq9-y-TcQLUsY3lAbz1ySc6EKB-A,13502
-msprobe/pytorch/service.py,sha256=PKnF-VfxXH9ISpK-aWdd7GZHgrClbN5tn49ctE7l1mw,14735
-msprobe/pytorch/api_accuracy_checker/.keep,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/config.yaml,sha256=2kktin5NoWkSp2MOGDoxlz3MPv0lYXXD5r5SPYArEYs,144
-msprobe/pytorch/api_accuracy_checker/common/.keep,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/common/config.py,sha256=A-4yJuWLrVw-J3o_cDjzenfz959YYnhAhjGtq3MfTrg,2626
-msprobe/pytorch/api_accuracy_checker/common/utils.py,sha256=nluFpct7Kjb5_JKMrbHqc9m_dvOm4jsPKfqJsjSHWuc,8605
-msprobe/pytorch/api_accuracy_checker/compare/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/compare/algorithm.py,sha256=yWVxDoOJxaEZ5_jcmQuNwRK3td4tljkuYTyv6_V0vg4,9143
-msprobe/pytorch/api_accuracy_checker/compare/api_precision_compare.py,sha256=peEWenqJ01oOvSjz0PgXxI6AXCvg0NyA1sOOK_4DUJg,32065
-msprobe/pytorch/api_accuracy_checker/compare/api_precision_standard.yaml,sha256=CnWLLvFwI44QOE6qwcMK6ZjLfOh4PIox9202hz6ZHEc,1998
-msprobe/pytorch/api_accuracy_checker/compare/api_precision_threshold.yaml,sha256=Pt5MQFazR6DwvI19SF8wUTHp2z_SLdoDz643AUdoWcY,8112
-msprobe/pytorch/api_accuracy_checker/compare/compare.py,sha256=WmfAdXQ5A-nhX_xLnorvrDhOoQq7cvCelAE2FvBUWT4,23194
-msprobe/pytorch/api_accuracy_checker/compare/compare_column.py,sha256=7W6DpfoWBzOi0ciZ_AiDCxxkZAGovUkZzHWWJESPDzA,4578
-msprobe/pytorch/api_accuracy_checker/compare/compare_utils.py,sha256=cquQl2OyXz7_sC6GKJJz71JIfPMGKkHScceukwVtcm8,9874
-msprobe/pytorch/api_accuracy_checker/run_ut/.keep,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/run_ut/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/run_ut/data_generate.py,sha256=XnDCv_rP5fz5QRRcRM2aF2MO0NS4lM85ohK9Qhpy68g,14950
-msprobe/pytorch/api_accuracy_checker/run_ut/multi_run_ut.py,sha256=HCn8-5jbs8g-NpWwkgF4QxJtzzXvaQ0Xf1EAGmHCJEs,9168
-msprobe/pytorch/api_accuracy_checker/run_ut/run_overflow_check.py,sha256=Bjarp1uxMM9wKEfOG8K40TeKRo6MmU9PZgsGvkmK1aY,6147
-msprobe/pytorch/api_accuracy_checker/run_ut/run_ut.py,sha256=XUwNQsoy3FB4Sanrmu5-oi_jV01yi4F_GAJ6xoMNCok,24211
-msprobe/pytorch/api_accuracy_checker/run_ut/run_ut_utils.py,sha256=TIwzS3ts62O5LbSH0zTaCtB6j-9IE7JCu8yLMXaebgo,9565
-msprobe/pytorch/api_accuracy_checker/run_ut/torch_ut_setting.json,sha256=6caXsuV9H9eDG2Zd0m_qJG_TsZ2Mk8z8RV8tS7r2m4o,92
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/attl.py,sha256=5wOFGk5fuUoa3cjJbp6NQujGKEKAY40oD3mFu3rmlow,8492
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/client.py,sha256=UYFSp8eTZC9_6k-o215skY4Tx_DCIOa4D3OaYgCb1Do,14441
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/device_dispatch.py,sha256=8zDT-TrD-JtZC-ZshPGqvqsd9UFcffx7BFCaoVjjTpw,10270
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/dump_dispatch.py,sha256=1hHEwuslHQHSEMOSYncBDJ4fY2hULjpc50oXhbjg_UA,3767
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/server.py,sha256=QXLNHDfAV7BLFfvI-7tT1zNKhU2kiZoaxMXEfgbPQCM,10124
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/torch_ops_config.yaml,sha256=AAEAsbIK8R1O9c9kDb6svfeklqt2W8GHf19ewKpEROo,1057
