diff --git a/dev/bench/data.js b/dev/bench/data.js index e823d62d2..71e910f08 100644 --- a/dev/bench/data.js +++ b/dev/bench/data.js @@ -1,5 +1,5 @@ window.BENCHMARK_DATA = { - "lastUpdate": 1729782581421, + "lastUpdate": 1729803593144, "repoUrl": "https://github.com/pytorch/tensordict", "entries": { "CPU Benchmark Results": [ @@ -289859,6 +289859,1542 @@ window.BENCHMARK_DATA = { "extra": "mean: 26.31436515384732 msec\nrounds: 39" } ] + }, + { + "commit": { + "author": { + "email": "vmoens@meta.com", + "name": "Vincent Moens", + "username": "vmoens" + }, + "committer": { + "email": "vmoens@meta.com", + "name": "Vincent Moens", + "username": "vmoens" + }, + "distinct": true, + "id": "3d3ea24aaae1eee03a2c9c80bcd1100f915b7962", + "message": "[Refactor] Better handling of params and buffers in bytes\n\nghstack-source-id: 87945c47b376d223bb3dc33bd6ec7cb9bb047455\nPull Request resolved: https://github.com/pytorch/tensordict/pull/1059", + "timestamp": "2024-10-24T13:51:13-07:00", + "tree_id": "09542959f93dbcb98a476f6492449c9660778461", + "url": "https://github.com/pytorch/tensordict/commit/3d3ea24aaae1eee03a2c9c80bcd1100f915b7962" + }, + "date": 1729803572904, + "tool": "pytest", + "benches": [ + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_nested", + "value": 41654.2034536691, + "unit": "iter/sec", + "range": "stddev: 0.0000036835778287046", + "extra": "mean: 24.007180958634205 usec\nrounds: 6239" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_stack_nested", + "value": 41446.0296140977, + "unit": "iter/sec", + "range": "stddev: 0.0000019478415797307658", + "extra": "mean: 24.12776348689994 usec\nrounds: 13661" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_nested_inplace", + "value": 38398.94055407199, + "unit": "iter/sec", + "range": "stddev: 0.0000018326182393670662", + "extra": "mean: 26.04238516924279 usec\nrounds: 24719" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_stack_nested_inplace", + "value": 38423.180552148486, + "unit": "iter/sec", + "range": "stddev: 0.0000014393728076307387", + "extra": "mean: 26.02595583264602 usec\nrounds: 27826" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items", + "value": 238265.05647857246, + "unit": "iter/sec", + "range": "stddev: 3.9938155440174587e-7", + "extra": "mean: 4.197006538765921 usec\nrounds: 70043" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested", + "value": 2636.3089812314092, + "unit": "iter/sec", + "range": "stddev: 0.000007704386420555872", + "extra": "mean: 379.31820857087246 usec\nrounds: 2287" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested_locked", + "value": 2626.677992045757, + "unit": "iter/sec", + "range": "stddev: 0.000019558617784822496", + "extra": "mean: 380.70901839823995 usec\nrounds: 2609" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested_leaf", + "value": 12440.036795869182, + "unit": "iter/sec", + "range": "stddev: 0.0000026204178661768916", + "extra": "mean: 80.38561432005235 usec\nrounds: 9469" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested", + "value": 2617.9438454775373, + "unit": "iter/sec", + "range": "stddev: 0.000006725932693944373", + "extra": "mean: 381.97916342914937 usec\nrounds: 2368" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested_leaf", + "value": 12032.931820686377, + "unit": "iter/sec", + "range": "stddev: 0.000005345349248251496", + "extra": "mean: 83.1052660234352 usec\nrounds: 10048" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested_locked", + "value": 2618.883987883871, + "unit": "iter/sec", + "range": "stddev: 0.000011297684529461147", + "extra": "mean: 381.8420382981634 usec\nrounds: 2585" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys", + "value": 285628.0837888373, + "unit": "iter/sec", + "range": "stddev: 2.8043902903738275e-7", + "extra": "mean: 3.5010562922772417 usec\nrounds: 98717" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested", + "value": 7469.93782078859, + "unit": "iter/sec", + "range": "stddev: 0.000007079218069521883", + "extra": "mean: 133.86992288169162 usec\nrounds: 6289" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested_locked", + "value": 7198.273587827721, + "unit": "iter/sec", + "range": "stddev: 0.000020798410059716038", + "extra": "mean: 138.92219958004927 usec\nrounds: 6674" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested_leaf", + "value": 8543.594847579996, + "unit": "iter/sec", + "range": "stddev: 0.000006338815343332715", + "extra": "mean: 117.04674880308183 usec\nrounds: 7727" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested", + "value": 7491.521675813563, + "unit": "iter/sec", + "range": "stddev: 0.000006714108063815243", + "extra": "mean: 133.48422967639644 usec\nrounds: 6827" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested_leaf", + "value": 8621.302028778859, + "unit": "iter/sec", + "range": "stddev: 0.000006263200597439671", + "extra": "mean: 115.99176048604835 usec\nrounds: 7319" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested_locked", + "value": 7278.410225301533, + "unit": "iter/sec", + "range": "stddev: 0.0000037049261549408652", + "extra": "mean: 137.39264056919404 usec\nrounds: 6744" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values", + "value": 941171.9275748729, + "unit": "iter/sec", + "range": "stddev: 7.390181815444509e-8", + "extra": "mean: 1.0625051286608846 usec\nrounds: 182816" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested", + "value": 10739.982702325202, + "unit": "iter/sec", + "range": "stddev: 0.0000027588480738553584", + "extra": "mean: 93.11001960771318 usec\nrounds: 7701" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested_locked", + "value": 10651.128995830108, + "unit": "iter/sec", + "range": "stddev: 0.000005035147664950523", + "extra": "mean: 93.88676077357599 usec\nrounds: 9932" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested_leaf", + "value": 12661.084537625407, + "unit": "iter/sec", + "range": "stddev: 0.000002358330119980513", + "extra": "mean: 78.98217542330308 usec\nrounds: 10449" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested", + "value": 