[Feature] TensorDictCatView

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] TensorDictCatView #1037

[github-actions]

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 216. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}16$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	60.6330μs	24.9939μs	40.0098 KOps/s	39.9619 KOps/s	$\color{#35bf28}+0.12\%$
test_plain_set_stack_nested	58.7490μs	25.2740μs	39.5664 KOps/s	39.2135 KOps/s	$\color{#35bf28}+0.90\%$
test_plain_set_nested_inplace	70.0000μs	27.4718μs	36.4009 KOps/s	36.5925 KOps/s	$\color{#d91a1a}-0.52\%$
test_plain_set_stack_nested_inplace	65.6420μs	27.4093μs	36.4839 KOps/s	36.7818 KOps/s	$\color{#d91a1a}-0.81\%$
test_items	24.4660μs	4.1839μs	239.0108 KOps/s	236.7527 KOps/s	$\color{#35bf28}+0.95\%$
test_items_nested	0.4745ms	0.3917ms	2.5533 KOps/s	2.5763 KOps/s	$\color{#d91a1a}-0.89\%$
test_items_nested_locked	0.6300ms	0.3917ms	2.5531 KOps/s	2.5883 KOps/s	$\color{#d91a1a}-1.36\%$
test_items_nested_leaf	0.1451ms	81.9545μs	12.2019 KOps/s	12.4064 KOps/s	$\color{#d91a1a}-1.65\%$
test_items_stack_nested	0.8350ms	0.3922ms	2.5496 KOps/s	2.5787 KOps/s	$\color{#d91a1a}-1.13\%$
test_items_stack_nested_leaf	0.1521ms	83.7155μs	11.9452 KOps/s	12.1643 KOps/s	$\color{#d91a1a}-1.80\%$
test_items_stack_nested_locked	0.6407ms	0.3986ms	2.5091 KOps/s	2.5877 KOps/s	$\color{#d91a1a}-3.04\%$
test_keys	32.1900μs	3.6189μs	276.3243 KOps/s	286.6842 KOps/s	$\color{#d91a1a}-3.61\%$
test_keys_nested	0.2282ms	0.1369ms	7.3067 KOps/s	7.4664 KOps/s	$\color{#d91a1a}-2.14\%$
test_keys_nested_locked	0.6492ms	0.1431ms	6.9875 KOps/s	7.2253 KOps/s	$\color{#d91a1a}-3.29\%$
test_keys_nested_leaf	0.2127ms	0.1203ms	8.3123 KOps/s	8.3585 KOps/s	$\color{#d91a1a}-0.55\%$
test_keys_stack_nested	0.5361ms	0.1420ms	7.0446 KOps/s	7.4483 KOps/s	$\textbf{\color{#d91a1a}-5.42\%}$
test_keys_stack_nested_leaf	0.1843ms	0.1197ms	8.3542 KOps/s	8.5344 KOps/s	$\color{#d91a1a}-2.11\%$
test_keys_stack_nested_locked	0.2396ms	0.1414ms	7.0724 KOps/s	7.1881 KOps/s	$\color{#d91a1a}-1.61\%$
test_values	7.0372μs	1.0522μs	950.4051 KOps/s	957.0757 KOps/s	$\color{#d91a1a}-0.70\%$
test_values_nested	0.1557ms	95.2079μs	10.5033 KOps/s	10.4245 KOps/s	$\color{#35bf28}+0.76\%$
test_values_nested_locked	0.3051ms	95.3930μs	10.4830 KOps/s	10.5562 KOps/s	$\color{#d91a1a}-0.69\%$
test_values_nested_leaf	0.1485ms	81.5929μs	12.2560 KOps/s	12.5215 KOps/s	$\color{#d91a1a}-2.12\%$
test_values_stack_nested	0.2140ms	93.8951μs	10.6502 KOps/s	10.7123 KOps/s	$\color{#d91a1a}-0.58\%$
test_values_stack_nested_leaf	0.1341ms	79.2534μs	12.6178 KOps/s	12.6195 KOps/s	$\color{#d91a1a}-0.01\%$
test_values_stack_nested_locked	0.1708ms	95.2544μs	10.4982 KOps/s	10.6905 KOps/s	$\color{#d91a1a}-1.80\%$
test_membership	24.6860μs	0.9139μs	1.0943 MOps/s	1.1588 MOps/s	$\textbf{\color{#d91a1a}-5.57\%}$
test_membership_nested	20.8190μs	2.8447μs	351.5292 KOps/s	366.9674 KOps/s	$\color{#d91a1a}-4.21\%$
test_membership_nested_leaf	36.7190μs	2.8554μs	350.2127 KOps/s	350.1097 KOps/s	$\color{#35bf28}+0.03\%$
test_membership_stacked_nested	20.5580μs	2.8365μs	352.5476 KOps/s	365.3951 KOps/s	$\color{#d91a1a}-3.52\%$
test_membership_stacked_nested_leaf	42.2080μs	2.8556μs	350.1899 KOps/s	368.7424 KOps/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_membership_nested_last	37.0290μs	4.2981μs	232.6625 KOps/s	238.8191 KOps/s	$\color{#d91a1a}-2.58\%$
test_membership_nested_leaf_last	32.3800μs	4.3654μs	229.0731 KOps/s	240.8519 KOps/s	$\color{#d91a1a}-4.89\%$
test_membership_stacked_nested_last	25.8580μs	4.3296μs	230.9673 KOps/s	238.4344 KOps/s	$\color{#d91a1a}-3.13\%$
test_membership_stacked_nested_leaf_last	34.3240μs	4.3044μs	232.3188 KOps/s	243.3476 KOps/s	$\color{#d91a1a}-4.53\%$
test_nested_getleaf	43.4500μs	10.7183μs	93.2980 KOps/s	95.1960 KOps/s	$\color{#d91a1a}-1.99\%$
test_nested_get	37.7810μs	10.3022μs	97.0663 KOps/s	99.1101 KOps/s	$\color{#d91a1a}-2.06\%$
test_stacked_getleaf	44.5030μs	10.6371μs	94.0103 KOps/s	94.5479 KOps/s	$\color{#d91a1a}-0.57\%$
test_stacked_get	39.8840μs	10.2416μs	97.6407 KOps/s	99.0488 KOps/s	$\color{#d91a1a}-1.42\%$
test_nested_getitemleaf	33.3220μs	11.4035μs	87.6922 KOps/s	90.9459 KOps/s	$\color{#d91a1a}-3.58\%$
test_nested_getitem	44.6430μs	10.5524μs	94.7655 KOps/s	99.5729 KOps/s	$\color{#d91a1a}-4.83\%$
test_stacked_getitemleaf	38.9130μs	10.9906μs	90.9866 KOps/s	91.3768 KOps/s	$\color{#d91a1a}-0.43\%$
test_stacked_getitem	0.5722ms	10.2911μs	97.1709 KOps/s	98.7044 KOps/s	$\color{#d91a1a}-1.55\%$
test_lock_nested	4.9771ms	0.5379ms	1.8591 KOps/s	2.0005 KOps/s	$\textbf{\color{#d91a1a}-7.07\%}$
test_lock_stack_nested	0.9117ms	0.4944ms	2.0227 KOps/s	2.1145 KOps/s	$\color{#d91a1a}-4.34\%$
test_unlock_nested	0.1022s	0.5506ms	1.8163 KOps/s	2.4094 KOps/s	$\textbf{\color{#d91a1a}-24.62\%}$
test_unlock_stack_nested	1.3462ms	0.4052ms	2.4682 KOps/s	2.5635 KOps/s	$\color{#d91a1a}-3.72\%$
test_flatten_speed	0.2287ms	0.1015ms	9.8547 KOps/s	10.0965 KOps/s	$\color{#d91a1a}-2.39\%$