-msprobe/pytorch/api_accuracy_checker/tensor_transport_layer/utils.py,sha256=aDQmCdDA_FV-5gx7WlvrVYxI77ZaO9kEnn4CwvSF17k,1791
-msprobe/pytorch/bench_functions/__init__.py,sha256=BoY2ZOpo_mj2gQ8pcl8oLwFxfHpZidOflJ1aGYDmPrM,1157
-msprobe/pytorch/bench_functions/apply_adam_w.py,sha256=tUkQEmmUf__kIkBTagtjphR7HaiZRpDjk_q7j6AGl00,1690
-msprobe/pytorch/bench_functions/confusion_transpose.py,sha256=XdXuilYWVNKAbpP81U1lb4I0-ZOJ4J-SllyNpm9ywsA,1294
-msprobe/pytorch/bench_functions/fast_gelu.py,sha256=L3J2FNiHY_FCjWXLKuf3fCPLf7oE7kBVaYBjXUstR4s,1883
-msprobe/pytorch/bench_functions/layer_norm_eval.py,sha256=kObMeu4JKPltaGFXqf64cFcNVEsH8W-jW1m2PHbPPzA,808
-msprobe/pytorch/bench_functions/linear.py,sha256=jZ7C4jcKEBmM9t74P1AkPgJK1IXoawyxaQtultiO0nE,944
-msprobe/pytorch/bench_functions/matmul_backward.py,sha256=LIpTJMkTKCGdU2c6qa0u-gDuEEnalbGcf2jitfKsoWs,3150
-msprobe/pytorch/bench_functions/npu_fusion_attention.py,sha256=051Bv2NkDP4ohvXR6SbW61lwq1FGhCsbH7VblA69_Bs,22422
-msprobe/pytorch/bench_functions/rms_norm.py,sha256=H5ukqQoxSWAHJoFeyZP_Jl6qn-nFNPrIfKedLNWPwOk,1072
-msprobe/pytorch/bench_functions/rotary_mul.py,sha256=kIkYe-1RIyxGaZiodQL2cnus40iiG17_D-lLK1SgdQc,2721
-msprobe/pytorch/bench_functions/scaled_mask_softmax.py,sha256=OAZq05MRKhBYuJ-962QeQMgO1xuCdTNd__XAgDf_5qQ,1518
-msprobe/pytorch/bench_functions/swiglu.py,sha256=Z1uyVdPYhPF27taYqZA3HyXIEW7Noq_7Y8JPlz6fqjk,3017
-msprobe/pytorch/common/__init__.py,sha256=87H-l6y1gbEL2EQiSbPceeJwKJdcsS6zv-9NmrTJlTo,709
-msprobe/pytorch/common/compare_script.template,sha256=CsfmJ5IuPycAYa0-IfVPXrXVu-o6g8BeEF9v17UK_1w,286
-msprobe/pytorch/common/log.py,sha256=yGLpZ6mGfB-s9sHj_gcL5_PMq_Ai4BJxumA0F0knAso,1121
-msprobe/pytorch/common/parse_json.py,sha256=Vq7qjvUA_2Rc5QKCpWb5TIk5akjhFLrqC0pJk8ihdpA,2136
-msprobe/pytorch/common/utils.py,sha256=fBRmBGning3_vfPweyUg0fd24h9rko5psiEOcAHq0_g,9918
-msprobe/pytorch/compare/distributed_compare.py,sha256=JR_qkymeduHErL1NFq5mWDnwxEz7Lwdq314ZFp2JEKg,3498
-msprobe/pytorch/compare/mapping.yaml,sha256=Udlbf1dWDAdnTs8VcDTiSkqsMNj93o2qFUU2ocFgt2k,11697
-msprobe/pytorch/compare/match.py,sha256=j6ydfhs9zNMu6vzxsp_qMei-7-zDMuPO8Uj9i_OIhSA,1886
-msprobe/pytorch/compare/pt_compare.py,sha256=hXWTzaWT9JBrI0GLDECoYyEFZGSZ4WLqTwTVliZfVE8,4034
-msprobe/pytorch/debugger/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/debugger/debugger_config.py,sha256=K6k_hKX8JW7a_yEu28nVCftNqomk5K9vchS9UelVN2w,5629
-msprobe/pytorch/debugger/precision_debugger.py,sha256=3WBFzTtufLnyNQQPSU40dWWC3wR1r6ng92c2Y4NbNtQ,6876
-msprobe/pytorch/free_benchmark/__init__.py,sha256=N2xdPic80U-Rl1B_6c8j09EMuU5Xn0knJjgmKQ5uiXQ,904
-msprobe/pytorch/free_benchmark/main.py,sha256=5SSJGQ-eYOsguCCHaCLiOPuSrlbEHuWkaIk6T4534UI,4634
-msprobe/pytorch/free_benchmark/common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/free_benchmark/common/constant.py,sha256=UeBE1OREX3cEJOEEcpq4O-yN7Xu-CfB-KXpAx1NGZXI,1929
-msprobe/pytorch/free_benchmark/common/counter.py,sha256=LqRx7cZ9rr5bD4ueOSe6i_2lyWb4yEDpBVXuWKepg4Y,2718