10513.03003480861, + "unit": "iter/sec", + "range": "stddev: 0.000009904515507525366", + "extra": "mean: 95.1200554634585 usec\nrounds: 9015" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested_leaf", + "value": 12826.347341306722, + "unit": "iter/sec", + "range": "stddev: 0.000004407896992896968", + "extra": "mean: 77.96451892266641 usec\nrounds: 10913" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested_locked", + "value": 10523.298339537927, + "unit": "iter/sec", + "range": "stddev: 0.000002806884607761242", + "extra": "mean: 95.02724029431153 usec\nrounds: 9788" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership", + "value": 1140579.468489921, + "unit": "iter/sec", + "range": "stddev: 1.282231464687489e-7", + "extra": "mean: 876.7473267986821 nsec\nrounds: 152161" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested", + "value": 366345.4519937219, + "unit": "iter/sec", + "range": "stddev: 3.9815054676673896e-7", + "extra": "mean: 2.7296640221894637 usec\nrounds: 144447" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_leaf", + "value": 366752.71926195733, + "unit": "iter/sec", + "range": "stddev: 3.041353798258717e-7", + "extra": "mean: 2.7266328168264744 usec\nrounds: 154250" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested", + "value": 368131.1518331802, + "unit": "iter/sec", + "range": "stddev: 3.0042765393603245e-7", + "extra": "mean: 2.7164231959732468 usec\nrounds: 103328" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_leaf", + "value": 367580.27527689544, + "unit": "iter/sec", + "range": "stddev: 2.7452501404943764e-7", + "extra": "mean: 2.7204941811600407 usec\nrounds: 114456" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_last", + "value": 239572.99947597156, + "unit": "iter/sec", + "range": "stddev: 6.962975202814503e-7", + "extra": "mean: 4.174093083057537 usec\nrounds: 92251" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_leaf_last", + "value": 236719.54375303726, + "unit": "iter/sec", + "range": "stddev: 0.0000018669545850407791", + "extra": "mean: 4.224408277177449 usec\nrounds: 100422" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_last", + "value": 205180.3057771084, + "unit": "iter/sec", + "range": "stddev: 5.128139357420076e-7", + "extra": "mean: 4.873762109928429 usec\nrounds: 48762" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_leaf_last", + "value": 204988.2646752109, + "unit": "iter/sec", + "range": "stddev: 4.4883358647940076e-7", + "extra": "mean: 4.878328042751266 usec\nrounds: 71195" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getleaf", + "value": 93781.98493200367, + "unit": "iter/sec", + "range": "stddev: 8.512540917854487e-7", + "extra": "mean: 10.663028733344115 usec\nrounds: 66543" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_get", + "value": 99297.93207420091, + "unit": "iter/sec", + "range": "stddev: 6.096488233982596e-7", + "extra": "mean: 10.070703176907498 usec\nrounds: 72491" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getleaf", + "value": 95656.37200221319, + "unit": "iter/sec", + "range": "stddev: 6.158459628371963e-7", + "extra": "mean: 10.45408663394492 usec\nrounds: 55544" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_get", + "value": 97615.40154633882, + "unit": "iter/sec", + "range": "stddev: 8.642179628977248e-7", + "extra": "mean: 10.24428506320585 usec\nrounds: 59029" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getitemleaf", + "value": 91416.99709874799, + "unit": "iter/sec", + "range": "stddev: 6.176742736594991e-7", + "extra": "mean: 10.938884799725013 usec\nrounds: 53663" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getitem", + "value": 97717.93954129572, + "unit": "iter/sec", + "range": "stddev: 7.658259585920442e-7", + "extra": "mean: 10.233535466406336 usec\nrounds: 66415" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getitemleaf", + "value": 91683.35433845203, + "unit": "iter/sec", + "range": "stddev: 7.777271254589427e-7", + "extra": "mean: 10.907105299708691 usec\nrounds: 50209" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getitem", + "value": 96919.3188421223, + "unit": "iter/sec", + "range": "stddev: 5.488456680236509e-7", + "extra": "mean: 10.317860380642584 usec\nrounds: 55329" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_lock_nested", + "value": 1947.5863990290827, + "unit": "iter/sec", + "range": "stddev: 0.00011702982114859284", + "extra": "mean: 513.4560399982888 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_lock_stack_nested", + "value": 2076.4972836413835, + "unit": "iter/sec", + "range": "stddev: 0.00003479497727456496", + "extra": "mean: 481.5802110014715 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unlock_nested", + "value": 2321.504978363557, + "unit": "iter/sec", + "range": "stddev: 0.00006926583698948252", + "extra": "mean: 430.75505300225814 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unlock_stack_nested", + "value": 2511.0822723997353, + "unit": "iter/sec", + "range": "stddev: 0.00002443056444907995", + "extra": "mean: 398.2346619986856 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_flatten_speed", + "value": 9942.908121693497, + "unit": "iter/sec", + "range": "stddev: 0.00000590945226571644", + "extra": "mean: 100.5741969814841 usec\nrounds: 7620" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unflatten_speed", + "value": 1929.9663681453146, + "unit": "iter/sec", + "range": "stddev: 0.00002629300472136902", + "extra": "mean: 518.1437441114549 usec\nrounds: 1868" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_common_ops", + "value": 878.8307583394004, + "unit": "iter/sec", + "range": "stddev: 0.00009495256385188975", + "extra": "mean: 1.1378755130163576 msec\nrounds: 653" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation", + "value": 487164.4330980753, + "unit": "iter/sec", + "range": "stddev: 2.7754049447150796e-7", + "extra": "mean: 2.0526950082143647 usec\nrounds: 104734" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_empty", + "value": 57814.48261389039, + "unit": "iter/sec", + "range": "stddev: 0.0000011074244267804492", + "extra": "mean: 