diff --git a/dev/bench/data.js b/dev/bench/data.js index 8f561003b..5fc44f4f5 100644 --- a/dev/bench/data.js +++ b/dev/bench/data.js @@ -1,5 +1,5 @@ window.BENCHMARK_DATA = { - "lastUpdate": 1737380974936, + "lastUpdate": 1737452017937, "repoUrl": "https://github.com/pytorch/tensordict", "entries": { "CPU Benchmark Results": [ @@ -810733,6 +810733,1633 @@ window.BENCHMARK_DATA = { "extra": "mean: 50.17030971413035 msec\nrounds: 21" } ] + }, + { + "commit": { + "author": { + "email": "vmoens@meta.com", + "name": "Vincent Moens", + "username": "vmoens" + }, + "committer": { + "email": "vmoens@meta.com", + "name": "Vincent Moens", + "username": "vmoens" + }, + "distinct": true, + "id": "259c94116773eddf995639fa6fda7d7400e14b10", + "message": "[BugFix] Fix mem leak when locking\n\nghstack-source-id: d6e44e1d9b9afc9903a0f45945c10a94dcf5a0ca\nPull Request resolved: https://github.com/pytorch/tensordict/pull/1188", + "timestamp": "2025-01-21T09:21:55Z", + "tree_id": "d6e44e1d9b9afc9903a0f45945c10a94dcf5a0ca", + "url": "https://github.com/pytorch/tensordict/commit/259c94116773eddf995639fa6fda7d7400e14b10" + }, + "date": 1737451979648, + "tool": "pytest", + "benches": [ + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_nested", + "value": 89934.69112042864, + "unit": "iter/sec", + "range": "stddev: 5.754609914799355e-7", + "extra": "mean: 11.119179790821011 usec\nrounds: 6263" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_stack_nested", + "value": 88300.9434135681, + "unit": "iter/sec", + "range": "stddev: 6.22586578296445e-7", + "extra": "mean: 11.324907315161736 usec\nrounds: 22712" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_nested_inplace", + "value": 81289.20024506551, + "unit": "iter/sec", + "range": "stddev: 6.876163989958702e-7", + "extra": "mean: 12.301757145909464 usec\nrounds: 46534" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_plain_set_stack_nested_inplace", + "value": 81782.04530519251, + "unit": "iter/sec", + "range": "stddev: 7.058195050649355e-7", + "extra": "mean: 12.227622777936418 usec\nrounds: 51230" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items", + "value": 339974.0332512815, + "unit": "iter/sec", + "range": "stddev: 2.774511427669604e-7", + "extra": "mean: 2.9414011135988156 usec\nrounds: 126423" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested", + "value": 2755.674859871413, + "unit": "iter/sec", + "range": "stddev: 0.000004619522293946117", + "extra": "mean: 362.8875142572744 usec\nrounds: 2491" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested_locked", + "value": 2756.6469659400664, + "unit": "iter/sec", + "range": "stddev: 0.000004669879271092771", + "extra": "mean: 362.7595453300935 usec\nrounds: 2681" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_nested_leaf", + "value": 17137.375173254473, + "unit": "iter/sec", + "range": "stddev: 0.000001601397184600744", + "extra": "mean: 58.351993224764946 usec\nrounds: 12689" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested", + "value": 2771.899089959161, + "unit": "iter/sec", + "range": "stddev: 0.000005056551968178958", + "extra": "mean: 360.76349374418726 usec\nrounds: 2560" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested_leaf", + "value": 16930.71278168528, + "unit": "iter/sec", + "range": "stddev: 0.0000017554932067175317", + "extra": "mean: 59.06425871695995 usec\nrounds: 13312" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_items_stack_nested_locked", + "value": 2736.8984126066566, + "unit": "iter/sec", + "range": "stddev: 0.000005145715753754919", + "extra": "mean: 365.3770981757366 usec\nrounds: 2628" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys", + "value": 254240.01824592578, + "unit": "iter/sec", + "range": "stddev: 3.189875659257823e-7", + "extra": "mean: 3.933291095946596 usec\nrounds: 142654" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested", + "value": 11452.665503744676, + "unit": "iter/sec", + "range": "stddev: 0.000004071323996627144", + "extra": "mean: 87.31591782480945 usec\nrounds: 8932" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested_locked", + "value": 10707.123540173194, + "unit": "iter/sec", + "range": "stddev: 0.00000765464690871456", + "extra": "mean: 93.39576556187045 usec\nrounds: 9627" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_nested_leaf", + "value": 12795.962108304988, + "unit": "iter/sec", + "range": "stddev: 0.0000019784546590646364", + "extra": "mean: 78.14965311213044 usec\nrounds: 7250" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested", + "value": 11409.305050776102, + "unit": "iter/sec", + "range": "stddev: 0.000002299293635627974", + "extra": "mean: 87.6477572954346 usec\nrounds: 9728" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested_leaf", + "value": 12679.002447530778, + "unit": "iter/sec", + "range": "stddev: 0.000001948711518029488", + "extra": "mean: 78.87055816404144 usec\nrounds: 10642" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_keys_stack_nested_locked", + "value": 10602.033144714507, + "unit": "iter/sec", + "range": "stddev: 0.0000025632471177221786", + "extra": "mean: 94.32153119597969 usec\nrounds: 9906" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values", + "value": 1160138.7343689369, + "unit": "iter/sec", + "range": "stddev: 6.397575735268609e-8", + "extra": "mean: 861.9658756105191 nsec\nrounds: 185186" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested", + "value": 26666.854290052604, + "unit": "iter/sec", + "range": "stddev: 0.0000011673525570932609", + "extra": "mean: 37.4997361564699 usec\nrounds: 20251" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested_locked", + "value": 25682.376537517586, + "unit": "iter/sec", + "range": "stddev: 0.000001231644939277366", + "extra": "mean: 38.93720655248435 usec\nrounds: 23447" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_nested_leaf", + "value": 24012.85697519562, + "unit": "iter/sec", + "range": "stddev: 0.000001273956018231167", + "extra": "mean: 41.64435748036823 usec\nrounds: 20404" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested", + "value": 26159.810479960288, + "unit": "iter/sec", + "range": "stddev: 0.0000012286048762031927", + "extra": "mean: 38.22657663235174 usec\nrounds: 20873" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested_leaf", + "value": 23758.098539237137, + "unit": "iter/sec", + "range": "stddev: 0.0000013765931014976494", + "extra": "mean: 42.090910530927935 usec\nrounds: 18833" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_values_stack_nested_locked", + "value": 