test_unflatten_speed	1.0516ms	0.5252ms	1.9040 KOps/s	2.0007 KOps/s	$\color{#d91a1a}-4.83\%$
test_common_ops	3.4345ms	1.1564ms	864.7750 Ops/s	846.6582 Ops/s	$\color{#35bf28}+2.14\%$
test_creation	0.1198ms	2.2482μs	444.7945 KOps/s	486.5028 KOps/s	$\textbf{\color{#d91a1a}-8.57\%}$
test_creation_empty	55.3940μs	19.1510μs	52.2165 KOps/s	49.1066 KOps/s	$\textbf{\color{#35bf28}+6.33\%}$
test_creation_nested_1	76.1720μs	23.0830μs	43.3220 KOps/s	42.3875 KOps/s	$\color{#35bf28}+2.20\%$
test_creation_nested_2	76.1320μs	27.6818μs	36.1248 KOps/s	36.2274 KOps/s	$\color{#d91a1a}-0.28\%$
test_clone	93.1330μs	16.9640μs	58.9485 KOps/s	57.1630 KOps/s	$\color{#35bf28}+3.12\%$
test_getitem[int]	0.9389ms	16.9662μs	58.9408 KOps/s	60.3052 KOps/s	$\color{#d91a1a}-2.26\%$
test_getitem[slice_int]	0.1801ms	31.6195μs	31.6260 KOps/s	32.0329 KOps/s	$\color{#d91a1a}-1.27\%$
test_getitem[range]	0.5499ms	59.3622μs	16.8457 KOps/s	17.2583 KOps/s	$\color{#d91a1a}-2.39\%$
test_getitem[tuple]	0.1338ms	25.6139μs	39.0413 KOps/s	39.1521 KOps/s	$\color{#d91a1a}-0.28\%$
test_getitem[list]	0.4006ms	53.9559μs	18.5337 KOps/s	18.9235 KOps/s	$\color{#d91a1a}-2.06\%$
test_setitem_dim[int]	64.6210μs	32.5527μs	30.7194 KOps/s	30.6024 KOps/s	$\color{#35bf28}+0.38\%$
test_setitem_dim[slice_int]	0.1168ms	60.6923μs	16.4766 KOps/s	16.2198 KOps/s	$\color{#35bf28}+1.58\%$
test_setitem_dim[range]	0.1618ms	85.1056μs	11.7501 KOps/s	11.9766 KOps/s	$\color{#d91a1a}-1.89\%$
test_setitem_dim[tuple]	0.1041ms	50.0711μs	19.9716 KOps/s	20.3466 KOps/s	$\color{#d91a1a}-1.84\%$
test_setitem	0.2107ms	31.3189μs	31.9296 KOps/s	30.9466 KOps/s	$\color{#35bf28}+3.18\%$
test_set	0.1324ms	30.3797μs	32.9167 KOps/s	31.6886 KOps/s	$\color{#35bf28}+3.88\%$
test_set_shared	3.7634ms	0.2197ms	4.5513 KOps/s	4.5338 KOps/s	$\color{#35bf28}+0.39\%$
test_update	0.1930ms	39.3902μs	25.3870 KOps/s	24.5088 KOps/s	$\color{#35bf28}+3.58\%$
test_update_nested	0.2557ms	50.1345μs	19.9463 KOps/s	19.6060 KOps/s	$\color{#35bf28}+1.74\%$
test_update__nested	0.8167ms	44.6026μs	22.4202 KOps/s	22.1690 KOps/s	$\color{#35bf28}+1.13\%$
test_set_nested	0.2182ms	33.9591μs	29.4471 KOps/s	28.6354 KOps/s	$\color{#35bf28}+2.83\%$
test_set_nested_new	0.2071ms	39.0745μs	25.5921 KOps/s	25.3087 KOps/s	$\color{#35bf28}+1.12\%$
test_select	0.1249ms	57.0224μs	17.5370 KOps/s	17.5466 KOps/s	$\color{#d91a1a}-0.05\%$
test_select_nested	0.1330ms	61.1437μs	16.3549 KOps/s	16.9287 KOps/s	$\color{#d91a1a}-3.39\%$
test_exclude_nested	0.1612ms	76.2054μs	13.1224 KOps/s	13.3820 KOps/s	$\color{#d91a1a}-1.94\%$
test_empty[True]	0.9334ms	0.3650ms	2.7394 KOps/s	2.8539 KOps/s	$\color{#d91a1a}-4.01\%$
test_empty[False]	12.6835μs	1.3459μs	742.9843 KOps/s	828.8382 KOps/s	$\textbf{\color{#d91a1a}-10.36\%}$
test_unbind_speed	0.5927ms	0.3205ms	3.1202 KOps/s	3.1872 KOps/s	$\color{#d91a1a}-2.10\%$
test_unbind_speed_stack0	0.5217ms	0.3200ms	3.1253 KOps/s	3.4434 KOps/s	$\textbf{\color{#d91a1a}-9.24\%}$
test_unbind_speed_stack1	0.1066s	0.8939ms	1.1187 KOps/s	1.3239 KOps/s	$\textbf{\color{#d91a1a}-15.50\%}$
test_split	0.1105s	2.2062ms	453.2651 Ops/s	446.6058 Ops/s	$\color{#35bf28}+1.49\%$
test_chunk	2.2593ms	1.9894ms	502.6708 Ops/s	450.9001 Ops/s	$\textbf{\color{#35bf28}+11.48\%}$
test_creation[device0]	0.2488ms	0.1175ms	8.5084 KOps/s	8.2680 KOps/s	$\color{#35bf28}+2.91\%$
test_creation_from_tensor	2.4658ms	0.1176ms	8.5040 KOps/s	8.5309 KOps/s	$\color{#d91a1a}-0.32\%$
test_add_one[memmap_tensor0]	0.3475ms	7.3433μs	136.1779 KOps/s	132.1683 KOps/s	$\color{#35bf28}+3.03\%$
test_contiguous[memmap_tensor0]	25.5470μs	1.8578μs	538.2644 KOps/s	514.5438 KOps/s	$\color{#35bf28}+4.61\%$
test_stack[memmap_tensor0]	38.5920μs	5.7597μs	173.6202 KOps/s	166.0962 KOps/s	$\color{#35bf28}+4.53\%$
test_memmaptd_index	1.1201ms	0.4156ms	2.4062 KOps/s	2.4456 KOps/s	$\color{#d91a1a}-1.61\%$
test_memmaptd_index_astensor	0.8266ms	0.5182ms	1.9299 KOps/s	1.9521 KOps/s	$\color{#d91a1a}-1.14\%$
test_memmaptd_index_op	1.9167ms	1.0631ms	940.6159 Ops/s	906.8116 Ops/s	$\color{#35bf28}+3.73\%$
test_serialize_model	0.1274s	0.1184s	8.4428 Ops/s	8.3237 Ops/s	$\color{#35bf28}+1.43\%$
test_serialize_model_pickle	0.4391s	0.4026s	2.4837 Ops/s	2.5012 Ops/s	$\color{#d91a1a}-0.70\%$
test_serialize_weights	0.1264s	0.1168s	8.5641 Ops/s	7.4171 Ops/s	$\textbf{\color{#35bf28}+15.46\%}$
test_serialize_weights_returnearly	0.1716s	0.1608s	6.2200 Ops/s	6.2190 Ops/s	$\color{#35bf28}+0.02\%$
test_serialize_weights_pickle	1.2491s	0.7543s	1.3257 Ops/s	2.4801 Ops/s	$\textbf{\color{#d91a1a}-46.55\%}$
test_serialize_weights_filesystem	0.1518s	0.1468s	6.8116 Ops/s	6.8018 Ops/s	$\color{#35bf28}+0.14\%$
test_serialize_model_filesystem	0.1626s	0.1488s	6.7202 Ops/s	5.9997 Ops/s	$\textbf{\color{#35bf28}+12.01\%}$
test_reshape_pytree	83.2550μs	38.5140μs	25.9646 KOps/s	26.3922 KOps/s	$\color{#d91a1a}-1.62\%$
test_reshape_td	0.1458ms	47.4982μs	21.0534 KOps/s	21.3533 KOps/s	$\color{#d91a1a}-1.40\%$