17.296704126514868 usec\nrounds: 16115" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_nested_1", + "value": 47320.51005989173, + "unit": "iter/sec", + "range": "stddev: 0.0000014645214708993572", + "extra": "mean: 21.132485654409454 usec\nrounds: 17218" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_nested_2", + "value": 40942.75165899735, + "unit": "iter/sec", + "range": "stddev: 0.0000026729511440681826", + "extra": "mean: 24.424347643479535 usec\nrounds: 15490" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_clone", + "value": 57171.52090954361, + "unit": "iter/sec", + "range": "stddev: 0.000001593665123111733", + "extra": "mean: 17.491226122568843 usec\nrounds: 12449" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[int]", + "value": 59349.292266476055, + "unit": "iter/sec", + "range": "stddev: 0.000016178905876470404", + "extra": "mean: 16.849400587795355 usec\nrounds: 9526" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[slice_int]", + "value": 33009.880639111085, + "unit": "iter/sec", + "range": "stddev: 0.000007479972537611556", + "extra": "mean: 30.293959888336293 usec\nrounds: 10371" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[range]", + "value": 17266.30857368423, + "unit": "iter/sec", + "range": "stddev: 0.000008558541562972554", + "extra": "mean: 57.91625903895353 usec\nrounds: 4868" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[tuple]", + "value": 40187.24933932979, + "unit": "iter/sec", + "range": "stddev: 0.000007090748755196727", + "extra": "mean: 24.88351445893403 usec\nrounds: 16149" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[list]", + "value": 18866.712796770218, + "unit": "iter/sec", + "range": "stddev: 0.000008593384006084584", + "extra": "mean: 53.0034039724816 usec\nrounds: 9414" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[int]", + "value": 30003.125424834907, + "unit": "iter/sec", + "range": "stddev: 0.000002302312088770226", + "extra": "mean: 33.3298610008228 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[slice_int]", + "value": 15793.65775668374, + "unit": "iter/sec", + "range": "stddev: 0.0000044481970629170374", + "extra": "mean: 63.316554999858006 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[range]", + "value": 11768.639654072753, + "unit": "iter/sec", + "range": "stddev: 0.0000071578208791177405", + "extra": "mean: 84.97158799946192 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[tuple]", + "value": 20132.569750669423, + "unit": "iter/sec", + "range": "stddev: 0.0000031517975432010858", + "extra": "mean: 49.670757999820125 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem", + "value": 32532.71932876328, + "unit": "iter/sec", + "range": "stddev: 0.000002591530229432144", + "extra": "mean: 30.73828504449261 usec\nrounds: 14296" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set", + "value": 33815.49348682192, + "unit": "iter/sec", + "range": "stddev: 0.0000023174034691863623", + "extra": "mean: 29.572243279244336 usec\nrounds: 14247" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_shared", + "value": 4558.798427125337, + "unit": "iter/sec", + "range": "stddev: 0.0001245042965114833", + "extra": "mean: 219.3560465516293 usec\nrounds: 2943" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update", + "value": 26656.477429213497, + "unit": "iter/sec", + "range": "stddev: 0.0000070438026101002075", + "extra": "mean: 37.5143340921736 usec\nrounds: 17576" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update_nested", + "value": 20670.34209949397, + "unit": "iter/sec", + "range": "stddev: 0.000003136516896872372", + "extra": "mean: 48.37849297252226 usec\nrounds: 12735" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update__nested", + "value": 21946.673329265122, + "unit": "iter/sec", + "range": "stddev: 0.00000472806734274293", + "extra": "mean: 45.56499224265278 usec\nrounds: 7735" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_nested", + "value": 30571.279392446315, + "unit": "iter/sec", + "range": "stddev: 0.0000023591576772884934", + "extra": "mean: 32.71043999051882 usec\nrounds: 13498" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_nested_new", + "value": 26139.351257398812, + "unit": "iter/sec", + "range": "stddev: 0.0000028217959154089042", + "extra": "mean: 38.256496504171935 usec\nrounds: 12300" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_select", + "value": 17837.048227812258, + "unit": "iter/sec", + "range": "stddev: 0.0000038075399218468558", + "extra": "mean: 56.06308775017825 usec\nrounds: 9094" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_select_nested", + "value": 15502.075405302878, + "unit": "iter/sec", + "range": "stddev: 0.000004689979597686071", + "extra": "mean: 64.5074916651434 usec\nrounds: 11038" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_exclude_nested", + "value": 12609.63719622395, + "unit": "iter/sec", + "range": "stddev: 0.0000030402701083923308", + "extra": "mean: 79.30442283458063 usec\nrounds: 9538" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_empty[True]", + "value": 2808.8581012961827, + "unit": "iter/sec", + "range": "stddev: 0.000019552594731517606", + "extra": "mean: 356.0165604444516 usec\nrounds: 2523" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_empty[False]", + "value": 698796.7492405836, + "unit": "iter/sec", + "range": "stddev: 1.008571807505031e-7", + "extra": "mean: 1.4310312706616748 usec\nrounds: 163372" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed", + "value": 3195.445489894306, + "unit": "iter/sec", + "range": "stddev: 0.00004321590526237778", + "extra": "mean: 312.94541032307717 usec\nrounds: 775" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed_stack0", + "value": 3224.6241339929866, + "unit": "iter/sec", + "range": "stddev: 0.00001022291611533358", + "extra": "mean: 310.11366238263565 usec\nrounds: 2879" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed_stack1", + "value": 1286.5736929925342, + "unit": "iter/sec", + "range": "stddev: 0.0026781050539393314", + "extra": "mean: 777.258236700012 usec\nrounds: 