25085.959728893664, + "unit": "iter/sec", + "range": "stddev: 0.000001261810089133335", + "extra": "mean: 39.86293571412433 usec\nrounds: 22650" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership", + "value": 1882834.3852443383, + "unit": "iter/sec", + "range": "stddev: 2.6640604178622853e-8", + "extra": "mean: 531.1141584395 nsec\nrounds: 87951" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested", + "value": 464299.84194157796, + "unit": "iter/sec", + "range": "stddev: 2.5476593525915965e-7", + "extra": "mean: 2.1537806168924525 usec\nrounds: 191939" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_leaf", + "value": 476822.46099407406, + "unit": "iter/sec", + "range": "stddev: 1.8676418334165941e-7", + "extra": "mean: 2.097216641001373 usec\nrounds: 170358" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested", + "value": 471476.4405618852, + "unit": "iter/sec", + "range": "stddev: 2.6061602693553534e-7", + "extra": "mean: 2.1209967539592083 usec\nrounds: 90580" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_leaf", + "value": 465555.2821727948, + "unit": "iter/sec", + "range": "stddev: 2.903771163158301e-7", + "extra": "mean: 2.1479726217107804 usec\nrounds: 136240" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_last", + "value": 319890.1132319267, + "unit": "iter/sec", + "range": "stddev: 3.4567765745103904e-7", + "extra": "mean: 3.126073481598914 usec\nrounds: 132101" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_nested_leaf_last", + "value": 320176.5320546538, + "unit": "iter/sec", + "range": "stddev: 4.126324781465416e-7", + "extra": "mean: 3.123277004665979 usec\nrounds: 166113" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_last", + "value": 120164.76481267264, + "unit": "iter/sec", + "range": "stddev: 5.819602041179542e-7", + "extra": "mean: 8.321907021238054 usec\nrounds: 37384" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_membership_stacked_nested_leaf_last", + "value": 120985.36785514349, + "unit": "iter/sec", + "range": "stddev: 5.503524350679867e-7", + "extra": "mean: 8.265462325967436 usec\nrounds: 59666" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getleaf", + "value": 162699.16044690786, + "unit": "iter/sec", + "range": "stddev: 3.983595140865674e-7", + "extra": "mean: 6.146313215465675 usec\nrounds: 103520" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_get", + "value": 171920.4135264414, + "unit": "iter/sec", + "range": "stddev: 4.230411935026075e-7", + "extra": "mean: 5.816644920099612 usec\nrounds: 113766" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getleaf", + "value": 162905.66562929287, + "unit": "iter/sec", + "range": "stddev: 4.4890932131584296e-7", + "extra": "mean: 6.138521923943356 usec\nrounds: 62419" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_get", + "value": 171436.32160894625, + "unit": "iter/sec", + "range": "stddev: 3.8093160180409584e-7", + "extra": "mean: 5.833069623839945 usec\nrounds: 86133" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getitemleaf", + "value": 155648.63207997422, + "unit": "iter/sec", + "range": "stddev: 4.539978334140393e-7", + "extra": "mean: 6.424727198927052 usec\nrounds: 54025" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_nested_getitem", + "value": 163438.2870652427, + "unit": "iter/sec", + "range": "stddev: 4.370468224271058e-7", + "extra": "mean: 6.118517380207316 usec\nrounds: 99592" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getitemleaf", + "value": 155805.2879578902, + "unit": "iter/sec", + "range": "stddev: 4.650527164329071e-7", + "extra": "mean: 6.418267397126289 usec\nrounds: 61388" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_stacked_getitem", + "value": 164014.50767696678, + "unit": "iter/sec", + "range": "stddev: 0.000001302967675605025", + "extra": "mean: 6.097021624267169 usec\nrounds: 68255" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_lock_nested", + "value": 2788.921937955042, + "unit": "iter/sec", + "range": "stddev: 0.000260737501310155", + "extra": "mean: 358.56148800394294 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_lock_stack_nested", + "value": 2866.251665753858, + "unit": "iter/sec", + "range": "stddev: 0.000005716264667957975", + "extra": "mean: 348.887716995705 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unlock_nested", + "value": 3413.052675681396, + "unit": "iter/sec", + "range": "stddev: 0.000006529445504623001", + "extra": "mean: 292.99284102035017 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unlock_stack_nested", + "value": 3488.0903566708353, + "unit": "iter/sec", + "range": "stddev: 0.000005420418798233413", + "extra": "mean: 286.6898210040745 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_flatten_speed", + "value": 13180.522247472285, + "unit": "iter/sec", + "range": "stddev: 0.000002023953890714276", + "extra": "mean: 75.86952787032217 usec\nrounds: 9004" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unflatten_speed", + "value": 3140.4398089704846, + "unit": "iter/sec", + "range": "stddev: 0.000004399145606811417", + "extra": "mean: 318.4267366448348 usec\nrounds: 2992" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_common_ops", + "value": 1720.2136350221526, + "unit": "iter/sec", + "range": "stddev: 0.00001859498265225746", + "extra": "mean: 581.3231447773766 usec\nrounds: 1202" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation", + "value": 574016.8846710068, + "unit": "iter/sec", + "range": "stddev: 2.5918286410748253e-7", + "extra": "mean: 1.7421090332092617 usec\nrounds: 134409" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_empty", + "value": 154934.06825241813, + "unit": "iter/sec", + "range": "stddev: 4.515873201053161e-7", + "extra": "mean: 6.454358368559734 usec\nrounds: 41136" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_nested_1", + "value": 124293.41927243028, + "unit": "iter/sec", + "range": "stddev: 4.7078823053995754e-7", + "extra": "mean: 8.045478238941742 usec\nrounds: 41544" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_creation_nested_2", + "value": 92784.4658128814, + "unit": "iter/sec", + "range": "stddev: 6.191042625021175e-7", + "extra": "mean: 10.77766618839733 usec\nrounds: 34154" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_clone", + "value": 89119.41041979614, + "unit": "iter/sec", + "range": "stddev: 7.129920939367923e-7", + "extra": "mean: 11.220900085508974 usec\nrounds: 18886" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[int]", + "value": 93206.9810914346, + "unit": "iter/sec", + "range": "stddev: 