-msprobe/pytorch/free_benchmark/common/enums.py,sha256=GFv7XBseR6EUdg7e1WSdrV9iM00QjFmq6N4fhqKqtiE,1631
-msprobe/pytorch/free_benchmark/common/params.py,sha256=6bgB5T2y4cGYQa9TVlzgbqpdkjEOdEEku8Is5dCFt9Q,4614
-msprobe/pytorch/free_benchmark/common/utils.py,sha256=q11OgMkDqrUYcCMwPxgC_230sOdHOGy03TyUPHK2GBg,4613
-msprobe/pytorch/free_benchmark/compare/grad_saver.py,sha256=oDOmQGtM0_JAEq6NUcVrIWXTh8S-8LUvHc6uHbd3fJ4,8278
-msprobe/pytorch/free_benchmark/compare/single_benchmark.py,sha256=8pLQ1H4wb54QzIJA9e-q8y8xwMXtedCdiq3DF0_9lho,4489
-msprobe/pytorch/free_benchmark/perturbed_layers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/free_benchmark/perturbed_layers/base_layer.py,sha256=Vn2KsiLmI69ssXPPrDGDWhBwnduJpI4xltdEMuW1U0s,939
-msprobe/pytorch/free_benchmark/perturbed_layers/layer_factory.py,sha256=P_QPdsRJDL6R38f3m9Moq1f8J-LNaj5MQJkjwhQEzjE,2391
-msprobe/pytorch/free_benchmark/perturbed_layers/run_cpu.py,sha256=lh6wPOmztdcbvfkhlu6HubAoEjfJaB2UEyjbeEWsYQ0,1480
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/add_noise.py,sha256=4yj7riWxVbiUfez8LpOvndiwIpAnbJ69-PblQTgfBCk,4100
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/bit_noise.py,sha256=1GJp8TTNmAvWpQifffks3s_m118ihPKQ8aXj60KNcFw,4571
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/change_value.py,sha256=3yRS1BfqHnsOtWMAIl8Lim9j8B-VTz0il47nGnhYeYQ,3740
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/improve_precision.py,sha256=R_IY4yHC2hETkUvFkhmXAqxtvfzJQNhVFzRt8MXXNqo,3342
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/no_change.py,sha256=D2jFon-qeJTIbF_vd_eXJiRDVlyIv_WWxDpnNR9NxSo,1542
-msprobe/pytorch/free_benchmark/perturbed_layers/npu/npu_base_layser.py,sha256=vVvV9s7XdE3yF5S06gippUEiEqsARtX6cwcA9bObRzo,2181
-msprobe/pytorch/free_benchmark/result_handlers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/free_benchmark/result_handlers/base_handler.py,sha256=r6hcfVPFtJi4nPRdHotqYKgghxwEBW9_CtlTDI96DZ0,10287
-msprobe/pytorch/free_benchmark/result_handlers/check_handler.py,sha256=d47cgxL0HIreeopj7pfL53_erhzxaI8-gDJ5nXvkFaY,2246
-msprobe/pytorch/free_benchmark/result_handlers/fix_handler.py,sha256=fBJ0J3Qkx_RJBCsoGQpZPRZVrJpLRWo6Ts1jSI7-DPw,1481
-msprobe/pytorch/free_benchmark/result_handlers/handler_factory.py,sha256=D-J2kezk9bhiLMs8I1SesZ91CuozO73KtJY7g2l1wT8,1941
-msprobe/pytorch/free_benchmark/result_handlers/preheat_handler.py,sha256=dKSntcNAF2B0Yqp5YJkfntDIlM8Zc6X66TuL9BcB6Po,7848
-msprobe/pytorch/functional/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/functional/module_dump.py,sha256=cFf_FL6F9p1DNr9eMB3D5o_t-EOCXXIicpPxYvGbwD8,3837
-msprobe/pytorch/grad_probe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/grad_probe/grad_monitor.py,sha256=o0QTmhmemxMKWTvHViW69ZFFyLKcJz9PButFMF02x8c,4282
-msprobe/pytorch/grad_probe/grad_stat_csv.py,sha256=ExksEWUaLsMoLnuz9SFub7VD1H8B6ye76o8NDb4p-F8,4046
-msprobe/pytorch/hook_module/__init__.py,sha256=ki_iOt1mnk-qTTl39Sm3ygQ86vG4EV2gwcSViQwkWRs,668
-msprobe/pytorch/hook_module/api_registry.py,sha256=V8bnCPlvrV2v3l2wtC3Dhr9SMeGp4HSc8K6i63K8Cd4,8104
-msprobe/pytorch/hook_module/hook_module.py,sha256=Kl5NzpGz2ZWiAZ3KwE3jCkYARA2Chx-E3Wq0e35iIms,4864