1297" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_split", + "value": 492.36425056129895, + "unit": "iter/sec", + "range": "stddev: 0.00010002467832164815", + "extra": "mean: 2.0310166687772164 msec\nrounds: 474" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_chunk", + "value": 413.14590487493643, + "unit": "iter/sec", + "range": "stddev: 0.006262061929639141", + "extra": "mean: 2.420452407249953 msec\nrounds: 469" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_creation[device0]", + "value": 8678.66283078572, + "unit": "iter/sec", + "range": "stddev: 0.00000958516724079596", + "extra": "mean: 115.2251239041931 usec\nrounds: 2736" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_creation_from_tensor", + "value": 8566.071164592431, + "unit": "iter/sec", + "range": "stddev: 0.000052308854476025283", + "extra": "mean: 116.73963253229398 usec\nrounds: 5244" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_add_one[memmap_tensor0]", + "value": 139428.41099482466, + "unit": "iter/sec", + "range": "stddev: 0.0000021616380965902434", + "extra": "mean: 7.172139400176613 usec\nrounds: 21406" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_contiguous[memmap_tensor0]", + "value": 534162.5876115761, + "unit": "iter/sec", + "range": "stddev: 2.4165703637409814e-7", + "extra": "mean: 1.8720891788235163 usec\nrounds: 57794" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_stack[memmap_tensor0]", + "value": 176413.0817166751, + "unit": "iter/sec", + "range": "stddev: 8.316455295994036e-7", + "extra": "mean: 5.668513866823273 usec\nrounds: 20120" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index", + "value": 2425.2373561370878, + "unit": "iter/sec", + "range": "stddev: 0.000056247788355489605", + "extra": "mean: 412.33077557109607 usec\nrounds: 1359" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index_astensor", + "value": 1952.2456979317367, + "unit": "iter/sec", + "range": "stddev: 0.000059467206599069", + "extra": "mean: 512.2306075815292 usec\nrounds: 1794" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index_op", + "value": 962.1468593293856, + "unit": "iter/sec", + "range": "stddev: 0.00006530687680885786", + "extra": "mean: 1.0393423730520701 msec\nrounds: 898" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_model", + "value": 8.262243430212274, + "unit": "iter/sec", + "range": "stddev: 0.003627939863651652", + "extra": "mean: 121.0325026666889 msec\nrounds: 6" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_model_pickle", + "value": 2.541957331649866, + "unit": "iter/sec", + "range": "stddev: 0.06361997282823152", + "extra": "mean: 393.39763399999583 msec\nrounds: 5" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights", + "value": 8.577138032632309, + "unit": "iter/sec", + "range": "stddev: 0.00407636398786374", + "extra": "mean: 116.58900628571344 msec\nrounds: 7" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights_returnearly", + "value": 6.256709330111014, + "unit": "iter/sec", + "range": "stddev: 0.007175081562529295", + "extra": "mean: 159.82842533333042 msec\nrounds: 6" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights_pickle", + "value": 1.168687561626759, + "unit": "iter/sec", + "range": "stddev: 0.29760328547051246", + "extra": "mean: 855.6606853999938 msec\nrounds: 5" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights_filesystem", + "value": 6.3232002310745665, + "unit": "iter/sec", + "range": "stddev: 0.034710577048605654", + "extra": "mean: 158.14776750001158 msec\nrounds: 8" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_model_filesystem", + "value": 6.73833195876762, + "unit": "iter/sec", + "range": "stddev: 0.009622483455616029", + "extra": "mean: 148.40468028572624 msec\nrounds: 7" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_reshape_pytree", + "value": 26616.678708486455, + "unit": "iter/sec", + "range": "stddev: 0.000002012465038774538", + "extra": "mean: 37.57042758611202 usec\nrounds: 6525" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_reshape_td", + "value": 22111.673405324294, + "unit": "iter/sec", + "range": "stddev: 0.00000308026624061731", + "extra": "mean: 45.2249805643027 usec\nrounds: 6483" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_view_pytree", + "value": 26415.395372124854, + "unit": "iter/sec", + "range": "stddev: 0.0000038016435496798876", + "extra": "mean: 37.856711433335626 usec\nrounds: 12621" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_view_td", + "value": 19352.000245379553, + "unit": "iter/sec", + "range": "stddev: 0.0000033764447297391984", + "extra": "mean: 51.67424490079562 usec\nrounds: 8187" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_unbind_pytree", + "value": 28540.6394192264, + "unit": "iter/sec", + "range": "stddev: 0.0000021170868964436415", + "extra": "mean: 35.037757399588955 usec\nrounds: 10812" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_unbind_td", + "value": 21914.420186246494, + "unit": "iter/sec", + "range": "stddev: 0.0000038459947998005406", + "extra": "mean: 45.63205375735201 usec\nrounds: 9822" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_split_pytree", + "value": 26924.407804295337, + "unit": "iter/sec", + "range": "stddev: 0.000002550737335379523", + "extra": "mean: 37.141021160750164 usec\nrounds: 7892" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_split_td", + "value": 17168.364741959493, + "unit": "iter/sec", + "range": "stddev: 0.000015038665116272309", + "extra": "mean: 58.246665598616936 usec\nrounds: 9049" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_add_pytree", + "value": 22474.959533879344, + "unit": "iter/sec", + "range": "stddev: 0.0000031845345267278633", + "extra": "mean: 44.49396220236009 usec\nrounds: 7699" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_add_td", + "value": 11654.011835454978, + "unit": "iter/sec", + "range": "stddev: 0.00000740156641114573", + "extra": "mean: 85.80736094309617 usec\nrounds: 6278" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[tensordict-compile]", + "value": 13890.40222680377, + "unit": "iter/sec", + "range": "stddev: 