0.000013450399914109662", + "extra": "mean: 10.72881009866649 usec\nrounds: 18199" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[slice_int]", + "value": 47059.84370955309, + "unit": "iter/sec", + "range": "stddev: 0.0000066365105466632215", + "extra": "mean: 21.24953933489161 usec\nrounds: 10437" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[range]", + "value": 25928.488065907128, + "unit": "iter/sec", + "range": "stddev: 0.000006984913994778898", + "extra": "mean: 38.56761711126847 usec\nrounds: 6007" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[tuple]", + "value": 53485.78105327917, + "unit": "iter/sec", + "range": "stddev: 0.000006607239994275295", + "extra": "mean: 18.696557857196904 usec\nrounds: 16748" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_getitem[list]", + "value": 30016.03257468362, + "unit": "iter/sec", + "range": "stddev: 0.000007394950254474946", + "extra": "mean: 33.31552887650544 usec\nrounds: 14389" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[int]", + "value": 48827.89135956672, + "unit": "iter/sec", + "range": "stddev: 0.0000011670622987419742", + "extra": "mean: 20.480097996369295 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[slice_int]", + "value": 25302.11163738698, + "unit": "iter/sec", + "range": "stddev: 0.0000021822880874109824", + "extra": "mean: 39.52239300542715 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[range]", + "value": 18811.02189911103, + "unit": "iter/sec", + "range": "stddev: 0.0000021773573274594565", + "extra": "mean: 53.16032299379003 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem_dim[tuple]", + "value": 30371.672118863495, + "unit": "iter/sec", + "range": "stddev: 0.00000174235656187194", + "extra": "mean: 32.92541800419713 usec\nrounds: 1000" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_setitem", + "value": 67165.35859041981, + "unit": "iter/sec", + "range": "stddev: 0.0000011638377128879408", + "extra": "mean: 14.888627426201754 usec\nrounds: 26089" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set", + "value": 70283.38986372076, + "unit": "iter/sec", + "range": "stddev: 0.0000011931696574705662", + "extra": "mean: 14.228112814976573 usec\nrounds: 28994" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_shared", + "value": 6160.043024651758, + "unit": "iter/sec", + "range": "stddev: 0.00001791910857191411", + "extra": "mean: 162.3365284947068 usec\nrounds: 3544" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update", + "value": 61270.44060427904, + "unit": "iter/sec", + "range": "stddev: 0.000003924946564980518", + "extra": "mean: 16.321083872378118 usec\nrounds: 35804" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update_nested", + "value": 46440.28678157897, + "unit": "iter/sec", + "range": "stddev: 0.0000013501627993827526", + "extra": "mean: 21.53302809483641 usec\nrounds: 25628" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_update__nested", + "value": 37974.86452487698, + "unit": "iter/sec", + "range": "stddev: 0.000006083738448009702", + "extra": "mean: 26.33320783395841 usec\nrounds: 10715" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_nested", + "value": 63307.50201463689, + "unit": "iter/sec", + "range": "stddev: 0.0000013560739821457361", + "extra": "mean: 15.795916252844677 usec\nrounds: 25170" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_set_nested_new", + "value": 55119.22851096558, + "unit": "iter/sec", + "range": "stddev: 0.0000013600721651028108", + "extra": "mean: 18.14248905535855 usec\nrounds: 22568" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_select", + "value": 34814.103847341925, + "unit": "iter/sec", + "range": "stddev: 0.0000015448690705732965", + "extra": "mean: 28.723990839601935 usec\nrounds: 14077" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_select_nested", + "value": 22404.613302466787, + "unit": "iter/sec", + "range": "stddev: 0.0000016061763476379992", + "extra": "mean: 44.633664794825904 usec\nrounds: 15295" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_exclude_nested", + "value": 15520.638810918434, + "unit": "iter/sec", + "range": "stddev: 0.000002131786758636674", + "extra": "mean: 64.43033770597906 usec\nrounds: 11125" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_empty[True]", + "value": 3392.442353003501, + "unit": "iter/sec", + "range": "stddev: 0.000009629082550524993", + "extra": "mean: 294.77287922509555 usec\nrounds: 2931" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_empty[False]", + "value": 1187395.4361866827, + "unit": "iter/sec", + "range": "stddev: 4.5155677550748396e-8", + "extra": "mean: 842.1794202035148 nsec\nrounds: 109650" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_to", + "value": 17536.607704517577, + "unit": "iter/sec", + "range": "stddev: 0.000013885841914588611", + "extra": "mean: 57.02357131147957 usec\nrounds: 7" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_to_nonblocking", + "value": 20909.15854265564, + "unit": "iter/sec", + "range": "stddev: 0.0000022747276509878117", + "extra": "mean: 47.82593225642984 usec\nrounds: 10909" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed", + "value": 4002.021682706052, + "unit": "iter/sec", + "range": "stddev: 0.00000547678834170902", + "extra": "mean: 249.87370866112568 usec\nrounds: 2722" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed_stack0", + "value": 4104.679347182351, + "unit": "iter/sec", + "range": "stddev: 0.000005947758329848257", + "extra": "mean: 243.62438948768266 usec\nrounds: 3027" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_unbind_speed_stack1", + "value": 1336.5432962481934, + "unit": "iter/sec", + "range": "stddev: 0.0035181139616653517", + "extra": "mean: 748.1987323621292 usec\nrounds: 1390" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_split", + "value": 620.6900957767594, + "unit": "iter/sec", + "range": "stddev: 0.0036955697559021955", + "extra": "mean: 1.6111099674444702 msec\nrounds: 645" + }, + { + "name": "benchmarks/common/common_ops_test.py::test_chunk", + "value": 623.1747682237317, + "unit": "iter/sec", + "range": "stddev: 0.003715647404090743", + "extra": "mean: 1.604686279019854 msec\nrounds: 663" + }, + { + "name": "benchmarks/common/h2d_test.py::TestConsolidate::test_consolidate[False-None]", + "value": 373.3716579085373, + "unit": "iter/sec", + "range": "stddev: 0.0000590914951038695", + "extra": "mean: 2.6782964877450985 msec\nrounds: 367" + }, + { + "name": "benchmarks/common/h2d_test.py::TestConsolidate::test_consolidate[default-None]", + "value": 