-msprobe/pytorch/hook_module/support_wrap_ops.yaml,sha256=jovioDnO0KW2tSRoPhxaxIpQV8ULAXkd-YtnsZutcRY,27870
-msprobe/pytorch/hook_module/utils.py,sha256=_m3M3Ad_Xtkm_ZZCo7rJzX9qeG3ocDYZQQtMqTnOuoc,1101
-msprobe/pytorch/hook_module/wrap_aten.py,sha256=tjD1CkresxQBr-iOhO5Hu4fGwSYpKNNhZv7XUBBFfus,3860
-msprobe/pytorch/hook_module/wrap_distributed.py,sha256=3P7Dn_1-OoOA1xiwJdfCggyLfji8skrz8K9vIG8JlBk,2690
-msprobe/pytorch/hook_module/wrap_functional.py,sha256=V12FL2RCW3vQeoD8LTSRXTtRSBR1aoT1eYhxizHvHUE,4474
-msprobe/pytorch/hook_module/wrap_npu_custom.py,sha256=gzhBVSygl32puWMZKmmSQfh9_f0lY_-V24pBNPOjVLo,3058
-msprobe/pytorch/hook_module/wrap_tensor.py,sha256=gWpEBg5BI1PqUYBz1XfSd9oPsau26w4iusmIrlH2m6c,2087
-msprobe/pytorch/hook_module/wrap_torch.py,sha256=eJL_fTCm1K4NVKrjEqX-41zaa0aTqdNKyDdWC7kMXyg,2617
-msprobe/pytorch/hook_module/wrap_vf.py,sha256=YpL1aPtVMszXosHiMtJgvLdLu4mBc2xjnNp9WvBrjGQ,1831
-msprobe/pytorch/online_dispatch/__init__.py,sha256=26TwsLkd9AoRbnUOjZ4oc4mv2S8sTXFucu_7M3r2v2M,734
-msprobe/pytorch/online_dispatch/compare.py,sha256=T9_UacfVbBmRtG9oO1ZfIIw7fKHwUZ4Qj9y77lnn--s,10593
-msprobe/pytorch/online_dispatch/dispatch.py,sha256=WQaU6dwNxbwycSnEVQ8cfU7PaqC6vK11uv30FoNnKI0,11971
-msprobe/pytorch/online_dispatch/dump_compare.py,sha256=F6pYomr0wk_j7z2U7kRvbpgRt6XfY7gODH1fMIwiefI,5870
-msprobe/pytorch/online_dispatch/single_compare.py,sha256=qoCbE8UCL2_NbydfsO_ueAjFC60De9rprG9uINt993M,16575
-msprobe/pytorch/online_dispatch/torch_ops_config.yaml,sha256=TvkG1_r78Ij10l2iFS0YqqY1sku2eiw3yG3JS65mCuc,872
-msprobe/pytorch/online_dispatch/utils.py,sha256=G55I1xojA-rQun3E0qmXa0Q-9do4XCMZR5Te3uOR7Sg,4083
-msprobe/pytorch/parse_tool/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/parse_tool/cli.py,sha256=aPk72KPC1Z10zU-tl869nf6J1pv8h05MuxB9OwI9YCQ,1048
-msprobe/pytorch/parse_tool/lib/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-msprobe/pytorch/parse_tool/lib/compare.py,sha256=Dt9AqbZfmyWUPp1aMk7tDJwRaPE3H28TI5eXjtI9908,13325
-msprobe/pytorch/parse_tool/lib/config.py,sha256=kLbz5EFkOgl3T4lRLmmoZgkpJ41yqH4StVjNgucs7xA,1992
-msprobe/pytorch/parse_tool/lib/file_desc.py,sha256=l0RCGU4-DCM5p5iPuCGLcjyMd710owdjnuTKYAZ0kdc,1147
-msprobe/pytorch/parse_tool/lib/interactive_cli.py,sha256=wkbpBK6LGufJvfFANoermyWLNLez8CR4ipcqv0DmBIo,3678
-msprobe/pytorch/parse_tool/lib/parse_exception.py,sha256=0gRawclT_z1U0K9FkawMgFuGz3zvvqGeIgAbxZB6pSw,1763
-msprobe/pytorch/parse_tool/lib/parse_tool.py,sha256=aJ846B-7Kk-JtqatoPvdjsSNZomeqAQEh3vPfZx53DU,8105
-msprobe/pytorch/parse_tool/lib/utils.py,sha256=004omfhfleECQ8FZ5B4eZnRY85owe6zH__vf4IVdD2I,12517
-msprobe/pytorch/parse_tool/lib/visualization.py,sha256=mppLLY-UBKQSXUA6Wbtj-rRxas4_SK-5rBV2GWvGb7Y,4063
-mindstudio_probe-1.1.0.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-mindstudio_probe-1.1.0.dist-info/METADATA,sha256=hKCEcts8M8OsIWLl4tzsc6uhL-l30nk8ChTH3Zyhrg8,1308
-mindstudio_probe-1.1.0.dist-info/WHEEL,sha256=G16H4A3IeoQmnOrYV4ueZGKSjhipXx8zc8nu9FGlvMA,92