0.000005019505455536484", + "extra": "mean: 71.99215571096558 usec\nrounds: 10558" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[tensordict-eager]", + "value": 4840.404435115292, + "unit": "iter/sec", + "range": "stddev: 0.000009110488284267783", + "extra": "mean: 206.5943070263676 usec\nrounds: 4270" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[pytree-compile]", + "value": 18555.01573876502, + "unit": "iter/sec", + "range": "stddev: 0.000005412101610558958", + "extra": "mean: 53.893783442651916 usec\nrounds: 13553" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[pytree-eager]", + "value": 6841.939070041137, + "unit": "iter/sec", + "range": "stddev: 0.000012972687900116382", + "extra": "mean: 146.15739628239447 usec\nrounds: 6132" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[tensordict-compile]", + "value": 36423.238696505745, + "unit": "iter/sec", + "range": "stddev: 0.000003123009893853218", + "extra": "mean: 27.45499949448303 usec\nrounds: 23726" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[tensordict-eager]", + "value": 12565.035947311186, + "unit": "iter/sec", + "range": "stddev: 0.0000057111488830348315", + "extra": "mean: 79.58592432152905 usec\nrounds: 11311" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[pytree-compile]", + "value": 12446.639332782854, + "unit": "iter/sec", + "range": "stddev: 0.000011210572351612176", + "extra": "mean: 80.34297236894525 usec\nrounds: 11002" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[pytree-eager]", + "value": 14925.757845462218, + "unit": "iter/sec", + "range": "stddev: 0.000004389349258826117", + "extra": "mean: 66.99827307623268 usec\nrounds: 13553" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensordict-compile]", + "value": 8336.320135091562, + "unit": "iter/sec", + "range": "stddev: 0.000005336029223098788", + "extra": "mean: 119.95700546461995 usec\nrounds: 7137" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensordict-eager]", + "value": 4088.5701194830453, + "unit": "iter/sec", + "range": "stddev: 0.000016143348758187448", + "extra": "mean: 244.58428515992753 usec\nrounds: 3868" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensorclass-compile]", + "value": 19103.876367986824, + "unit": "iter/sec", + "range": "stddev: 0.0000035869663134267587", + "extra": "mean: 52.345397381012276 usec\nrounds: 13438" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensorclass-eager]", + "value": 12502.650597430158, + "unit": "iter/sec", + "range": "stddev: 0.000006810847090501958", + "extra": "mean: 79.98303977282575 usec\nrounds: 9328" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[pytree-compile]", + "value": 9075.85541227526, + "unit": "iter/sec", + "range": "stddev: 0.000004804798446694358", + "extra": "mean: 110.1824516339784 usec\nrounds: 7743" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[pytree-eager]", + "value": 3362.9824202497166, + "unit": "iter/sec", + "range": "stddev: 0.000018432978082227046", + "extra": "mean: 297.35510776941425 usec\nrounds: 3192" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensordict-eager]", + "value": 3586.4530450874636, + "unit": "iter/sec", + "range": "stddev: 0.000008633911589132481", + "extra": "mean: 278.8270158366489 usec\nrounds: 3031" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensordict-compile]", + "value": 8167.087438654323, + "unit": "iter/sec", + "range": "stddev: 0.000005072531848302654", + "extra": "mean: 122.44267096579148 usec\nrounds: 6978" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensorclass-eager]", + "value": 13204.34636049301, + "unit": "iter/sec", + "range": "stddev: 0.000004806261548105116", + "extra": "mean: 75.73263929155696 usec\nrounds: 10557" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensorclass-compile]", + "value": 18637.588068104, + "unit": "iter/sec", + "range": "stddev: 0.000004529234277581233", + "extra": "mean: 53.65501138590891 usec\nrounds: 13174" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[pytree-eager]", + "value": 4120.851479386407, + "unit": "iter/sec", + "range": "stddev: 0.000007965478728603736", + "extra": "mean: 242.6682944052377 usec\nrounds: 3879" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[pytree-compile]", + "value": 9004.788035628395, + "unit": "iter/sec", + "range": "stddev: 0.00000482944072106881", + "extra": "mean: 111.05203099100106 usec\nrounds: 7712" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[tensordict-compile]", + "value": 34165.565717527854, + "unit": "iter/sec", + "range": "stddev: 0.000002847889782090789", + "extra": "mean: 29.269235822633345 usec\nrounds: 23823" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[tensordict-eager]", + "value": 12638.421073784757, + "unit": "iter/sec", + "range": "stddev: 0.0000033592244915825007", + "extra": "mean: 79.12380780493616 usec\nrounds: 11634" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[pytree-compile]", + "value": 12237.56272044264, + "unit": "iter/sec", + "range": "stddev: 0.00000619338251242377", + "extra": "mean: 81.7156179579384 usec\nrounds: 10881" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[pytree-eager]", + "value": 14472.64447240224, + "unit": "iter/sec", + "range": "stddev: 0.000002722138254028507", + "extra": "mean: 69.09587269326565 usec\nrounds: 12898" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[tensordict-compile]", + "value": 4662.3703067220495, + "unit": "iter/sec", + "range": "stddev: 0.000015763862848515363", + "extra": "mean: 214.48317791451132 usec\nrounds: 3586" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[tensordict-eager]", + "value": 553.687675914073, + "unit": "iter/sec", + "range": "stddev: 0.00013643235292545592", + "extra": "mean: 1.8060723463803272 msec\nrounds: 511" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[pytree-compile]", + "value": 4693.549682330922, + "unit": "iter/sec", + "range": "stddev: 0.00000905000090064654", + "extra": "mean: 213.05836044828604 usec\nrounds: 3393" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[pytree-eager]", + "value": 875.3289775083019, + "unit": "iter/sec", + "range": "stddev: 