test_view_pytree	0.1137ms	38.4895μs	25.9811 KOps/s	26.4645 KOps/s	$\color{#d91a1a}-1.83\%$
test_view_td	0.1495ms	53.5453μs	18.6758 KOps/s	19.6637 KOps/s	$\textbf{\color{#d91a1a}-5.02\%}$
test_unbind_pytree	81.6720μs	35.5487μs	28.1304 KOps/s	28.0865 KOps/s	$\color{#35bf28}+0.16\%$
test_unbind_td	0.3669ms	46.6757μs	21.4244 KOps/s	22.8422 KOps/s	$\textbf{\color{#d91a1a}-6.21\%}$
test_split_pytree	0.1022ms	37.9528μs	26.3485 KOps/s	26.9855 KOps/s	$\color{#d91a1a}-2.36\%$
test_split_td	0.6620ms	58.8184μs	17.0015 KOps/s	17.4697 KOps/s	$\color{#d91a1a}-2.68\%$
test_add_pytree	0.1057ms	44.6144μs	22.4143 KOps/s	21.2840 KOps/s	$\textbf{\color{#35bf28}+5.31\%}$
test_add_td	0.3086ms	87.2721μs	11.4584 KOps/s	10.6622 KOps/s	$\textbf{\color{#35bf28}+7.47\%}$
test_compile_add_one_nested[tensordict-compile]	0.1292ms	58.8733μs	16.9856 KOps/s	17.0892 KOps/s	$\color{#d91a1a}-0.61\%$
test_compile_add_one_nested[tensordict-eager]	1.2885ms	0.2058ms	4.8587 KOps/s	5.0873 KOps/s	$\color{#d91a1a}-4.49\%$
test_compile_add_one_nested[pytree-compile]	0.1460ms	57.8573μs	17.2839 KOps/s	17.5063 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_add_one_nested[pytree-eager]	0.2633ms	0.1400ms	7.1410 KOps/s	7.0326 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_copy_nested[tensordict-compile]	60.2820μs	23.5428μs	42.4758 KOps/s	42.7893 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_copy_nested[tensordict-eager]	0.1870ms	73.7357μs	13.5619 KOps/s	13.7223 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_copy_nested[pytree-compile]	0.3019ms	76.7849μs	13.0234 KOps/s	13.2391 KOps/s	$\color{#d91a1a}-1.63\%$
test_compile_copy_nested[pytree-eager]	0.1237ms	69.2665μs	14.4370 KOps/s	14.7260 KOps/s	$\color{#d91a1a}-1.96\%$
test_compile_add_one_flat[tensordict-compile]	0.2954ms	0.1839ms	5.4391 KOps/s	5.4539 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_add_one_flat[tensordict-eager]	0.4052ms	0.2403ms	4.1622 KOps/s	4.1529 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_add_one_flat[tensorclass-compile]	0.1083ms	48.7732μs	20.5031 KOps/s	20.4375 KOps/s	$\color{#35bf28}+0.32\%$
test_compile_add_one_flat[tensorclass-eager]	0.5342ms	77.8095μs	12.8519 KOps/s	12.4605 KOps/s	$\color{#35bf28}+3.14\%$
test_compile_add_one_flat[pytree-compile]	0.2885ms	0.1766ms	5.6616 KOps/s	5.7683 KOps/s	$\color{#d91a1a}-1.85\%$
test_compile_add_one_flat[pytree-eager]	0.5164ms	0.2868ms	3.4869 KOps/s	3.4779 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_self_flat[tensordict-eager]	0.5106ms	0.2759ms	3.6250 KOps/s	3.5920 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_add_self_flat[tensordict-compile]	0.3356ms	0.1874ms	5.3365 KOps/s	5.5811 KOps/s	$\color{#d91a1a}-4.38\%$
test_compile_add_self_flat[tensorclass-eager]	0.2515ms	75.7161μs	13.2072 KOps/s	13.3171 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_self_flat[tensorclass-compile]	0.1947ms	49.1591μs	20.3421 KOps/s	20.7412 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_add_self_flat[pytree-eager]	0.5415ms	0.2346ms	4.2633 KOps/s	4.2622 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_add_self_flat[pytree-compile]	0.2861ms	0.1761ms	5.6784 KOps/s	5.5835 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_copy_flat[tensordict-compile]	0.2247ms	0.1118ms	8.9447 KOps/s	8.9869 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_copy_flat[tensordict-eager]	0.1474ms	79.2196μs	12.6231 KOps/s	12.8831 KOps/s	$\color{#d91a1a}-2.02\%$
test_compile_copy_flat[pytree-compile]	0.1441ms	79.1601μs	12.6326 KOps/s	13.2677 KOps/s	$\color{#d91a1a}-4.79\%$
test_compile_copy_flat[pytree-eager]	0.1242ms	70.5874μs	14.1668 KOps/s	14.9946 KOps/s	$\textbf{\color{#d91a1a}-5.52\%}$
test_compile_assign_and_add[tensordict-compile]	0.2867ms	0.1969ms	5.0800 KOps/s	5.2132 KOps/s	$\color{#d91a1a}-2.56\%$
test_compile_assign_and_add[tensordict-eager]	2.7342ms	1.7540ms	570.1185 Ops/s	584.3041 Ops/s	$\color{#d91a1a}-2.43\%$
test_compile_assign_and_add[pytree-compile]	0.2793ms	0.1950ms	5.1270 KOps/s	5.1753 KOps/s	$\color{#d91a1a}-0.93\%$
test_compile_assign_and_add[pytree-eager]	1.2529ms	1.1068ms	903.4927 Ops/s	912.7660 Ops/s	$\color{#d91a1a}-1.02\%$
test_compile_assign_and_add_stack[compile]	0.5628ms	0.4257ms	2.3492 KOps/s	2.3969 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_assign_and_add_stack[eager]	6.3882ms	4.2024ms	237.9618 Ops/s	231.0427 Ops/s	$\color{#35bf28}+2.99\%$
test_compile_indexing[tensor-tensordict-compile]	92.0810μs	35.0407μs	28.5382 KOps/s	28.8084 KOps/s	$\color{#d91a1a}-0.94\%$
test_compile_indexing[tensor-tensordict-eager]	0.6303ms	48.7879μs	20.4969 KOps/s	20.5725 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_indexing[tensor-tensorclass-compile]	76.2820μs	31.5544μs	31.6913 KOps/s	32.4633 KOps/s	$\color{#d91a1a}-2.38\%$
test_compile_indexing[tensor-tensorclass-eager]	98.4630μs	29.7289μs	33.6373 KOps/s	35.1193 KOps/s	$\color{#d91a1a}-4.22\%$