586.8735488194686, + "unit": "iter/sec", + "range": "stddev: 0.000015173277473140624", + "extra": "mean: 1.7039445754738138 msec\nrounds: 570" + }, + { + "name": "benchmarks/common/h2d_test.py::TestConsolidate::test_consolidate[reduce-overhead-None]", + "value": 572.7073941109427, + "unit": "iter/sec", + "range": "stddev: 0.00003066367013309106", + "extra": "mean: 1.7460923506188988 msec\nrounds: 559" + }, + { + "name": "benchmarks/common/h2d_test.py::TestConsolidate::test_consolidate_njt[False-None]", + "value": 153.42266200424334, + "unit": "iter/sec", + "range": "stddev: 0.00016886589560218378", + "extra": "mean: 6.517941918986794 msec\nrounds: 148" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to[False-False-None]", + "value": 585.3532691641097, + "unit": "iter/sec", + "range": "stddev: 0.000018053346809208804", + "extra": "mean: 1.70837006074641 msec\nrounds: 576" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to[True-False-None]", + "value": 751.8452207995068, + "unit": "iter/sec", + "range": "stddev: 0.00004704151516900713", + "extra": "mean: 1.3300609917246093 msec\nrounds: 724" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to[within-False-None]", + "value": 244.4726483305295, + "unit": "iter/sec", + "range": "stddev: 0.00005922324539090394", + "extra": "mean: 4.090437138178296 msec\nrounds: 246" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to[True-default-None]", + "value": 183.9494130309288, + "unit": "iter/sec", + "range": "stddev: 0.00016092401832611583", + "extra": "mean: 5.436277199927038 msec\nrounds: 180" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to_njt[False-False-None]", + "value": 144.96855915416862, + "unit": "iter/sec", + "range": "stddev: 0.000051819457295466535", + "extra": "mean: 6.898047451354866 msec\nrounds: 144" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to_njt[True-False-None]", + "value": 176.2018210335298, + "unit": "iter/sec", + "range": "stddev: 0.0001172612429506842", + "extra": "mean: 5.675310244436737 msec\nrounds: 180" + }, + { + "name": "benchmarks/common/h2d_test.py::TestTo::test_to_njt[within-False-None]", + "value": 82.12551340709712, + "unit": "iter/sec", + "range": "stddev: 0.00015519394662358359", + "extra": "mean: 12.176484000081539 msec\nrounds: 83" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_creation[device0]", + "value": 11771.46342688831, + "unit": "iter/sec", + "range": "stddev: 0.000011416263612614832", + "extra": "mean: 84.95120476828784 usec\nrounds: 2842" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_creation_from_tensor", + "value": 11310.068321500938, + "unit": "iter/sec", + "range": "stddev: 0.000015623167928155318", + "extra": "mean: 88.4167956880469 usec\nrounds: 7058" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_add_one[memmap_tensor0]", + "value": 139215.10523959127, + "unit": "iter/sec", + "range": "stddev: 0.0000017263649885947897", + "extra": "mean: 7.183128571278132 usec\nrounds: 23878" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_contiguous[memmap_tensor0]", + "value": 2483680.784247787, + "unit": "iter/sec", + "range": "stddev: 2.452991925603649e-8", + "extra": "mean: 402.6282307864544 nsec\nrounds: 118064" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_stack[memmap_tensor0]", + "value": 214040.64255580187, + "unit": "iter/sec", + "range": "stddev: 5.37075094331111e-7", + "extra": "mean: 4.672009895220218 usec\nrounds: 21930" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index", + "value": 3975.5082230188705, + "unit": "iter/sec", + "range": "stddev: 0.000028521894389604302", + "extra": "mean: 251.54016641440447 usec\nrounds: 1941" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index_astensor", + "value": 3188.234577742748, + "unit": "iter/sec", + "range": "stddev: 0.000008680582464937568", + "extra": "mean: 313.65320700711874 usec\nrounds: 2826" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_memmaptd_index_op", + "value": 1754.8070981058518, + "unit": "iter/sec", + "range": "stddev: 0.000018782095721402187", + "extra": "mean: 569.8632066620914 usec\nrounds: 1621" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_model", + "value": 7.644661290725596, + "unit": "iter/sec", + "range": "stddev: 0.0008095614653094127", + "extra": "mean: 130.81024285708867 msec\nrounds: 7" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_model_pickle", + "value": 0.8231274358192101, + "unit": "iter/sec", + "range": "stddev: 0.3191921720390888", + "extra": "mean: 1.2148787131663994 sec\nrounds: 6" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights", + "value": 7.6632650294456495, + "unit": "iter/sec", + "range": "stddev: 0.0013083365308296125", + "extra": "mean: 130.49268114277115 msec\nrounds: 7" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights_returnearly", + "value": 14.022864783532473, + "unit": "iter/sec", + "range": "stddev: 0.13781636146139964", + "extra": "mean: 71.31210458324706 msec\nrounds: 24" + }, + { + "name": "benchmarks/common/memmap_benchmarks_test.py::test_serialize_weights_pickle", + "value": 0.8220995991035457, + "unit": "iter/sec", + "range": "stddev: 0.33176050406302726", + "extra": "mean: 1.2163976251666402 sec\nrounds: 6" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_reshape_pytree", + "value": 45087.83649711019, + "unit": "iter/sec", + "range": "stddev: 0.0000011007490323884962", + "extra": "mean: 22.17893067599491 usec\nrounds: 12290" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_reshape_td", + "value": 36269.06866360328, + "unit": "iter/sec", + "range": "stddev: 0.0000013694207111499352", + "extra": "mean: 27.571703295583095 usec\nrounds: 8362" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_view_pytree", + "value": 45732.19284248713, + "unit": "iter/sec", + "range": "stddev: 0.0000010972805127761802", + "extra": "mean: 21.86643451461523 usec\nrounds: 17737" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_view_td", + "value": 32029.808257030432, + "unit": "iter/sec", + "range": "stddev: 0.0000015892981310007973", + "extra": "mean: 31.220917464609027 usec\nrounds: 11959" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_unbind_pytree", + "value": 35109.740742431255, + "unit": "iter/sec", + "range": "stddev: 0.000001737463168466828", + "extra": "mean: 28.48212430094842 usec\nrounds: 14449" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_unbind_td", + "value": 26273.910863384364, + "unit": "iter/sec", + "range": "stddev: 0.0000064585538947630675", + "extra": "mean: 38.06056910216636 usec\nrounds: 11794" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_split_pytree", + "value": 