0.000022742682521870062", + "extra": "mean: 1.1424276194381051 msec\nrounds: 854" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add_stack[compile]", + "value": 2156.3327247975403, + "unit": "iter/sec", + "range": "stddev: 0.00001673409437050547", + "extra": "mean: 463.750324103573 usec\nrounds: 1842" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add_stack[eager]", + "value": 224.14270038238104, + "unit": "iter/sec", + "range": "stddev: 0.0005983813834822742", + "extra": "mean: 4.461443528136444 msec\nrounds: 231" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensordict-compile]", + "value": 23175.671851990643, + "unit": "iter/sec", + "range": "stddev: 0.000004954136693214785", + "extra": "mean: 43.14869516562069 usec\nrounds: 15205" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensordict-eager]", + "value": 19785.762576196965, + "unit": "iter/sec", + "range": "stddev: 0.000014014914957622972", + "extra": "mean: 50.54139289041296 usec\nrounds: 14994" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensorclass-compile]", + "value": 27182.549167997036, + "unit": "iter/sec", + "range": "stddev: 0.0000030721266758730285", + "extra": "mean: 36.78830833045397 usec\nrounds: 17478" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensorclass-eager]", + "value": 34666.27680790099, + "unit": "iter/sec", + "range": "stddev: 0.0000030291789970510258", + "extra": "mean: 28.84647825151169 usec\nrounds: 22162" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-pytree-compile]", + "value": 26373.384597460103, + "unit": "iter/sec", + "range": "stddev: 0.0000029646099866070647", + "extra": "mean: 37.917014265067266 usec\nrounds: 16123" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-pytree-eager]", + "value": 34690.7216182043, + "unit": "iter/sec", + "range": "stddev: 0.000003950771007697493", + "extra": "mean: 28.826151586170525 usec\nrounds: 22634" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensordict-compile]", + "value": 13390.036805958305, + "unit": "iter/sec", + "range": "stddev: 0.000004832643154348451", + "extra": "mean: 74.68239367012193 usec\nrounds: 10016" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensordict-eager]", + "value": 32854.20196836337, + "unit": "iter/sec", + "range": "stddev: 0.00001346673120145911", + "extra": "mean: 30.437506927209494 usec\nrounds: 23530" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensorclass-compile]", + "value": 14530.984205539835, + "unit": "iter/sec", + "range": "stddev: 0.000005520369065703354", + "extra": "mean: 68.81846307552637 usec\nrounds: 10711" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensorclass-eager]", + "value": 43045.78447669752, + "unit": "iter/sec", + "range": "stddev: 0.0000016698695874884322", + "extra": "mean: 23.231078540137695 usec\nrounds: 28062" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-pytree-compile]", + "value": 14365.248472174118, + "unit": "iter/sec", + "range": "stddev: 0.000004434889931074087", + "extra": "mean: 69.61244018417275 usec\nrounds: 10432" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-pytree-eager]", + "value": 43210.045994631175, + "unit": "iter/sec", + "range": "stddev: 0.000002472136226103331", + "extra": "mean: 23.14276638641508 usec\nrounds: 27828" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensordict-compile]", + "value": 13177.749164134588, + "unit": "iter/sec", + "range": "stddev: 0.000004885632033217798", + "extra": "mean: 75.88549361082578 usec\nrounds: 9469" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensordict-eager]", + "value": 34223.68139887135, + "unit": "iter/sec", + "range": "stddev: 0.000013039075507716865", + "extra": "mean: 29.21953335017251 usec\nrounds: 24093" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensorclass-compile]", + "value": 14384.164356536994, + "unit": "iter/sec", + "range": "stddev: 0.000005828543570731501", + "extra": "mean: 69.52089639781836 usec\nrounds: 10299" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensorclass-eager]", + "value": 43607.04343595484, + "unit": "iter/sec", + "range": "stddev: 0.0000028132275688106496", + "extra": "mean: 22.932075215525412 usec\nrounds: 30047" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-pytree-compile]", + "value": 14373.451575271118, + "unit": "iter/sec", + "range": "stddev: 0.000004488813246035914", + "extra": "mean: 69.57271152048513 usec\nrounds: 10746" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-pytree-eager]", + "value": 43195.51333791699, + "unit": "iter/sec", + "range": "stddev: 0.0000022989597910807784", + "extra": "mean: 23.150552516346664 usec\nrounds: 29505" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[eager]", + "value": 37720.009441899485, + "unit": "iter/sec", + "range": "stddev: 0.0000030425968604290268", + "extra": "mean: 26.51112804041871 usec\nrounds: 23063" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[compile]", + "value": 22256.473322234997, + "unit": "iter/sec", + "range": "stddev: 0.0000032947332978166833", + "extra": "mean: 44.93074826014618 usec\nrounds: 9915" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[compile-overhead]", + "value": 22535.429595803053, + "unit": "iter/sec", + "range": "stddev: 0.0000032556145867581665", + "extra": "mean: 44.37457008524203 usec\nrounds: 10316" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[eager]", + "value": 4580.441935611221, + "unit": "iter/sec", + "range": "stddev: 0.000013295668974365824", + "extra": "mean: 218.31954515684927 usec\nrounds: 3975" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[compile]", + "value": 4925.641074604453, + "unit": "iter/sec", + "range": "stddev: 0.00002901400244814789", + "extra": "mean: 203.01925878354905 usec\nrounds: 3586" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[compile-overhead]", + "value": 4982.272419640845, + "unit": "iter/sec", + "range": "stddev: 0.000012941464347772739", + "extra": "mean: 200.71162629683874 usec\nrounds: 3856" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[eager]", + "value": 89.36606311637298, + "unit": "iter/sec", + "range": "stddev: 