-mindstudio_probe-1.1.0.dist-info/entry_points.txt,sha256=D2Vi0uZD_DqVIrJ9_Cw7y9SP5AHH4-8VHLjlPXUjIpk,50
-mindstudio_probe-1.1.0.dist-info/top_level.txt,sha256=LxFEFqelENSyWmRtocCiEUF04IE8aZvwTl7ADB598Tk,8
-mindstudio_probe-1.1.0.dist-info/RECORD,,

msprobe/docs/04.acl_config_examples.md

@@ -1,78 +0,0 @@
-# acl_config.json 配置文件说明
-
-当 PyTorch 场景 level 取"L2"时，须配置 acl_config 参数，并指定 acl_config.json 文件（用于指定 L2 kernel 级 dump 的配置），此时 **config.json** 文件配置示例如下：
-
-## 1 前向 kernel dump 配置示例
-
-"scope"配置为前向 API 名称，仅支持配置一个 API。
-
-```json
-{
-	"task": "tensor",
-	"dump_path": "/home/data_dump",
-	"level": "L2",
-	"rank": [0],
-	"step": [0],
-	"is_deterministic": false,
-
-	"tensor": {
-		"scope": ["Tensor.__mul__.10.forward"],
-		"list":[],
-		"data_mode": ["all"],
-		"backward_input": [""],
-		"file_format": "npy"
-		},
-	"acl_config": "acl_config.json"
-}
-```
-
-## 2 反向 kernel dump 配置示例
-
-执行反向 kernel dump 前需要先使用工具 dump 该 API 的反向输入，保存 pt 文件，"backward_input"参数中传入该 pt 文件路径。
-
-"scope"配置为反向 API 名称，仅支持配置一个 API。
-
-```json
-{
-	"task": "tensor",
-	"dump_path": "/home/data_dump",
-	"level": "L2",
-	"rank": [0],
-	"step": [0],
-	"is_deterministic": false,
-
-	"tensor": {
-		"scope": ["Tensor.__mul__.10.backward"],
-		"list":[],
-		"data_mode": ["all"],
-		"backward_input": ["Tensor.__mul__.10.backward.input.0.pt"],
-		"file_format": "npy"
-		},
-	"acl_config": "acl_config.json"
-}
-```
-
-## 3 acl_config.json 配置示例
-
-该文件须自行创建，配置示例如下：
-
-```json
-{
-	"dump":
-	{
-		"dump_list":[],
-		"dump_path":"./dump/output",
-		"dump_mode":"all",
-		"dump_op_switch":"on"
-		}
-}
-```
-
-**acl_config.json 参数说明**
-
-| 字段名         | 解释                                                         |
-| -------------- | ------------------------------------------------------------ |
-| dump_list      | 待 dump 数据的 API 模型。为空，无需配置。                        |
-| dump_path      | dump 数据文件存储到运行环境的目录，主要配置的是 kernel 级数据的存放路径。支持配置绝对路径或相对路径。dump_path 须为已存在目录。 |
-| dump_mode      | dump 数据模式，可取值，<br/> output：dump API 的输出数据，默认值；<br/> input：dump API 的输入数据；<br/> all：dump API 的输入、输出数据。 |
-| dump_op_switch | 单 API 模型 dump 数据开关，可取值，<br>        off：关闭单 API 模型 dump，默认值；<br>        on：开启单 API 模型 dump。 |

msprobe/mindspore/compare/layer_mapping.py

@@ -1,146 +0,0 @@
-import re
-
-from msprobe.core.common.const import Const
-from msprobe.core.common.log import logger
-from msprobe.core.common.utils import CompareException
-
-
-class Trie:
-    def __init__(self, type_name=None, has_data=False):
-        self.type_name = type_name
-        self.call_count_list = []
-        self.children = {}
-        self.has_data = has_data
-        self.node_type = None
-
-    def __repr__(self):
-        return (f"Node(type_name={self.type_name}, "
-                f"has_data={self.has_data}, call number={len(self.call_count_list)})")
-
-    def insert(self, word, word_type="func"):
-        parts = word.split(Const.SEP)
-        if len(parts) < 2:
-            logger.error('result dataframe elements can not be access.')
-            raise CompareException(CompareException.INDEX_OUT_OF_BOUNDS_ERROR)
-        """