32587.578703875555, + "unit": "iter/sec", + "range": "stddev: 0.0000019083780738176504", + "extra": "mean: 30.68653885233494 usec\nrounds: 13744" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_split_td", + "value": 24492.467901609347, + "unit": "iter/sec", + "range": "stddev: 0.00001539240127154413", + "extra": "mean: 40.828878658417764 usec\nrounds: 11752" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_add_pytree", + "value": 27479.114121117054, + "unit": "iter/sec", + "range": "stddev: 0.000001593567173187583", + "extra": "mean: 36.39127504592746 usec\nrounds: 10209" + }, + { + "name": "benchmarks/common/pytree_benchmarks_test.py::test_add_td", + "value": 20815.641925456337, + "unit": "iter/sec", + "range": "stddev: 0.000002701601760618329", + "extra": "mean: 48.040795646905195 usec\nrounds: 8446" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[tensordict-compile]", + "value": 7855.27109032456, + "unit": "iter/sec", + "range": "stddev: 0.000003927903135835776", + "extra": "mean: 127.30305402594102 usec\nrounds: 2073" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[tensordict-eager]", + "value": 7481.755258289897, + "unit": "iter/sec", + "range": "stddev: 0.0000037031620166875127", + "extra": "mean: 133.65847524776822 usec\nrounds: 6504" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[pytree-compile]", + "value": 10365.348144926029, + "unit": "iter/sec", + "range": "stddev: 0.000002930403312977419", + "extra": "mean: 96.47529306475951 usec\nrounds: 3146" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_nested[pytree-eager]", + "value": 6627.526503712119, + "unit": "iter/sec", + "range": "stddev: 0.000008387925377076832", + "extra": "mean: 150.88585454013557 usec\nrounds: 6036" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[tensordict-compile]", + "value": 40627.92894436823, + "unit": "iter/sec", + "range": "stddev: 0.000001370053011356232", + "extra": "mean: 24.613610045673227 usec\nrounds: 28288" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[tensordict-eager]", + "value": 34056.30532723554, + "unit": "iter/sec", + "range": "stddev: 0.000001166657402429155", + "extra": "mean: 29.363138202789106 usec\nrounds: 28465" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[pytree-compile]", + "value": 15320.68123854602, + "unit": "iter/sec", + "range": "stddev: 0.0000023710967263844405", + "extra": "mean: 65.27124900190816 usec\nrounds: 13486" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_nested[pytree-eager]", + "value": 20192.24539157582, + "unit": "iter/sec", + "range": "stddev: 0.0000019037039136439765", + "extra": "mean: 49.52396232353629 usec\nrounds: 16987" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensordict-compile]", + "value": 6946.13230955687, + "unit": "iter/sec", + "range": "stddev: 0.000005037982991808166", + "extra": "mean: 143.96500893369756 usec\nrounds: 1680" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensordict-eager]", + "value": 4696.315291370474, + "unit": "iter/sec", + "range": "stddev: 0.0000053335823842114445", + "extra": "mean: 212.93289269515398 usec\nrounds: 4464" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensorclass-compile]", + "value": 9871.529448702604, + "unit": "iter/sec", + "range": "stddev: 0.000003891003445653546", + "extra": "mean: 101.30142499158812 usec\nrounds: 3400" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[tensorclass-eager]", + "value": 18105.622911668368, + "unit": "iter/sec", + "range": "stddev: 0.0000027913219928430527", + "extra": "mean: 55.23146068371605 usec\nrounds: 13721" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[pytree-compile]", + "value": 7388.651489374289, + "unit": "iter/sec", + "range": "stddev: 0.000003926543468667908", + "extra": "mean: 135.34269432495392 usec\nrounds: 1747" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_one_flat[pytree-eager]", + "value": 2045.753821769655, + "unit": "iter/sec", + "range": "stddev: 0.0000070085212422047995", + "extra": "mean: 488.817368619144 usec\nrounds: 2021" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensordict-eager]", + "value": 3846.3459684475083, + "unit": "iter/sec", + "range": "stddev: 0.000006185396782070665", + "extra": "mean: 259.9870131816633 usec\nrounds: 3563" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensordict-compile]", + "value": 7036.548495887617, + "unit": "iter/sec", + "range": "stddev: 0.0000045272371363364886", + "extra": "mean: 142.11512939681037 usec\nrounds: 1708" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensorclass-eager]", + "value": 14711.340785056229, + "unit": "iter/sec", + "range": "stddev: 0.00000264550541626201", + "extra": "mean: 67.9747695747623 usec\nrounds: 12195" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[tensorclass-compile]", + "value": 10209.439929767806, + "unit": "iter/sec", + "range": "stddev: 0.00000328870741681504", + "extra": "mean: 97.94856592321838 usec\nrounds: 3398" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[pytree-eager]", + "value": 2489.1324350670598, + "unit": "iter/sec", + "range": "stddev: 0.000005482104213025963", + "extra": "mean: 401.74640204431665 usec\nrounds: 2450" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_add_self_flat[pytree-compile]", + "value": 7421.617955718589, + "unit": "iter/sec", + "range": "stddev: 0.000003880313060259706", + "extra": "mean: 134.7415086530382 usec\nrounds: 1791" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[tensordict-compile]", + "value": 55320.26218139977, + "unit": "iter/sec", + "range": "stddev: 0.0000012203676712044313", + "extra": "mean: 18.076559303369102 usec\nrounds: 35310" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[tensordict-eager]", + "value": 31920.754306085033, + "unit": "iter/sec", + "range": "stddev: 0.0000011009421795926665", + "extra": "mean: 31.32758049547002 usec\nrounds: 29206" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[pytree-compile]", + "value": 14014.934201726952, + "unit": "iter/sec", + "range": "stddev: 0.000002008322315975576", + "extra": "mean: 71.35245771448415 usec\nrounds: 13043" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_copy_flat[pytree-eager]", + "value": 19267.68426403401, + "unit": "iter/sec", + "range": "stddev: 0.0000015612331105417437", + "extra": "mean: 51.90037299223594 usec\nrounds: 17271" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[tensordict-compile]", + "value": 2195.7512208599155, + "unit": "iter/sec", + "range": "stddev: 0.0006568340668553962", + "extra": "mean: 455.42500010924414 usec\nrounds: 5" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[tensordict-eager]", + "value": 375.8072418918844, + "unit": "iter/sec", + "range": "stddev: 0.00007890425367003603", + "extra": "mean: 2.6609386103519763 msec\nrounds: 367" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[pytree-compile]", + "value": 2208.1791852468755, + "unit": "iter/sec", + "range": "stddev: 0.0006710612642137398", + "extra": "mean: 452.86179974937113 usec\nrounds: 5" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_assign_and_add[pytree-eager]", + "value": 365.79474677265927, + "unit": "iter/sec", + "range": "stddev: 0.00009648714569611537", + "extra": "mean: 2.733773540551959 msec\nrounds: 370" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensordict-compile]", + "value": 8747.707158576794, + "unit": "iter/sec", + "range": "stddev: 0.000005448680719493165", + "extra": "mean: 114.3156694516846 usec\nrounds: 7542" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensordict-eager]", + "value": 12604.091288156207, + "unit": "iter/sec", + "range": "stddev: 0.000012444755819132251", + "extra": "mean: 79.33931745953622 usec\nrounds: 11066" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensorclass-compile]", + "value": 9544.581822417676, + "unit": "iter/sec", + "range": "stddev: 0.0000035497780687506503", + "extra": "mean: 104.77148382250407 usec\nrounds: 8189" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-tensorclass-eager]", + "value": 14624.409905332037, + "unit": "iter/sec", + "range": "stddev: 0.0000027890355835700298", + "extra": "mean: 68.37882734915695 usec\nrounds: 12708" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-pytree-compile]", + "value": 9469.703645984626, + "unit": "iter/sec", + "range": "stddev: 0.0000035922685825695505", + "extra": "mean: 105.59992555036537 usec\nrounds: 8623" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[tensor-pytree-eager]", + "value": 14650.994802413352, + "unit": "iter/sec", + "range": "stddev: 0.0000028283139782658707", + "extra": "mean: 68.25475085386539 usec\nrounds: 13181" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensordict-compile]", + "value": 10021.792166051564, + "unit": "iter/sec", + "range": "stddev: 0.0000035072835521103323", + "extra": "mean: 99.78255220532927 usec\nrounds: 3908" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensordict-eager]", + "value": 56130.255720951005, + "unit": "iter/sec", + "range": "stddev: 0.000008233581907769368", + "extra": "mean: 17.81570361930033 usec\nrounds: 42017" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensorclass-compile]", + "value": 10399.000949032556, + "unit": "iter/sec", + "range": "stddev: 0.000003974660137817653", + "extra": "mean: 96.1630838290319 usec\nrounds: 4211" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-tensorclass-eager]", + "value": 62318.321542944155, + "unit": "iter/sec", + "range": "stddev: 0.000001106848959957387", + "extra": "mean: 16.046645276074873 usec\nrounds: 43764" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-pytree-compile]", + "value": 10168.087617003024, + "unit": "iter/sec", + "range": "stddev: 0.0000039707329738770484", + "extra": "mean: 98.34691022211543 usec\nrounds: 9268" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[slice-pytree-eager]", + "value": 61780.76713444373, + "unit": "iter/sec", + "range": "stddev: 0.0000010208824217694408", + "extra": "mean: 16.186267124586813 usec\nrounds: 44904" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensordict-compile]", + "value": 9755.602084954202, + "unit": "iter/sec", + "range": "stddev: 0.00000410874269992973", + "extra": "mean: 102.5052058593362 usec\nrounds: 8593" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensordict-eager]", + "value": 57396.33742002035, + "unit": "iter/sec", + "range": "stddev: 0.000008500614891157653", + "extra": "mean: 17.422714496259673 usec\nrounds: 42588" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensorclass-compile]", + "value": 10464.651420971246, + "unit": "iter/sec", + "range": "stddev: 0.000003718113790799897", + "extra": "mean: 95.55980030027487 usec\nrounds: 9244" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-tensorclass-eager]", + "value": 62143.74754629855, + "unit": "iter/sec", + "range": "stddev: 0.000001209720629356511", + "extra": "mean: 16.091723455444594 usec\nrounds: 48589" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-pytree-compile]", + "value": 10465.780571508994, + "unit": "iter/sec", + "range": "stddev: 0.000003245506266257375", + "extra": "mean: 95.5494903765039 usec\nrounds: 9450" + }, + { + "name": "benchmarks/compile/compile_td_test.py::test_compile_indexing[int-pytree-eager]", + "value": 51879.25060253497, + "unit": "iter/sec", + "range": "stddev: 0.000004203430437037465", + "extra": "mean: 19.275529009880053 usec\nrounds: 47640" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[eager]", + "value": 26577.268143965925, + "unit": "iter/sec", + "range": "stddev: 0.0000020273586214672306", + "extra": "mean: 37.62613954839595 usec\nrounds: 19857" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[compile]", + "value": 12400.93938370913, + "unit": "iter/sec", + "range": "stddev: 0.0000031746474050854787", + "extra": "mean: 80.63905233773501 usec\nrounds: 7680" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_add[compile-overhead]", + "value": 5593.954949446536, + "unit": "iter/sec", + "range": "stddev: 0.00008297894955089996", + "extra": "mean: 178.7643999705324 usec\nrounds: 5" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[eager]", + "value": 3812.095481149011, + "unit": "iter/sec", + "range": "stddev: 0.00001195518326472409", + "extra": "mean: 262.32291529554976 usec\nrounds: 3601" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[compile]", + "value": 3468.3454162944076, + "unit": "iter/sec", + "range": "stddev: 0.000021993735574527562", + "extra": "mean: 288.3219172179233 usec\nrounds: 2839" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap[compile-overhead]", + "value": 270.7478233629181, + "unit": "iter/sec", + "range": "stddev: 0.0017019841136222502", + "extra": "mean: 3.6934738295553036 msec\nrounds: 1279" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[eager]", + "value": 676.117079183314, + "unit": "iter/sec", + "range": "stddev: 0.000027342872496374906", + "extra": "mean: 1.479033780965726 msec\nrounds: 662" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[compile]", + "value": 721.4558871242456, + "unit": "iter/sec", + "range": "stddev: 0.00001907348985196379", + "extra": "mean: 1.3860861320101543 