test_compile_indexing[tensor-pytree-compile]	87.1520μs	31.5923μs	31.6532 KOps/s	32.5847 KOps/s	$\color{#d91a1a}-2.86\%$
test_compile_indexing[tensor-pytree-eager]	89.2960μs	29.7613μs	33.6007 KOps/s	35.4145 KOps/s	$\textbf{\color{#d91a1a}-5.12\%}$
test_compile_indexing[slice-tensordict-compile]	0.1400ms	74.1429μs	13.4875 KOps/s	13.5755 KOps/s	$\color{#d91a1a}-0.65\%$
test_compile_indexing[slice-tensordict-eager]	0.5566ms	27.8144μs	35.9525 KOps/s	36.1683 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_indexing[slice-tensorclass-compile]	0.1747ms	69.8109μs	14.3244 KOps/s	14.4711 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_indexing[slice-tensorclass-eager]	68.1460μs	23.4632μs	42.6200 KOps/s	43.9078 KOps/s	$\color{#d91a1a}-2.93\%$
test_compile_indexing[slice-pytree-compile]	0.1482ms	70.0763μs	14.2702 KOps/s	14.4778 KOps/s	$\color{#d91a1a}-1.43\%$
test_compile_indexing[slice-pytree-eager]	80.2300μs	23.1115μs	43.2685 KOps/s	44.3936 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_indexing[int-tensordict-compile]	0.1491ms	73.3826μs	13.6272 KOps/s	13.7837 KOps/s	$\color{#d91a1a}-1.14\%$
test_compile_indexing[int-tensordict-eager]	0.9074ms	27.3117μs	36.6143 KOps/s	36.1108 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_indexing[int-tensorclass-compile]	0.1463ms	69.1847μs	14.4541 KOps/s	14.4714 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_indexing[int-tensorclass-eager]	65.0810μs	23.3005μs	42.9175 KOps/s	44.3649 KOps/s	$\color{#d91a1a}-3.26\%$
test_compile_indexing[int-pytree-compile]	0.1511ms	69.3337μs	14.4230 KOps/s	14.6782 KOps/s	$\color{#d91a1a}-1.74\%$
test_compile_indexing[int-pytree-eager]	86.9820μs	23.3123μs	42.8959 KOps/s	44.9228 KOps/s	$\color{#d91a1a}-4.51\%$
test_mod_add[eager]	73.9080μs	26.9842μs	37.0587 KOps/s	36.4473 KOps/s	$\color{#35bf28}+1.68\%$
test_mod_add[compile]	95.2870μs	39.6057μs	25.2489 KOps/s	25.8747 KOps/s	$\color{#d91a1a}-2.42\%$
test_mod_add[compile-overhead]	0.1114ms	40.6875μs	24.5776 KOps/s	25.7934 KOps/s	$\color{#d91a1a}-4.71\%$
test_mod_wrap[eager]	0.3912ms	0.2102ms	4.7565 KOps/s	4.7771 KOps/s	$\color{#d91a1a}-0.43\%$
test_mod_wrap[compile]	0.4998ms	0.2381ms	4.1993 KOps/s	4.2884 KOps/s	$\color{#d91a1a}-2.08\%$
test_mod_wrap[compile-overhead]	0.4787ms	0.2376ms	4.2084 KOps/s	4.4131 KOps/s	$\color{#d91a1a}-4.64\%$
test_mod_wrap_and_backward[eager]	13.0462ms	11.0922ms	90.1531 Ops/s	76.9059 Ops/s	$\textbf{\color{#35bf28}+17.23\%}$
test_mod_wrap_and_backward[compile]	12.5540ms	11.1391ms	89.7737 Ops/s	79.8768 Ops/s	$\textbf{\color{#35bf28}+12.39\%}$
test_mod_wrap_and_backward[compile-overhead]	14.7358ms	11.3310ms	88.2537 Ops/s	79.4496 Ops/s	$\textbf{\color{#35bf28}+11.08\%}$
test_seq_add[eager]	0.2062ms	96.8401μs	10.3263 KOps/s	10.4430 KOps/s	$\color{#d91a1a}-1.12\%$
test_seq_add[compile]	0.1333ms	66.5799μs	15.0196 KOps/s	14.9757 KOps/s	$\color{#35bf28}+0.29\%$
test_seq_add[compile-overhead]	0.1355ms	65.4874μs	15.2701 KOps/s	15.6930 KOps/s	$\color{#d91a1a}-2.69\%$
test_seq_wrap[eager]	1.0723ms	0.3959ms	2.5261 KOps/s	2.5205 KOps/s	$\color{#35bf28}+0.22\%$
test_seq_wrap[compile]	0.5143ms	0.2764ms	3.6175 KOps/s	3.6772 KOps/s	$\color{#d91a1a}-1.62\%$
test_seq_wrap[compile-overhead]	0.6283ms	0.2772ms	3.6075 KOps/s	3.6767 KOps/s	$\color{#d91a1a}-1.88\%$
test_func_call_runtime[False-eager]	0.8074ms	0.5281ms	1.8936 KOps/s	1.8606 KOps/s	$\color{#35bf28}+1.77\%$
test_func_call_runtime[False-compile]	0.6316ms	0.5078ms	1.9693 KOps/s	1.9879 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_runtime[False-compile-overhead]	0.6264ms	0.5052ms	1.9793 KOps/s	2.0119 KOps/s	$\color{#d91a1a}-1.62\%$
test_func_call_runtime[True-eager]	1.3818ms	0.7564ms	1.3220 KOps/s	1.3137 KOps/s	$\color{#35bf28}+0.63\%$
test_func_call_runtime[True-compile]	0.7303ms	0.5201ms	1.9227 KOps/s	1.9199 KOps/s	$\color{#35bf28}+0.15\%$
test_func_call_runtime[True-compile-overhead]	0.8476ms	0.5270ms	1.8976 KOps/s	1.9029 KOps/s	$\color{#d91a1a}-0.28\%$
test_func_call_cm_runtime[False-eager]	1.0016ms	0.5318ms	1.8803 KOps/s	1.8406 KOps/s	$\color{#35bf28}+2.16\%$
test_func_call_cm_runtime[False-compile]	0.6768ms	0.5081ms	1.9679 KOps/s	1.9492 KOps/s	$\color{#35bf28}+0.96\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7148ms	0.5090ms	1.9645 KOps/s	1.9807 KOps/s	$\color{#d91a1a}-0.82\%$
test_func_call_cm_runtime[True-eager]	1.4040ms	0.9205ms	1.0864 KOps/s	1.0871 KOps/s	$\color{#d91a1a}-0.07\%$
test_func_call_cm_runtime[True-compile]	1.0283ms	0.7456ms	1.3413 KOps/s	1.3311 KOps/s	$\color{#35bf28}+0.76\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9318ms	0.7390ms	1.3531 KOps/s	1.3198 KOps/s	$\color{#35bf28}+2.53\%$
test_vmap_func_call_cm_runtime[eager]	2.7383ms	1.9315ms	517.7386 Ops/s	506.8652 Ops/s	$\color{#35bf28}+2.15\%$
test_vmap_func_call_cm_runtime[compile]	2.7934ms	1.9870ms	503.2639 Ops/s	489.7777 Ops/s	$\color{#35bf28}+2.75\%$
test_vmap_func_call_cm_runtime[compile-overhead]	3.3023ms	2.0034ms	499.1483 Ops/s	488.3736 Ops/s	$\color{#35bf28}+2.21\%$