-        xxx, node_name, type_name, execute_num
-        etc: Cell.network_with_loss.language_model.encoder.layers.1.attention.out_proj.RowParallelLinear.1
-        prefix_name_list: Cell.network_with_loss.language_model.encoder.layers.1.attention
-        node_name: out_proj
-        type_name: RowParallelLinear
-        call_count: 1
-        """
-        type_name = parts[-2]
-        call_count = parts[-1]
-        node = self
-        prefix_name_list = parts[:-2]
-
-        for name in prefix_name_list:
-            if name not in node.children:
-                node.children[name] = Trie()
-            node = node.children[name]
-            if node.type_name is None:
-                node.type_name = name
-
-        node.type_name = type_name
-        node.has_data = True
-        node.call_count_list.append(call_count)
-        node.node_type = word_type
-
-
-class DFSConverter:
-    def __init__(self, mapping, max_depth=100):
-        self.mapping = mapping
-        self.max_depth = max_depth
-        self.result = {}
-
-    def traverse_and_collect(self, node, path="", mapping_path="", depth=0):
-        if depth > self.max_depth:
-            logger.error("The converted data depth is too large, please check the data")
-            raise CompareException(CompareException.RECURSION_LIMIT_ERROR)
-
-        if node is None:
-            return self.result
-
-        type_name = node.type_name
-        if node.has_data:
-            for count in node.call_count_list:
-                origin_name = f"{path}.{count}" if node.node_type == "Cell" else f"{path}.{type_name}.{count}"
-                mapping_name = f"{mapping_path}.{count}" if node.node_type == "Cell" else f"{mapping_path}.{type_name}.{count}"
-                self.result[origin_name] = mapping_name
-
-        name_mapping = self.mapping.get(type_name, {})
-
-        for child_name, child_node in node.children.items():
-            new_path = f"{path}.{child_name}" if path else child_name
-            converted_name = name_mapping.get(child_name, child_name)
-            new_mapping_path = f"{mapping_path}.{converted_name}" if mapping_path else converted_name
-            self.traverse_and_collect(child_node, new_path, new_mapping_path, depth+1)
-
-        return self.result
-
-
-def get_mapping_list(ms_tree, mapping):
-    dfs_converter = DFSConverter(mapping)
-    ms_pt_mapping = dfs_converter.traverse_and_collect(ms_tree)
-    mapping_list = []
-    for ms_name, pt_name in ms_pt_mapping.items():
-        pt_name = re.sub(r"^Cell", "Module", pt_name)
-        mapping_list.append((ms_name, pt_name))
-    return mapping_list
-
-
-def get_prefix_mapping(scope_list):
-    """layer name to layer name.class_name"""
-    layer_mapping = {}
-    for name, v in scope_list.items():
-        origin_data = v.get("origin_data")
-        if not origin_data.startswith(("Cell", "Module")):
-            continue
-        name_list = name.split(Const.SEP)
-        if len(name_list) < 2:
-            logger.error('result dataframe elements can not be access.')