msec\nrounds: 606" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_mod_wrap_and_backward[compile-overhead]", + "value": 960.8254183491566, + "unit": "iter/sec", + "range": "stddev: 0.0002710777906582275", + "extra": "mean: 1.0407717998532462 msec\nrounds: 5" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[eager]", + "value": 8646.368555200102, + "unit": "iter/sec", + "range": "stddev: 0.0000044190251207257", + "extra": "mean: 115.65549092845222 usec\nrounds: 6056" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[compile]", + "value": 11182.92936874088, + "unit": "iter/sec", + "range": "stddev: 0.0000031431291583716045", + "extra": "mean: 89.42200804694816 usec\nrounds: 9199" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_add[compile-overhead]", + "value": 7768.326498488606, + "unit": "iter/sec", + "range": "stddev: 0.0000042624375731005994", + "extra": "mean: 128.72785408730678 usec\nrounds: 2995" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[eager]", + "value": 2352.9304760441005, + "unit": "iter/sec", + "range": "stddev: 0.00000872847114626274", + "extra": "mean: 425.001932773324 usec\nrounds: 2261" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[compile]", + "value": 3293.4411271498643, + "unit": "iter/sec", + "range": "stddev: 0.000006328370700826049", + "extra": "mean: 303.6337864844111 usec\nrounds: 3016" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_seq_wrap[compile-overhead]", + "value": 4439.03104187046, + "unit": "iter/sec", + "range": "stddev: 0.000005675849886494949", + "extra": "mean: 225.27438771381807 usec\nrounds: 1888" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-eager]", + "value": 1335.9306105356889, + "unit": "iter/sec", + "range": "stddev: 0.000011176719972704925", + "extra": "mean: 748.5418719457401 usec\nrounds: 1312" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-compile]", + "value": 1337.4153411858194, + "unit": "iter/sec", + "range": "stddev: 0.00001035638772368243", + "extra": "mean: 747.7108787411919 usec\nrounds: 1270" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[False-compile-overhead]", + "value": 2757.3295835402855, + "unit": "iter/sec", + "range": "stddev: 0.0000071954895050846825", + "extra": "mean: 362.6697388551011 usec\nrounds: 1390" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-eager]", + "value": 1099.2170060654068, + "unit": "iter/sec", + "range": "stddev: 0.000013570374917659454", + "extra": "mean: 909.7384724600021 usec\nrounds: 1071" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-compile]", + "value": 1314.2814554694703, + "unit": "iter/sec", + "range": "stddev: 0.00001093968337279581", + "extra": "mean: 760.8720307499075 usec\nrounds: 1236" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_runtime[True-compile-overhead]", + "value": 2626.016681418078, + "unit": "iter/sec", + "range": "stddev: 0.000007230562296082079", + "extra": "mean: 380.8048924731084 usec\nrounds: 1302" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-eager]", + "value": 1345.8919334268996, + "unit": "iter/sec", + "range": "stddev: 0.000010632739090520401", + "extra": "mean: 743.0017040475218 usec\nrounds: 1284" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-compile]", + "value": 1336.8908211983896, + "unit": "iter/sec", + "range": "stddev: 0.000009903775887954818", + "extra": "mean: 748.0042380002276 usec\nrounds: 1290" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[False-compile-overhead]", + "value": 2759.16843383482, + "unit": "iter/sec", + "range": "stddev: 0.000006797056469542463", + "extra": "mean: 362.42803728011404 usec\nrounds: 1422" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-eager]", + "value": 985.8403988257912, + "unit": "iter/sec", + "range": "stddev: 0.000013443004282132027", + "extra": "mean: 1.0143629751743528 msec\nrounds: 967" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-compile]", + "value": 1002.0953887042862, + "unit": "iter/sec", + "range": "stddev: 0.000013810722769509051", + "extra": "mean: 997.9089927686471 usec\nrounds: 968" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_func_call_cm_runtime[True-compile-overhead]", + "value": 943.250121479377, + "unit": "iter/sec", + "range": "stddev: 0.00003289226793838074", + "extra": "mean: 1.0601641889339144 msec\nrounds: 958" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[eager]", + "value": 471.74662682063143, + "unit": "iter/sec", + "range": "stddev: 0.00007189905553841326", + "extra": "mean: 2.1197819828401703 msec\nrounds: 466" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[compile]", + "value": 1195.997355655228, + "unit": "iter/sec", + "range": "stddev: 0.000025612928688416338", + "extra": "mean: 836.122249996238 usec\nrounds: 1176" + }, + { + "name": "benchmarks/compile/tensordict_nn_test.py::test_vmap_func_call_cm_runtime[compile-overhead]", + "value": 2405.331745899985, + "unit": "iter/sec", + "range": "stddev: 0.000007264053713211402", + "extra": "mean: 415.7430681670222 usec\nrounds: 1159" + }, + { + "name": "benchmarks/distributed/distributed_benchmark_test.py::test_distributed", + "value": 8184.80235212835, + "unit": "iter/sec", + "range": "stddev: 0.00009217415630607307", + "extra": "mean: 122.1776601288317 usec\nrounds: 3925" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdmodule", + "value": 48912.12110711551, + "unit": "iter/sec", + "range": "stddev: 0.000014565096701825616", + "extra": "mean: 20.444829980078794 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdmodule_dispatch", + "value": 29345.418180503882, + "unit": "iter/sec", + "range": "stddev: 0.0000016971287939248085", + "extra": "mean: 34.07686998525605 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdseq", + "value": 50482.776860786675, + "unit": "iter/sec", + "range": "stddev: 0.0000014262710018703464", + "extra": "mean: 19.808736012237205 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_tdseq_dispatch", + "value": 27026.18776002629, + "unit": "iter/sec", + "range": "stddev: 0.000002269520350333977", + "extra": "mean: 37.001148992203525 usec\nrounds: 1000" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_instantiation_functorch", + "value": 624.2024543112058, + "unit": "iter/sec", + "range": "stddev: 0.00002453541985008714", + "extra": "mean: 1.6020443256723156 msec\nrounds: 522" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_functorch", + "value": 6609.110522793326, + "unit": "iter/sec", + "range": "stddev: 0.00000432566007464088", + "extra": "mean: 151.30629099804375 usec\nrounds: 3409" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_functional_call", + "value": 