0.00024023639764272374", + "extra": "mean: 11.189930104650514 msec\nrounds: 86" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[compile]", + "value": 93.0597919433891, + "unit": "iter/sec", + "range": "stddev: 0.0002538524978655385", + "extra": "mean: 10.74577945121915 msec\nrounds: 82" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[compile-overhead]", + "value": 93.80110599281518, + "unit": "iter/sec", + "range": "stddev: 0.00030921151327875066", + "extra": "mean: 10.660855108431198 msec\nrounds: 83" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[eager]", + "value": 10788.448864861508, + "unit": "iter/sec", + "range": "stddev: 0.000006011684517299101", + "extra": "mean: 92.69173099175059 usec\nrounds: 8628" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[compile]", + "value": 17119.171612342372, + "unit": "iter/sec", + "range": "stddev: 0.000004934308059004971", + "extra": "mean: 58.41404144106086 usec\nrounds: 12572" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[compile-overhead]", + "value": 17518.53254488733, + "unit": "iter/sec", + "range": "stddev: 0.0000034425231540846957", + "extra": "mean: 57.08240672771668 usec\nrounds: 12812" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[eager]", + "value": 2557.44131447697, + "unit": "iter/sec", + "range": "stddev: 0.000017371327625641044", + "extra": "mean: 391.01581504110214 usec\nrounds: 2287" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[compile]", + "value": 4430.626751387205, + "unit": "iter/sec", + "range": "stddev: 0.000013454288426284746", + "extra": "mean: 225.7017022900666 usec\nrounds: 3930" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[compile-overhead]", + "value": 4473.90271110049, + "unit": "iter/sec", + "range": "stddev: 0.000014834730089317433", + "extra": "mean: 223.518494829768 usec\nrounds: 3965" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-eager]", + "value": 1818.412146534423, + "unit": "iter/sec", + "range": "stddev: 0.00001953292708787759", + "extra": "mean: 549.9303344986041 usec\nrounds: 1716" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-compile]", + "value": 2363.839624925823, + "unit": "iter/sec", + "range": "stddev: 0.00002291182936231767", + "extra": "mean: 423.0405436372951 usec\nrounds: 2200" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-compile-overhead]", + "value": 2349.5175166397876, + "unit": "iter/sec", + "range": "stddev: 0.000011724096400257175", + "extra": "mean: 425.61929967228815 usec\nrounds: 2139" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-eager]", + "value": 1283.8316298015632, + "unit": "iter/sec", + "range": "stddev: 0.000019316018030500384", + "extra": "mean: 778.9183385009497 usec\nrounds: 1161" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-compile]", + "value": 2145.7117974117677, + "unit": "iter/sec", + "range": "stddev: 0.000025213438297624273", + "extra": "mean: 466.04581342482004 usec\nrounds: 2026" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-compile-overhead]", + "value": 2143.554223669972, + "unit": "iter/sec", + "range": "stddev: 0.000011809410564535238", + "extra": "mean: 466.51490732429585 usec\nrounds: 2007" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-eager]", + "value": 1827.0099497867097, + "unit": "iter/sec", + "range": "stddev: 0.000025456092625066157", + "extra": "mean: 547.3423941214676 usec\nrounds: 1667" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-compile]", + "value": 2344.2073788420666, + "unit": "iter/sec", + "range": "stddev: 0.000012118177199651004", + "extra": "mean: 426.5834196349792 usec\nrounds: 2190" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-compile-overhead]", + "value": 2347.1757899453423, + "unit": "iter/sec", + "range": "stddev: 0.000012547733244911185", + "extra": "mean: 426.04393087374444 usec\nrounds: 2141" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-eager]", + "value": 1084.7472135172404, + "unit": "iter/sec", + "range": "stddev: 0.000031100797363225696", + "extra": "mean: 921.8737670295998 usec\nrounds: 1013" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-compile]", + "value": 2029.9325513018623, + "unit": "iter/sec", + "range": "stddev: 0.000026511340458947222", + "extra": "mean: 492.62720545008614 usec\nrounds: 1908" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-compile-overhead]", + "value": 2042.2735687861934, + "unit": "iter/sec", + "range": "stddev: 0.00001249699269551539", + "extra": "mean: 489.65036579028964 usec\nrounds: 1941" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[eager]", + "value": 514.2585133300557, + "unit": "iter/sec", + "range": "stddev: 0.00012828778352196667", + "extra": "mean: 1.9445472930035699 msec\nrounds: 529" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[compile]", + "value": 1954.0327290878042, + "unit": "iter/sec", + "range": "stddev: 0.000012695867835271859", + "extra": "mean: 511.762154806295 usec\nrounds: 1841" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[compile-overhead]", + "value": 1937.659535726377, + "unit": "iter/sec", + "range": "stddev: 0.000015184204683991577", + "extra": "mean: 516.0865371661522 usec\nrounds: 1843" + }, + { + "name": "benchmarks/distributed/distributed_benchmark_test.py::test_distributed", + "value": 7782.107506007178, + "unit": "iter/sec", + "range": "stddev: 0.00000995659874060583", + "extra": "mean: 128.49989533401822 usec\nrounds: 3621" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdmodule", + "value": 55025.25576825242, + "unit": "iter/sec", + "range": "stddev: 0.0000021737319594864612", + "extra": "mean: 18.173472999592377 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdmodule_dispatch", + "value": 27804.20181007207, + "unit": "iter/sec", + "range": "stddev: 0.0000028350130042348125", + "extra": "mean: 35.96578699978181 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdseq", + "value": 44953.841394446, + "unit": "iter/sec", + "range": "stddev: 0.000005337154358958546", + "extra": "mean: 22.245040000598237 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdseq_dispatch", + "value": 