mindstudio-probe 1.0.4__py3-none-any.whl → 1.1.0__py3-none-any.whl

{mindstudio_probe-1.0.4.dist-info → mindstudio_probe-1.1.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mindstudio-probe
-Version: 1.0.4
+Version: 1.1.0
 Summary: Pytorch Ascend Probe Utils
 Home-page: https://gitee.com/ascend/mstt/tree/master/debug/accuracy_tools/msprobe
 Author: Ascend Team

mindstudio_probe-1.1.0.dist-info/RECORD

@@ -0,0 +1,287 @@
+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/README.md

@@ -1,12 +1,12 @@
 # 📖 msprobe 使用手册
 
-![version](https://img.shields.io/badge/version-1.0.3-blueviolet)
+![version](https://img.shields.io/badge/version-1.0.4-blueviolet)
 ![python](https://img.shields.io/badge/python-3.8|3.9|3.10-blue)
 ![platform](https://img.shields.io/badge/platform-Linux-yellow)
 
-[**msprobe**](./) 是 MindStudio Training Tools 工具链下精度调试部分的工具包。主要包括精度预检、溢出检测和精度比对等功能，目前适配 [PyTorch](https://pytorch.org/) 和 [MindSpore](https://www.mindspore.cn/) 框架。这些子工具侧重不同的训练场景，可以定位模型训练中的精度问题。
+**msprobe** 是 MindStudio Training Tools 工具链下精度调试部分的工具包。主要包括精度预检、溢出检测和精度比对等功能，目前适配 [PyTorch](https://pytorch.org/) 和 [MindSpore](https://www.mindspore.cn/) 框架。这些子工具侧重不同的训练场景，可以定位模型训练中的精度问题。
 
-为方便使用，本工具提供了一个统一、简易的程序接口，**PrecisionDebugger**，以 PyTorch 框架为例，通过以下示例模板和 **config.json** 可轻松使用各种功能。
+为方便使用，本工具提供了统一、简易的程序接口：**PrecisionDebugger**。以 PyTorch 框架为例，通过以下示例模板和 **config.json** 可以轻松使用各种功能。
 
 ```python
 from msprobe.pytorch import PrecisionDebugger  
@@ -19,7 +19,19 @@ debugger.stop() # 一般在训练循环末尾结束工具
 debugger.step() # 在训练循环的最后需要重置工具，非循环场景不需要
 ```
 
-除了在训练脚本中调用接口函数，还可以通过命令行使用 **msprobe** 的其他功能，具体的使用规则和 **config.json** 的配置要求详见以下章节。
+此外，根据以下规则，可以通过环境变量设置日志级别。
+- MSPROBE_LOG_LEVEL=4，不打印任何日志；
+- MSPROBE_LOG_LEVEL=3，仅打印 ERROR；
+- MSPROBE_LOG_LEVEL=2，仅打印 WARNING、ERROR；
+- MSPROBE_LOG_LEVEL=1，仅打印 INFO、WARNING、ERROR（默认配置）；
+- MSPROBE_LOG_LEVEL=0，打印 DEBUG、INFO、WARNING、ERROR。
+
+例如在 shell 脚本：
+
+```shell
+export MSPROBE_LOG_LEVEL={x}
+```
+**config.json** 的配置要求和各功能具体的使用指导详见后续章节。
 
 ## ⚙️ [安装](./docs/01.installation.md)
 
@@ -67,9 +79,9 @@ MindSpore 动态图场景的[离线预检](./docs/09.accuracy_checker_MindSpore.
 
 ### 6 无标杆比对
 
-[PyTorch 场景的无标杆比对](./docs/15.free_benchmarking_PyTorch.md)（待补充）
+[PyTorch 场景的无标杆比对](./docs/15.free_benchmarking_PyTorch.md)
 
-[MindSpore 场景的无标杆比对](./docs/16.overflow_check_MindSpore.md)（待补充）
+[MindSpore 场景的无标杆比对](./docs/16.free_benchmarking_MindSpore.md)
 
 ### 7 梯度状态监测
 
@@ -77,25 +89,43 @@ MindSpore 动态图场景的[离线预检](./docs/09.accuracy_checker_MindSpore.
 