test_distributed	0.3073ms	0.1297ms	7.7120 KOps/s	7.6736 KOps/s	$\color{#35bf28}+0.50\%$
test_tdmodule	90.0670μs	17.9980μs	55.5619 KOps/s	50.3022 KOps/s	$\textbf{\color{#35bf28}+10.46\%}$
test_tdmodule_dispatch	69.5800μs	36.8207μs	27.1586 KOps/s	24.3357 KOps/s	$\textbf{\color{#35bf28}+11.60\%}$
test_tdseq	46.5560μs	20.6232μs	48.4890 KOps/s	44.1162 KOps/s	$\textbf{\color{#35bf28}+9.91\%}$
test_tdseq_dispatch	74.6190μs	42.7420μs	23.3962 KOps/s	22.3521 KOps/s	$\color{#35bf28}+4.67\%$
test_instantiation_functorch	2.5894ms	1.5830ms	631.7114 Ops/s	627.2070 Ops/s	$\color{#35bf28}+0.72\%$
test_exec_functorch	0.4337ms	0.1892ms	5.2861 KOps/s	5.3827 KOps/s	$\color{#d91a1a}-1.79\%$
test_exec_functional_call	0.2688ms	0.1772ms	5.6430 KOps/s	5.6205 KOps/s	$\color{#35bf28}+0.40\%$
test_exec_td_decorator	0.6035ms	0.2350ms	4.2546 KOps/s	4.3080 KOps/s	$\color{#d91a1a}-1.24\%$
test_vmap_mlp_speed_decorator[True-True]	1.4562ms	0.6590ms	1.5174 KOps/s	1.5309 KOps/s	$\color{#d91a1a}-0.89\%$
test_vmap_mlp_speed_decorator[True-False]	1.1141ms	0.6570ms	1.5222 KOps/s	1.4814 KOps/s	$\color{#35bf28}+2.75\%$
test_vmap_mlp_speed_decorator[False-True]	0.8519ms	0.5397ms	1.8530 KOps/s	1.8583 KOps/s	$\color{#d91a1a}-0.28\%$
test_vmap_mlp_speed_decorator[False-False]	0.9726ms	0.5374ms	1.8608 KOps/s	1.8655 KOps/s	$\color{#d91a1a}-0.25\%$
test_to_module_speed[True]	2.5408ms	1.4536ms	687.9291 Ops/s	700.6840 Ops/s	$\color{#d91a1a}-1.82\%$
test_to_module_speed[False]	1.5200ms	1.3934ms	717.6720 Ops/s	728.8604 Ops/s	$\color{#d91a1a}-1.54\%$
test_tc_init	98.7640μs	47.5293μs	21.0396 KOps/s	19.7238 KOps/s	$\textbf{\color{#35bf28}+6.67\%}$
test_tc_init_nested	0.1757ms	96.0922μs	10.4067 KOps/s	9.9529 KOps/s	$\color{#35bf28}+4.56\%$
test_tc_first_layer_tensor	33.8430μs	1.5471μs	646.3876 KOps/s	647.4782 KOps/s	$\color{#d91a1a}-0.17\%$
test_tc_first_layer_nontensor	29.2240μs	4.6780μs	213.7655 KOps/s	220.5462 KOps/s	$\color{#d91a1a}-3.07\%$
test_tc_second_layer_tensor	25.2170μs	2.8414μs	351.9343 KOps/s	358.0101 KOps/s	$\color{#d91a1a}-1.70\%$
test_tc_second_layer_nontensor	0.1693ms	6.0018μs	166.6175 KOps/s	169.6431 KOps/s	$\color{#d91a1a}-1.78\%$
test_unbind	8.3194ms	7.7398ms	129.2020 Ops/s	73.2024 Ops/s	$\textbf{\color{#35bf28}+76.50\%}$
test_full_like	23.2310ms	13.9526ms	71.6714 Ops/s	119.8505 Ops/s	$\textbf{\color{#d91a1a}-40.20\%}$
test_zeros_like	15.5192ms	8.0245ms	124.6191 Ops/s	313.2111 Ops/s	$\textbf{\color{#d91a1a}-60.21\%}$
test_ones_like	17.0311ms	8.1184ms	123.1768 Ops/s	120.9727 Ops/s	$\color{#35bf28}+1.82\%$
test_clone	16.5694ms	10.2050ms	97.9913 Ops/s	97.2555 Ops/s	$\color{#35bf28}+0.76\%$
test_squeeze	72.6950μs	12.3614μs	80.8970 KOps/s	82.6306 KOps/s	$\color{#d91a1a}-2.10\%$
test_unsqueeze	0.1742ms	92.0910μs	10.8588 KOps/s	10.7289 KOps/s	$\color{#35bf28}+1.21\%$
test_split	0.4974ms	0.1921ms	5.2052 KOps/s	4.9972 KOps/s	$\color{#35bf28}+4.16\%$
test_permute	0.3670ms	0.2178ms	4.5903 KOps/s	4.4803 KOps/s	$\color{#35bf28}+2.45\%$
test_stack	34.6362ms	28.4559ms	35.1421 Ops/s	36.3687 Ops/s	$\color{#d91a1a}-3.37\%$
test_cat	36.8149ms	28.2119ms	35.4461 Ops/s	35.9415 Ops/s	$\color{#d91a1a}-1.38\%$

[github-actions]

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 218. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1455ms	16.9090μs	59.1401 KOps/s	59.9299 KOps/s	$\color{#d91a1a}-1.32\%$
test_plain_set_stack_nested	45.4400μs	17.0521μs	58.6437 KOps/s	59.4081 KOps/s	$\color{#d91a1a}-1.29\%$
test_plain_set_nested_inplace	48.6500μs	18.1886μs	54.9795 KOps/s	55.5015 KOps/s	$\color{#d91a1a}-0.94\%$
test_plain_set_stack_nested_inplace	48.8400μs	18.2093μs	54.9169 KOps/s	55.9320 KOps/s	$\color{#d91a1a}-1.81\%$
test_items	25.7610μs	2.8956μs	345.3459 KOps/s	340.5744 KOps/s	$\color{#35bf28}+1.40\%$
test_items_nested	0.3721ms	0.3406ms	2.9356 KOps/s	2.9664 KOps/s	$\color{#d91a1a}-1.04\%$
test_items_nested_locked	0.3749ms	0.3432ms	2.9137 KOps/s	2.9372 KOps/s	$\color{#d91a1a}-0.80\%$
test_items_nested_leaf	0.1137ms	63.7953μs	15.6751 KOps/s	15.9089 KOps/s	$\color{#d91a1a}-1.47\%$
test_items_stack_nested	0.5136ms	0.3440ms	2.9068 KOps/s	2.9190 KOps/s	$\color{#d91a1a}-0.42\%$
test_items_stack_nested_leaf	93.4410μs	64.6424μs	15.4697 KOps/s	15.5473 KOps/s	$\color{#d91a1a}-0.50\%$
test_items_stack_nested_locked	0.3901ms	0.3419ms	2.9245 KOps/s	2.9203 KOps/s	$\color{#35bf28}+0.15\%$
test_keys	23.3700μs	3.4059μs	293.6054 KOps/s	273.1073 KOps/s	$\textbf{\color{#35bf28}+7.51\%}$
test_keys_nested	97.1610μs	71.6385μs	13.9590 KOps/s	14.1039 KOps/s	$\color{#d91a1a}-1.03\%$
test_keys_nested_locked	0.8463ms	76.9576μs	12.9942 KOps/s	12.9290 KOps/s	$\color{#35bf28}+0.50\%$
test_keys_nested_leaf	93.5310μs	62.2056μs	16.0757 KOps/s	16.1785 KOps/s	$\color{#d91a1a}-0.64\%$
test_keys_stack_nested	0.1105ms	71.0869μs	14.0673 KOps/s	14.0431 KOps/s	$\color{#35bf28}+0.17\%$
test_keys_stack_nested_leaf	91.2110μs	64.4255μs	15.5218 KOps/s	15.7982 KOps/s	$\color{#d91a1a}-1.75\%$