24622.475887257755, + "unit": "iter/sec", + "range": "stddev: 0.0000034225729888098638", + "extra": "mean: 40.61330000195085 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_instantiation_functorch", + "value": 660.2203607634555, + "unit": "iter/sec", + "range": "stddev: 0.000046489009936964705", + "extra": "mean: 1.5146458052939102 msec\nrounds: 529" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_functorch", + "value": 5645.102295581224, + "unit": "iter/sec", + "range": "stddev: 0.000008045856666352695", + "extra": "mean: 177.14470839310792 usec\nrounds: 3347" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_functional_call", + "value": 5864.375462037887, + "unit": "iter/sec", + "range": "stddev: 0.000006763235103842066", + "extra": "mean: 170.52114184593788 usec\nrounds: 3814" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_td_decorator", + "value": 4300.713643313884, + "unit": "iter/sec", + "range": "stddev: 0.000012130932474403403", + "extra": "mean: 232.5195497623174 usec\nrounds: 2954" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[True-True]", + "value": 1536.196797707401, + "unit": "iter/sec", + "range": "stddev: 0.00001904370654131978", + "extra": "mean: 650.9582636107472 usec\nrounds: 1396" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[True-False]", + "value": 1535.48446695586, + "unit": "iter/sec", + "range": "stddev: 0.00002750275982030681", + "extra": "mean: 651.2602514192329 usec\nrounds: 1412" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[False-True]", + "value": 1851.7281129425282, + "unit": "iter/sec", + "range": "stddev: 0.000016592602247459473", + "extra": "mean: 540.0360846771011 usec\nrounds: 1736" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[False-False]", + "value": 1839.7362572783181, + "unit": "iter/sec", + "range": "stddev: 0.00002339857006603758", + "extra": "mean: 543.5561733611681 usec\nrounds: 1742" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_to_module_speed[True]", + "value": 711.9418797125674, + "unit": "iter/sec", + "range": "stddev: 0.00010446886743723881", + "extra": "mean: 1.4046090397206727 msec\nrounds: 428" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_to_module_speed[False]", + "value": 741.6009505175946, + "unit": "iter/sec", + "range": "stddev: 0.000021235797475449767", + "extra": "mean: 1.3484340861511273 msec\nrounds: 592" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_init", + "value": 21753.791868044642, + "unit": "iter/sec", + "range": "stddev: 0.0000027443335434393695", + "extra": "mean: 45.96899731623137 usec\nrounds: 4471" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_init_nested", + "value": 11017.392876565382, + "unit": "iter/sec", + "range": "stddev: 0.000004463802273153757", + "extra": "mean: 90.76557505061443 usec\nrounds: 5523" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_first_layer_tensor", + "value": 669142.4410639683, + "unit": "iter/sec", + "range": "stddev: 1.9743381633137987e-7", + "extra": "mean: 1.4944501179897547 usec\nrounds: 146135" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_first_layer_nontensor", + "value": 217522.93644421833, + "unit": "iter/sec", + "range": "stddev: 4.0956013879454485e-7", + "extra": "mean: 4.597216350361473 usec\nrounds: 78285" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_second_layer_tensor", + "value": 364773.6550714562, + "unit": "iter/sec", + "range": "stddev: 2.933537853848371e-7", + "extra": "mean: 2.7414260489949807 usec\nrounds: 140174" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_second_layer_nontensor", + "value": 168130.88533419953, + "unit": "iter/sec", + "range": "stddev: 3.946594901016445e-7", + "extra": "mean: 5.94774718524955 usec\nrounds: 80822" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_unbind", + "value": 68.48376830446931, + "unit": "iter/sec", + "range": "stddev: 0.04758862478703575", + "extra": "mean: 14.60200022221527 msec\nrounds: 54" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_full_like", + "value": 133.69606911935213, + "unit": "iter/sec", + "range": "stddev: 0.0003373827034953891", + "extra": "mean: 7.479651470585029 msec\nrounds: 102" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_zeros_like", + "value": 139.59144128801873, + "unit": "iter/sec", + "range": "stddev: 0.00044226858751988994", + "extra": "mean: 7.163762984126671 msec\nrounds: 126" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_ones_like", + "value": 132.92405415018163, + "unit": "iter/sec", + "range": "stddev: 0.0006063188436258514", + "extra": "mean: 7.523092839691525 msec\nrounds: 131" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_clone", + "value": 109.52439426083276, + "unit": "iter/sec", + "range": "stddev: 0.0008117206788351543", + "extra": "mean: 9.13038603636096 msec\nrounds: 110" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_squeeze", + "value": 80390.60055494489, + "unit": "iter/sec", + "range": "stddev: 0.000001098666677728932", + "extra": "mean: 12.439265201365501 usec\nrounds: 16199" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_unsqueeze", + "value": 10432.634761249086, + "unit": "iter/sec", + "range": "stddev: 0.0000048566902650035755", + "extra": "mean: 95.85306328506715 usec\nrounds: 3587" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_split", + "value": 5024.508009426249, + "unit": "iter/sec", + "range": "stddev: 0.000024172367150002907", + "extra": "mean: 199.02446132515774 usec\nrounds: 2883" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_permute", + "value": 4396.063584525032, + "unit": "iter/sec", + "range": "stddev: 0.000008584849087612763", + "extra": "mean: 227.47623658588276 usec\nrounds: 2050" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_stack", + "value": 40.1305810417131, + "unit": "iter/sec", + "range": "stddev: 0.0007979353904125368", + "extra": "mean: 24.918652410254555 msec\nrounds: 39" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_cat", + "value": 40.80178669707416, + "unit": "iter/sec", + "range": "stddev: 0.00047830374338209006", + "extra": "mean: 24.508730645163357 msec\nrounds: 31" + } + ] } ], "GPU Benchmark Results": [