 [兼容 PyTorch 和 MindSpore 框架的梯度监测](./docs/17.grad_probe.md)
 
+### 8 在线精度比对
+
+在线精度比对是实现在PyTorch训练过程中直接完成精度比对并输出比对结果的功能，是NPU与CPU之间的精度比对。
+
+[PyTorch 场景的在线精度比对](./docs/18.online_dispatch.md)
+
 ## 🌟 新版本特性
 
+若查看历史版本特性，请点击[安装](./docs/01.installation.md)。
+
+【数据采集】
+- 支持 config.json 中的 step 传入范围；
+- 优化了指定 step 的机制，指定 step 结束后工具不再采集数据，但训练会继续运行。工具结束运行后，日志提示信息如下：
+    ```bash
+    ****************************************
+    *      msprobe ends successfully.      *
+    ****************************************
+    ```
+    注：在多卡场景，每张卡进程训练到指定 step 之后都会打印一次上述信息。
+
 【精度预检】
-- 落盘数据小。
-- 支持随机生成模式和真实数据模式。
-- 单 API 测试，排除整网中的累计误差问题。
+- 在 PyTorch 场景，支持部分 NPU 融合算子预检。
 
-【梯度检测】
-- 使用便捷，无需在训练流程里插入代码。
-- 可以精准定位问题出现的 step。
+【精度比对】
+- 解决了使用 MindSpore 需要安装 PyTorch 的问题。
+
+【无标杆比对】
+- 补充在 PyTorch 场景的性能基线报告；
+- 支持 MindSpore 场景的 change_value 扰动模式。
 
 ## 📑 补充材料
 
-[msprobe 标准性能基线报告](./docs/S01.report_msprobe_dump_standard_performance_baseline.md)
+[无标杆比对功能在 PyTorch 场景的性能基线报告](./docs/S02.report_free_benchmarking_validation_performance_baseline.md)
 
-[无标杆工具场景验证和性能基线报告](./docs/S02.report_free_benchmarking_validation_performance_baseline.md)
+## ❗ 免责声明
+本工具建议执行用户与安装用户保持一致，如果您要使用 root 执行，请自行关注 root 高权限触及的安全风险。
 
 ## ❓ FAQ
 
 [FAQ for PyTorch](./docs/FAQ_PyTorch.md)
-
-FAQ for MindSpore

msprobe/__init__.py

@@ -1 +1,16 @@
-from msprobe.core.grad_probe.grad_compare import GradComparator
+# Copyright (c) 2024-2024, Huawei Technologies Co., Ltd.
+# All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0  (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from msprobe.core.grad_probe.grad_compare import GradComparator

msprobe/config.json

@@ -4,8 +4,6 @@
     "rank": [],  
     "step": [],  
     "level": "L1", 
-    "seed": 1234,
-    "is_deterministic": false, 
     "enable_dataloader": false,
     "acl_config": "",
     "tensor": {

msprobe/core/advisor/advisor.py

@@ -1,8 +1,7 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-# Copyright (C) 2022-2024. Huawei Technologies Co., Ltd. All rights reserved.
-# Licensed under the Apache License, Version 2.0 (the "License");
+# Copyright (c) 2022-2024, Huawei Technologies Co., Ltd.
+# All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0  (the "License");
 # you may not use this file except in compliance with the License.
 # You may obtain a copy of the License at
 #
@@ -13,7 +12,6 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""
 
 import os
 
@@ -24,15 +22,17 @@ from msprobe.core.common.utils import CompareException
 from msprobe.core.common.file_utils import FileChecker
 from msprobe.core.common.const import Const, CompareConst, FileCheckConst
 
+
 class Advisor:
     """
     Class for generate advisor
     """
 
-    def __init__(self, input_data, out_path=""):
+    def __init__(self, input_data, out_path="", suffix=""):
         self.input_data = input_data
         self.out_path = os.path.realpath(out_path)
         self.file_type = None
+        self.suffix = suffix
 