-            raise CompareException(CompareException.INDEX_OUT_OF_BOUNDS_ERROR)
-        prefix_name_list = name_list[:-2] + [name_list[-1]]
-        prefix_name = Const.SEP.join(prefix_name_list)
-        layer_mapping[prefix_name] = name
-    return layer_mapping
-
-
-def get_layer_mapping(ms_scope_list, pt_scope_list, mapping):
-    # 1. get layer prefix to full name mapping
-    # ect: Cell.network_with_loss.language_model.embedding.3 : Cell.network_with_loss.language_model.embedding.Embedding.3
-    ms_prefix2fullname = get_prefix_mapping(ms_scope_list)
-    # 2. build trie tree
-    ms_tree = Trie(type_name="Cell")
-    for k, r in ms_scope_list.items():
-        origin_data_name = r.get('origin_data')
-        data_type = origin_data_name.split(Const.SEP)[0]
-        ms_tree.insert(k, data_type)
-    msname2ptname = get_mapping_list(ms_tree, mapping)
-    # 3. get pt layer prefix to full name mapping
-    # ect: Module.network_with_loss.language_model.embedding.3 : Module.network_with_loss.language_model.embedding.Embedding.3
-    pt_prefix2fullname = get_prefix_mapping(pt_scope_list)
-
-    final_mapping = []
-    for ms_name, pt_name in msname2ptname:
-        final_ms_name = ms_name
-        final_pt_name = pt_name
-        # cell
-        if ms_name in ms_prefix2fullname:
-            final_ms_name = ms_prefix2fullname.get(ms_name)
-            final_pt_name = pt_prefix2fullname.get(pt_name, None)
-        # func
-        elif final_ms_name in ms_scope_list:
-            final_ms_name = ms_scope_list.get(ms_name)['origin_data']
-            # remove forward/backward
-            final_ms_name = Const.SEP.join(final_ms_name.split(Const.SEP)[:-1])
-            final_pt_name = pt_scope_list.get(pt_name, None)
-            if final_pt_name:
-                final_pt_name = final_pt_name['origin_data']
-                final_pt_name = Const.SEP.join(final_pt_name.split(Const.SEP)[:-1])
-        else:
-            continue
-        final_mapping.append((final_ms_name, final_pt_name))
-
-    return final_mapping

msprobe/mindspore/compare/modify_mapping.py

@@ -1,107 +0,0 @@
-from msprobe.core.common.const import Const
-from msprobe.core.common.log import logger
-
-def find_regard_scope(lines, start_sign, end_sign):
-    # 找出 start_pos 和 end_pos
-    start_pos = end_pos = -1
-    for idx, ii in enumerate(lines):
-        if start_sign in ii:
-            start_pos = idx
-        elif end_sign in ii:
-            end_pos = idx
-            break
-    return start_pos, end_pos
-
-
-def find_stack_func_list(lines):
-    res_list = []
-    # 过滤和处理 regard_scope
-    for line in lines:
-        ele_list = line.split(',')
-        file_ele = ele_list[Const.STACK_FILE_INDEX]
-        if any(ii in file_ele for ii in Const.FILE_SKIP_LIST):
-            continue
-
-        func_ele = ele_list[Const.STACK_FUNC_INDEX]
-        if any(ii in func_ele for ii in Const.FUNC_SKIP_LIST):
-            continue
-
-        in_func_name = func_ele.split()[Const.STACK_FUNC_ELE_INDEX]
-
-        res_list.append(in_func_name)
-    # 反转res_list并生成final_res
-    reversed_list = res_list[::-1]
-    return reversed_list
-
-
-def get_duplicated_name(components):
-    duplicated_components = components
-    if len(components) < 3 or components[Const.CONSTRUCT_NAME_INDEX].isdigit():
-        logger.warning("key in construct.json is shorter than 3 parts or not name valid.")