6945.518459888099, + "unit": "iter/sec", + "range": "stddev: 0.000004063638141385506", + "extra": "mean: 143.9777326595877 usec\nrounds: 4874" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_exec_td_decorator", + "value": 5196.611907373576, + "unit": "iter/sec", + "range": "stddev: 0.000007749869437528532", + "extra": "mean: 192.43307328397566 usec\nrounds: 3698" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[True-True]", + "value": 1456.0011903939414, + "unit": "iter/sec", + "range": "stddev: 0.000011133097632159988", + "extra": "mean: 686.8126252901181 usec\nrounds: 1361" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[True-False]", + "value": 1459.6850468376242, + "unit": "iter/sec", + "range": "stddev: 0.000011750995559092988", + "extra": "mean: 685.0792930752276 usec\nrounds: 1399" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[False-True]", + "value": 1657.2622107897378, + "unit": "iter/sec", + "range": "stddev: 0.00001143762541981022", + "extra": "mean: 603.4048163829599 usec\nrounds: 1612" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_mlp_speed_decorator[False-False]", + "value": 1640.460416963889, + "unit": "iter/sec", + "range": "stddev: 0.000015385448233225712", + "extra": "mean: 609.5849614285529 usec\nrounds: 1634" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_transformer_speed_decorator[True-True]", + "value": 51.84335070178973, + "unit": "iter/sec", + "range": "stddev: 0.00013284285076173134", + "extra": "mean: 19.288876711540908 msec\nrounds: 52" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_transformer_speed_decorator[True-False]", + "value": 51.92411311632726, + "unit": "iter/sec", + "range": "stddev: 0.00006162949015154175", + "extra": "mean: 19.25887492309532 msec\nrounds: 52" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_transformer_speed_decorator[False-True]", + "value": 51.98436880625763, + "unit": "iter/sec", + "range": "stddev: 0.0002591344066490869", + "extra": "mean: 19.23655173590614 msec\nrounds: 53" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_vmap_transformer_speed_decorator[False-False]", + "value": 52.29163120874572, + "unit": "iter/sec", + "range": "stddev: 0.00003112853329648637", + "extra": "mean: 19.12351894336681 msec\nrounds: 53" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_to_module_speed[True]", + "value": 1027.3678237909246, + "unit": "iter/sec", + "range": "stddev: 0.000021870881707132746", + "extra": "mean: 973.3612216022699 usec\nrounds: 916" + }, + { + "name": "benchmarks/nn/functional_benchmarks_test.py::test_to_module_speed[False]", + "value": 1045.7042955730017, + "unit": "iter/sec", + "range": "stddev: 0.000011990311483681259", + "extra": "mean: 956.2932888709636 usec\nrounds: 630" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_init", + "value": 30075.392111876426, + "unit": "iter/sec", + "range": "stddev: 0.0000016801377384190726", + "extra": "mean: 33.24977430984554 usec\nrounds: 5729" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_init_nested", + "value": 14862.695878034521, + "unit": "iter/sec", + "range": "stddev: 0.0000028101193909360576", + "extra": "mean: 67.28254471504684 usec\nrounds: 9192" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_first_layer_tensor", + "value": 1410408.71170203, + "unit": "iter/sec", + "range": "stddev: 5.2108964991008916e-8", + "extra": "mean: 709.0143386828888 nsec\nrounds: 187266" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_first_layer_nontensor", + "value": 436275.522270961, + "unit": "iter/sec", + "range": "stddev: 2.7990664840232867e-7", + "extra": "mean: 2.292129512090578 usec\nrounds: 60025" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_second_layer_tensor", + "value": 644470.9684354748, + "unit": "iter/sec", + "range": "stddev: 1.899714424473031e-7", + "extra": "mean: 1.5516602748260508 usec\nrounds: 192679" + }, + { + "name": "benchmarks/tensorclass/test_tensorclass_speed.py::test_tc_second_layer_nontensor", + "value": 326420.48651434877, + "unit": "iter/sec", + "range": "stddev: 3.1245269762357423e-7", + "extra": "mean: 3.0635332073621 usec\nrounds: 135870" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_unbind", + "value": 142.87051950612965, + "unit": "iter/sec", + "range": "stddev: 0.00016928942940003637", + "extra": "mean: 6.999344605568515 msec\nrounds: 71" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_full_like", + "value": 110.30145914163782, + "unit": "iter/sec", + "range": "stddev: 0.00037676674687883103", + "extra": "mean: 9.066063203351666 msec\nrounds: 59" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_zeros_like", + "value": 231.0249876885732, + "unit": "iter/sec", + "range": "stddev: 0.00017586477849275805", + "extra": "mean: 4.328536103410693 msec\nrounds: 174" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_ones_like", + "value": 231.24741049265202, + "unit": "iter/sec", + "range": "stddev: 0.00005154515251403355", + "extra": "mean: 4.324372748086516 msec\nrounds: 262" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_clone", + "value": 159.09288347311193, + "unit": "iter/sec", + "range": "stddev: 0.0000320392449939204", + "extra": "mean: 6.285636278438618 msec\nrounds: 158" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_squeeze", + "value": 103161.0451749322, + "unit": "iter/sec", + "range": "stddev: 9.231308036899064e-7", + "extra": "mean: 9.693581509418411 usec\nrounds: 14141" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_unsqueeze", + "value": 13663.565588237381, + "unit": "iter/sec", + "range": "stddev: 0.000002873791095060512", + "extra": "mean: 73.18733851293361 usec\nrounds: 5270" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_split", + "value": 6121.010258562773, + "unit": "iter/sec", + "range": "stddev: 0.000017055293946733807", + "extra": "mean: 163.37172423475113 usec\nrounds: 3532" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_permute", + "value": 5470.210667057281, + "unit": "iter/sec", + "range": "stddev: 0.000005574083337529157", + "extra": "mean: 182.80831596161423 usec\nrounds: 2206" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_stack", + "value": 19.949563968395054, + "unit": "iter/sec", + "range": "stddev: 0.00020941588751306417", + "extra": "mean: 50.12640885706787 msec\nrounds: 21" + }, + { + "name": "benchmarks/tensorclass/test_torch_functions.py::test_cat", + "value": 19.951198589228035, + "unit": "iter/sec", + "range": "stddev: 0.00012172864904487259", + "extra": "mean: 50.12230195232058 msec\nrounds: 21" + } + ] } ] }