     @staticmethod
     def deterministic_advisor(message, node_name):
@@ -103,7 +103,7 @@ class Advisor:
         else:
             result = self.gen_advisor_result(failing_data)
         message_list = result.print_advisor_log()
-        result.gen_summary_file(self.out_path, message_list)
+        result.gen_summary_file(self.out_path, message_list, suffix=self.suffix)
 
     def _parse_input_data(self):
         data_columns = self.input_data.columns.values

msprobe/core/advisor/advisor_const.py

@@ -1,8 +1,7 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-# Copyright (C) 2022-2024. Huawei Technologies Co., Ltd. All rights reserved.
-# Licensed under the Apache License, Version 2.0 (the "License");
+# Copyright (c) 2022-2024, Huawei Technologies Co., Ltd.
+# All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0  (the "License");
 # you may not use this file except in compliance with the License.
 # You may obtain a copy of the License at
 #
@@ -13,7 +12,6 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""
 
 
 class AdvisorConst:
@@ -35,7 +33,8 @@ class AdvisorConst:
     # advisor message
     NO_ERR_SUGGEST = "All data in comparison result meets the accuracy requirements."
     FORWARD_INPUT_SUGGEST = "1. Analyze the model to view the input source.\n" \
-                            "2. Check whether an inplace API causes the output result to overwrite the input result. That is, the fault is actually caused by a computation error.\n" \
+                            "2. Check whether an inplace API causes the output result to overwrite the input result. "\
+                            "That is, the fault is actually caused by a computation error.\n" \
                             "3. The fault may be caused by memory corruption and further analysis is required."
     FORWARD_OUTPUT_SUGGEST = "This is a forward API computation error. Check the computation implementation."
     BACKWARD_INPUT_SUGGEST = "Check whether the forward computation result is affected."

msprobe/core/advisor/advisor_result.py

@@ -1,8 +1,7 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-# Copyright (C) 2022-2024. Huawei Technologies Co., Ltd. All rights reserved.
-# Licensed under the Apache License, Version 2.0 (the "License");
+# Copyright (c) 2022-2024, Huawei Technologies Co., Ltd.
+# All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0  (the "License");
 # you may not use this file except in compliance with the License.
 # You may obtain a copy of the License at
 #
@@ -13,13 +12,12 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""
 import os
 import time
 
 from msprobe.core.advisor.advisor_const import AdvisorConst
 from msprobe.core.common.log import logger
-from msprobe.core.common.const import Const, FileCheckConst
+from msprobe.core.common.const import FileCheckConst
 from msprobe.core.common.file_utils import change_mode, FileOpen
 
 
@@ -34,8 +32,8 @@ class AdvisorResult:
         self.advisor_message = message
 
     @staticmethod
-    def gen_summary_file(out_path, message_list):
-        file_name = 'advisor_{}.txt'.format(time.strftime("%Y%m%d%H%M%S", time.localtime(time.time())))
+    def gen_summary_file(out_path, message_list, suffix):
+        file_name = 'advisor{}_{}.txt'.format(suffix, time.strftime("%Y%m%d%H%M%S", time.localtime(time.time())))
         result_file = os.path.join(out_path, file_name)
         try:
             with FileOpen(result_file, 'w+') as output_file:
@@ -50,9 +48,11 @@ class AdvisorResult:
 
     def print_advisor_log(self):
         logger.info("The summary of the expert advice is as follows: ")
-        message_list = [AdvisorConst.LINE + AdvisorConst.COLON + str(self.line),
-                        AdvisorConst.SUSPECT_NODES + AdvisorConst.COLON + self.suspect_node,
-                        AdvisorConst.ADVISOR_SUGGEST + AdvisorConst.COLON + self.advisor_message]
+        message_list = [
+            AdvisorConst.LINE + AdvisorConst.COLON + str(self.line),
+            AdvisorConst.SUSPECT_NODES + AdvisorConst.COLON + self.suspect_node,
+            AdvisorConst.ADVISOR_SUGGEST + AdvisorConst.COLON + self.advisor_message
+        ]
         for message in message_list:
             logger.info(message)
         return message_list