-    else:
-        # 重复name，如Functional.add.add.X ward
-        duplicated_components = components[:Const.CONSTRUCT_NAME_INDEX + 1] + components[Const.CONSTRUCT_NAME_INDEX:]
-    return duplicated_components
-
-
-def modify_mapping_with_stack(stack, construct):
-    if not stack or not construct:
-        return {}
-
-    # 是否是mindspore的数据结构
-    is_ms = any("Cell" in ii for ii in construct)
-    # 调整后的mapping结构
-    final_pres = {}
-    # 查看归属关系
-    for key in construct:
-        key_components = key.split(Const.SEP)
-        code_list = stack.get(key, None)
-        parent_node = construct.get(key, None)
-        # 名称如果非标准开头，转为标准开头
-        if not key.startswith(("Module", "Cell")):
-            # 如果没有拿到父属scope name，默认顶级域名为Module或Cell
-            if not parent_node:
-                # 将节点名字转为标准的Module或Cell
-                key_components[0] = "Cell" if is_ms else "Module"
-                # 重复该节点的名字作为类型 如add.add add在-3位置
-                duplicated_components = get_duplicated_name(key_components)
-                modified_key = Const.SEP.join(duplicated_components)
-
-                modified_key = modified_key.replace(".forward", "").replace(".backward", "")
-                final_pres[modified_key] = {Const.ORIGIN_DATA: key, Const.SCOPE: None, Const.STACK: None}
-                continue
-            parent = parent_node.split(Const.SEP)
-            if len(parent) < 4:
-                logger.info(f"Parent name in construct.json is not valid")
-                continue
-            parent_idx = Const.NAME_FIRST_POSSIBLE_INDEX if not \
-            parent[Const.NAME_FIRST_POSSIBLE_INDEX].isdigit() else Const.NAME_SECOND_POSSIBLE_INDEX
-            parent_name = parent[parent_idx]
-
-            if code_list:
-                # {name}.Class.count_number.X ward Or {name}.Class.count_number.X ward.ele_number
-                if parent_name.endswith('s'):
-                    parent_name = parent_name[:-1]
-                if len(key_components) < 3:
-                    logger.info("The length of key in construct is less than 3, please check")
-                    continue
-                # {name}.count_number.X ward
-                func_name = key_components[-3]
-                start_pos, end_pos = find_regard_scope(code_list, func_name, parent_name)
-
-                # 获取指定范围的代码
-                regard_scope = code_list[start_pos:end_pos]
-
-                func_stack_list = find_stack_func_list(regard_scope)
-            else:
-                func_stack_list = []
-            # 组合逻辑：parent的节点名（到节点名字为止）加上调用栈名[reversed_list]加上原来key重复key的节点名[key_components[1:-2] + key_components[-3:]]
-            final_res_key = Const.SEP.join(parent[:parent_idx + 1] + func_stack_list +
-                                     key_components[1:Const.CONSTRUCT_NAME_INDEX + 1] + key_components[Const.CONSTRUCT_NAME_INDEX:])
-            final_res_key = final_res_key.strip(".forward").strip(".backward")
-        else:
-            final_res_key = Const.SEP.join(key_components[:-2] + [key_components[-1]])
-            func_stack_list = []
-        final_pres[final_res_key] = {Const.ORIGIN_DATA: key, Const.SCOPE: parent_node,
-                                     Const.STACK: Const.SEP.join(func_stack_list) if func_stack_list else None}
-    return final_pres