test_keys_stack_nested_locked	0.1096ms	77.8614μs	12.8433 KOps/s	12.8803 KOps/s	$\color{#d91a1a}-0.29\%$
test_values	5.1150μs	0.8373μs	1.1944 MOps/s	1.1734 MOps/s	$\color{#35bf28}+1.79\%$
test_values_nested	89.9500μs	49.0904μs	20.3706 KOps/s	20.4056 KOps/s	$\color{#d91a1a}-0.17\%$
test_values_nested_locked	84.7900μs	50.9008μs	19.6460 KOps/s	19.8632 KOps/s	$\color{#d91a1a}-1.09\%$
test_values_nested_leaf	70.5010μs	42.8154μs	23.3561 KOps/s	23.4080 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_stack_nested	95.2500μs	50.7651μs	19.6986 KOps/s	19.9352 KOps/s	$\color{#d91a1a}-1.19\%$
test_values_stack_nested_leaf	79.0400μs	44.2564μs	22.5956 KOps/s	22.9851 KOps/s	$\color{#d91a1a}-1.69\%$
test_values_stack_nested_locked	79.1810μs	52.1548μs	19.1737 KOps/s	19.4990 KOps/s	$\color{#d91a1a}-1.67\%$
test_membership	2.0945μs	0.5021μs	1.9917 MOps/s	1.9896 MOps/s	$\color{#35bf28}+0.11\%$
test_membership_nested	19.6900μs	1.8538μs	539.4205 KOps/s	524.0761 KOps/s	$\color{#35bf28}+2.93\%$
test_membership_nested_leaf	18.4603μs	1.8281μs	547.0020 KOps/s	553.0605 KOps/s	$\color{#d91a1a}-1.10\%$
test_membership_stacked_nested	16.4610μs	1.9054μs	524.8125 KOps/s	526.3816 KOps/s	$\color{#d91a1a}-0.30\%$
test_membership_stacked_nested_leaf	25.9400μs	1.9251μs	519.4411 KOps/s	520.3190 KOps/s	$\color{#d91a1a}-0.17\%$
test_membership_nested_last	29.9000μs	2.9446μs	339.6010 KOps/s	335.0828 KOps/s	$\color{#35bf28}+1.35\%$
test_membership_nested_leaf_last	34.7100μs	2.9578μs	338.0859 KOps/s	332.2136 KOps/s	$\color{#35bf28}+1.77\%$
test_membership_stacked_nested_last	30.4700μs	2.9335μs	340.8888 KOps/s	263.8504 KOps/s	$\textbf{\color{#35bf28}+29.20\%}$
test_membership_stacked_nested_leaf_last	30.8210μs	2.9597μs	337.8752 KOps/s	267.7475 KOps/s	$\textbf{\color{#35bf28}+26.19\%}$
test_nested_getleaf	28.4400μs	6.1052μs	163.7942 KOps/s	165.8281 KOps/s	$\color{#d91a1a}-1.23\%$
test_nested_get	36.1100μs	5.7628μs	173.5255 KOps/s	175.3144 KOps/s	$\color{#d91a1a}-1.02\%$
test_stacked_getleaf	31.2200μs	5.9871μs	167.0271 KOps/s	165.9284 KOps/s	$\color{#35bf28}+0.66\%$
test_stacked_get	33.4010μs	5.5842μs	179.0753 KOps/s	177.5318 KOps/s	$\color{#35bf28}+0.87\%$
test_nested_getitemleaf	27.5810μs	6.1545μs	162.4827 KOps/s	163.6222 KOps/s	$\color{#d91a1a}-0.70\%$
test_nested_getitem	35.7300μs	5.7886μs	172.7541 KOps/s	174.0278 KOps/s	$\color{#d91a1a}-0.73\%$
test_stacked_getitemleaf	36.9400μs	6.1335μs	163.0386 KOps/s	165.0611 KOps/s	$\color{#d91a1a}-1.23\%$
test_stacked_getitem	30.4900μs	5.7489μs	173.9466 KOps/s	174.6411 KOps/s	$\color{#d91a1a}-0.40\%$
test_lock_nested	0.8723ms	0.4336ms	2.3061 KOps/s	2.3084 KOps/s	$\color{#d91a1a}-0.10\%$
test_lock_stack_nested	0.4480ms	0.3960ms	2.5250 KOps/s	2.5353 KOps/s	$\color{#d91a1a}-0.41\%$
test_unlock_nested	0.8427ms	0.3739ms	2.6742 KOps/s	2.7107 KOps/s	$\color{#d91a1a}-1.35\%$
test_unlock_stack_nested	0.4002ms	0.3356ms	2.9799 KOps/s	2.9997 KOps/s	$\color{#d91a1a}-0.66\%$
test_flatten_speed	0.1548ms	76.6394μs	13.0481 KOps/s	13.0112 KOps/s	$\color{#35bf28}+0.28\%$
test_unflatten_speed	0.3661ms	0.3246ms	3.0811 KOps/s	3.1332 KOps/s	$\color{#d91a1a}-1.66\%$
test_common_ops	1.7620ms	1.3077ms	764.7159 Ops/s	771.9915 Ops/s	$\color{#d91a1a}-0.94\%$
test_creation	27.7300μs	1.5014μs	666.0490 KOps/s	668.7669 KOps/s	$\color{#d91a1a}-0.41\%$
test_creation_empty	45.6000μs	16.1936μs	61.7529 KOps/s	64.0031 KOps/s	$\color{#d91a1a}-3.52\%$
test_creation_nested_1	47.2100μs	18.1614μs	55.0620 KOps/s	58.3236 KOps/s	$\textbf{\color{#d91a1a}-5.59\%}$
test_creation_nested_2	50.0400μs	20.9522μs	47.7277 KOps/s	50.4056 KOps/s	$\textbf{\color{#d91a1a}-5.31\%}$
test_clone	0.1078ms	30.3450μs	32.9544 KOps/s	33.4907 KOps/s	$\color{#d91a1a}-1.60\%$
test_getitem[int]	1.2715ms	15.8808μs	62.9691 KOps/s	61.9973 KOps/s	$\color{#35bf28}+1.57\%$
test_getitem[slice_int]	0.1167ms	27.4818μs	36.3878 KOps/s	36.2327 KOps/s	$\color{#35bf28}+0.43\%$
test_getitem[range]	0.1565ms	0.1101ms	9.0799 KOps/s	8.9529 KOps/s	$\color{#35bf28}+1.42\%$
test_getitem[tuple]	0.1161ms	24.1641μs	41.3837 KOps/s	40.6976 KOps/s	$\color{#35bf28}+1.69\%$
test_getitem[list]	0.2052ms	0.1011ms	9.8903 KOps/s	9.9140 KOps/s	$\color{#d91a1a}-0.24\%$
test_setitem_dim[int]	72.6110μs	45.9547μs	21.7606 KOps/s	21.8131 KOps/s	$\color{#d91a1a}-0.24\%$
test_setitem_dim[slice_int]	90.8410μs	67.6977μs	14.7716 KOps/s	14.6474 KOps/s	$\color{#35bf28}+0.85\%$
test_setitem_dim[range]	0.1586ms	0.1300ms	7.6915 KOps/s	7.7157 KOps/s	$\color{#d91a1a}-0.31\%$
test_setitem_dim[tuple]	0.1023ms	61.8220μs	16.1755 KOps/s	16.0679 KOps/s	$\color{#35bf28}+0.67\%$
test_setitem	90.4100μs	44.5168μs	22.4634 KOps/s	23.2550 KOps/s	$\color{#d91a1a}-3.40\%$
test_set	88.7210μs	47.0298μs	21.2631 KOps/s	24.1544 KOps/s	$\textbf{\color{#d91a1a}-11.97\%}$
test_set_shared	0.3858ms	60.0430μs	16.6547 KOps/s	18.2958 KOps/s	$\textbf{\color{#d91a1a}-8.97\%}$
test_update	95.1600μs	56.3266μs	17.7536 KOps/s	19.5420 KOps/s	$\textbf{\color{#d91a1a}-9.15\%}$
test_update_nested	0.1045ms	64.3558μs	15.5386 KOps/s	17.0307 KOps/s	$\textbf{\color{#d91a1a}-8.76\%}$
test_update__nested	0.1871ms	63.1737μs	15.8294 KOps/s	15.8930 KOps/s	$\color{#d91a1a}-0.40\%$
test_set_nested	97.5310μs	45.2791μs	22.0852 KOps/s	22.3337 KOps/s	$\color{#d91a1a}-1.11\%$
test_set_nested_new	99.6510μs	49.6334μs	20.1477 KOps/s	20.9941 KOps/s	$\color{#d91a1a}-4.03\%$
test_select	0.1033ms	62.5070μs	15.9982 KOps/s	16.4412 KOps/s	$\color{#d91a1a}-2.69\%$
test_select_nested	78.6900μs	41.6091μs	24.0332 KOps/s	23.7177 KOps/s	$\color{#35bf28}+1.33\%$
test_exclude_nested	0.5184ms	60.7567μs	16.4591 KOps/s	17.1078 KOps/s	$\color{#d91a1a}-3.79\%$
test_empty[True]	0.2988ms	0.2594ms	3.8545 KOps/s	3.8590 KOps/s	$\color{#d91a1a}-0.12\%$
test_empty[False]	3.6480μs	0.8437μs	1.1853 MOps/s	1.3587 MOps/s	$\textbf{\color{#d91a1a}-12.76\%}$
test_to	56.7500μs	27.6397μs	36.1798 KOps/s	37.6510 KOps/s	$\color{#d91a1a}-3.91\%$
test_to_nonblocking	56.8800μs	26.9051μs	37.1677 KOps/s	40.8404 KOps/s	$\textbf{\color{#d91a1a}-8.99\%}$
test_unbind_speed	1.2464ms	0.2854ms	3.5039 KOps/s	3.5394 KOps/s	$\color{#d91a1a}-1.00\%$
test_unbind_speed_stack0	0.3210ms	0.2799ms	3.5727 KOps/s	3.5468 KOps/s	$\color{#35bf28}+0.73\%$
test_unbind_speed_stack1	92.4051ms	0.7214ms	1.3861 KOps/s	1.4032 KOps/s	$\color{#d91a1a}-1.22\%$
test_split	93.6898ms	2.1896ms	456.7096 Ops/s	456.7079 Ops/s	$+0.00\%$
test_chunk	95.7193ms	2.1716ms	460.4883 Ops/s	449.1022 Ops/s	$\color{#35bf28}+2.54\%$
test_creation[device0]	0.3299ms	0.1283ms	7.7955 KOps/s	7.3847 KOps/s	$\textbf{\color{#35bf28}+5.56\%}$
test_creation_from_tensor	0.3534ms	0.1297ms	7.7075 KOps/s	7.3874 KOps/s	$\color{#35bf28}+4.33\%$
test_add_one[memmap_tensor0]	0.2947ms	9.1390μs	109.4206 KOps/s	112.3311 KOps/s	$\color{#d91a1a}-2.59\%$
test_contiguous[memmap_tensor0]	29.9600μs	2.1996μs	454.6291 KOps/s	446.2509 KOps/s	$\color{#35bf28}+1.88\%$
test_stack[memmap_tensor0]	34.7400μs	6.9111μs	144.6940 KOps/s	147.8753 KOps/s	$\color{#d91a1a}-2.15\%$
test_memmaptd_index	1.1135ms	0.4357ms	2.2951 KOps/s	2.2887 KOps/s	$\color{#35bf28}+0.28\%$
test_memmaptd_index_astensor	0.7584ms	0.5087ms	1.9656 KOps/s	1.9775 KOps/s	$\color{#d91a1a}-0.60\%$
test_memmaptd_index_op	1.4869ms	1.0628ms	940.9472 Ops/s	957.4025 Ops/s	$\color{#d91a1a}-1.72\%$
test_serialize_model	0.1320s	0.1302s	7.6818 Ops/s	7.6702 Ops/s	$\color{#35bf28}+0.15\%$
test_serialize_model_pickle	1.3477s	1.2149s	0.8231 Ops/s	0.8217 Ops/s	$\color{#35bf28}+0.17\%$
test_serialize_weights	0.1304s	0.1298s	7.7024 Ops/s	7.7090 Ops/s	$\color{#d91a1a}-0.09\%$
test_serialize_weights_returnearly	0.2338s	63.8411ms	15.6639 Ops/s	21.7070 Ops/s	$\textbf{\color{#d91a1a}-27.84\%}$
test_serialize_weights_pickle	1.3464s	1.1869s	0.8426 Ops/s	0.8212 Ops/s	$\color{#35bf28}+2.60\%$
test_reshape_pytree	68.9510μs	35.9509μs	27.8157 KOps/s	27.2994 KOps/s	$\color{#35bf28}+1.89\%$
test_reshape_td	64.7700μs	41.6206μs	24.0266 KOps/s	23.5485 KOps/s	$\color{#35bf28}+2.03\%$
test_view_pytree	68.1410μs	35.4729μs	28.1905 KOps/s	27.7521 KOps/s	$\color{#35bf28}+1.58\%$
test_view_td	87.3110μs	47.9377μs	20.8604 KOps/s	21.5910 KOps/s	$\color{#d91a1a}-3.38\%$
test_unbind_pytree	62.5610μs	33.5538μs	29.8029 KOps/s	28.9633 KOps/s	$\color{#35bf28}+2.90\%$
test_unbind_td	0.4897ms	43.8448μs	22.8077 KOps/s	22.6734 KOps/s	$\color{#35bf28}+0.59\%$
test_split_pytree	73.4500μs	44.9214μs	22.2611 KOps/s	21.8305 KOps/s	$\color{#35bf28}+1.97\%$
test_split_td	0.6885ms	55.7707μs	17.9306 KOps/s	17.8082 KOps/s	$\color{#35bf28}+0.69\%$
test_add_pytree	0.1034ms	58.9430μs	16.9655 KOps/s	17.1917 KOps/s	$\color{#d91a1a}-1.32\%$
test_add_td	0.1348ms	96.8157μs	10.3289 KOps/s	10.5965 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_add_one_nested[tensordict-compile]	0.2103ms	0.1615ms	6.1908 KOps/s	6.1999 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_add_one_nested[tensordict-eager]	0.3412ms	0.1615ms	6.1909 KOps/s	6.1510 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_add_one_nested[pytree-compile]	0.2078ms	0.1553ms	6.4382 KOps/s	6.4377 KOps/s	$+0.01\%$
test_compile_add_one_nested[pytree-eager]	0.5916ms	0.1941ms	5.1510 KOps/s	5.3417 KOps/s	$\color{#d91a1a}-3.57\%$
test_compile_copy_nested[tensordict-compile]	0.4134ms	21.9716μs	45.5132 KOps/s	46.5544 KOps/s	$\color{#d91a1a}-2.24\%$
test_compile_copy_nested[tensordict-eager]	0.4353ms	49.5398μs	20.1858 KOps/s	20.7673 KOps/s	$\color{#d91a1a}-2.80\%$
test_compile_copy_nested[pytree-compile]	0.4489ms	65.4155μs	15.2869 KOps/s	15.4423 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_copy_nested[pytree-eager]	0.4339ms	49.5775μs	20.1705 KOps/s	20.0369 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_add_one_flat[tensordict-compile]	0.3693ms	0.3211ms	3.1142 KOps/s	3.1267 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_add_one_flat[tensordict-eager]	0.6151ms	0.2325ms	4.3016 KOps/s	4.2608 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_add_one_flat[tensorclass-compile]	0.1834ms	0.1289ms	7.7551 KOps/s	7.7670 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_add_one_flat[tensorclass-eager]	0.4560ms	66.6373μs	15.0066 KOps/s	14.8355 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_add_one_flat[pytree-compile]	0.7377ms	0.3302ms	3.0280 KOps/s	3.0441 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_add_one_flat[pytree-eager]	1.0559ms	0.6605ms	1.5139 KOps/s	1.5582 KOps/s	$\color{#d91a1a}-2.84\%$
test_compile_add_self_flat[tensordict-eager]	0.6944ms	0.2854ms	3.5037 KOps/s	3.5109 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_add_self_flat[tensordict-compile]	0.3986ms	0.3242ms	3.0846 KOps/s	3.1045 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_add_self_flat[tensorclass-eager]	0.4973ms	78.1709μs	12.7925 KOps/s	12.6191 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_add_self_flat[tensorclass-compile]	0.1934ms	0.1318ms	7.5890 KOps/s	7.5102 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_add_self_flat[pytree-eager]	0.6651ms	0.5569ms	1.7956 KOps/s	1.8717 KOps/s	$\color{#d91a1a}-4.06\%$
test_compile_add_self_flat[pytree-compile]	0.3799ms	0.3279ms	3.0501 KOps/s	3.0493 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_copy_flat[tensordict-compile]	52.2000μs	21.1580μs	47.2634 KOps/s	46.1194 KOps/s	$\color{#35bf28}+2.48\%$
test_compile_copy_flat[tensordict-eager]	94.0310μs	38.8919μs	25.7123 KOps/s	26.0708 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_copy_flat[pytree-compile]	0.1142ms	70.7920μs	14.1259 KOps/s	14.2819 KOps/s	$\color{#d91a1a}-1.09\%$
test_compile_copy_flat[pytree-eager]	86.8700μs	52.0999μs	19.1939 KOps/s	19.3094 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_assign_and_add[tensordict-compile]	2.3703ms	0.8291ms	1.2061 KOps/s	1.1142 KOps/s	$\textbf{\color{#35bf28}+8.25\%}$
test_compile_assign_and_add[tensordict-eager]	3.3489ms	3.2448ms	308.1824 Ops/s	301.8191 Ops/s	$\color{#35bf28}+2.11\%$
test_compile_assign_and_add[pytree-compile]	2.3956ms	0.8404ms	1.1899 KOps/s	1.0914 KOps/s	$\textbf{\color{#35bf28}+9.03\%}$
test_compile_assign_and_add[pytree-eager]	3.4542ms	3.3412ms	299.2960 Ops/s	302.1734 Ops/s	$\color{#d91a1a}-0.95\%$
test_compile_indexing[tensor-tensordict-compile]	0.2623ms	0.1207ms	8.2878 KOps/s	8.4021 KOps/s	$\color{#d91a1a}-1.36\%$
test_compile_indexing[tensor-tensordict-eager]	0.1983ms	64.7590μs	15.4419 KOps/s	15.2296 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1604ms	0.1154ms	8.6663 KOps/s	8.3779 KOps/s	$\color{#35bf28}+3.44\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1087ms	47.0703μs	21.2448 KOps/s	20.5564 KOps/s	$\color{#35bf28}+3.35\%$
test_compile_indexing[tensor-pytree-compile]	0.1605ms	0.1189ms	8.4136 KOps/s	8.3278 KOps/s	$\color{#35bf28}+1.03\%$
test_compile_indexing[tensor-pytree-eager]	91.9010μs	46.8301μs	21.3538 KOps/s	20.6768 KOps/s	$\color{#35bf28}+3.27\%$
test_compile_indexing[slice-tensordict-compile]	0.1898ms	0.1482ms	6.7498 KOps/s	6.8547 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_indexing[slice-tensordict-eager]	0.1505ms	24.2609μs	41.2186 KOps/s	39.7948 KOps/s	$\color{#35bf28}+3.58\%$
test_compile_indexing[slice-tensorclass-compile]	0.1910ms	0.1436ms	6.9628 KOps/s	7.1399 KOps/s	$\color{#d91a1a}-2.48\%$
test_compile_indexing[slice-tensorclass-eager]	60.0510μs	20.9214μs	47.7981 KOps/s	46.9555 KOps/s	$\color{#35bf28}+1.79\%$
test_compile_indexing[slice-pytree-compile]	0.2016ms	0.1452ms	6.8862 KOps/s	7.0965 KOps/s	$\color{#d91a1a}-2.96\%$
test_compile_indexing[slice-pytree-eager]	66.1210μs	20.6572μs	48.4094 KOps/s	47.6845 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_indexing[int-tensordict-compile]	0.2798ms	0.1506ms	6.6421 KOps/s	6.7805 KOps/s	$\color{#d91a1a}-2.04\%$
test_compile_indexing[int-tensordict-eager]	0.5104ms	24.0576μs	41.5669 KOps/s	40.4454 KOps/s	$\color{#35bf28}+2.77\%$
test_compile_indexing[int-tensorclass-compile]	0.1908ms	0.1458ms	6.8570 KOps/s	6.7937 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_indexing[int-tensorclass-eager]	52.3010μs	20.7712μs	48.1435 KOps/s	48.1111 KOps/s	$\color{#35bf28}+0.07\%$
test_compile_indexing[int-pytree-compile]	0.1875ms	0.1410ms	7.0915 KOps/s	6.8702 KOps/s	$\color{#35bf28}+3.22\%$
test_compile_indexing[int-pytree-eager]	54.1600μs	20.6073μs	48.5264 KOps/s	45.3173 KOps/s	$\textbf{\color{#35bf28}+7.08\%}$
test_mod_add[eager]	75.4110μs	34.0144μs	29.3993 KOps/s	29.8975 KOps/s	$\color{#d91a1a}-1.67\%$
test_mod_add[compile]	0.1300ms	83.9008μs	11.9188 KOps/s	12.3315 KOps/s	$\color{#d91a1a}-3.35\%$
test_mod_add[compile-overhead]	0.3086ms	0.1541ms	6.4910 KOps/s	6.3617 KOps/s	$\color{#35bf28}+2.03\%$
test_mod_wrap[eager]	0.3678ms	0.2459ms	4.0664 KOps/s	3.9926 KOps/s	$\color{#35bf28}+1.85\%$
test_mod_wrap[compile]	1.4557ms	0.3086ms	3.2405 KOps/s	3.2780 KOps/s	$\color{#d91a1a}-1.14\%$
test_mod_wrap[compile-overhead]	7.7718ms	4.0832ms	244.9030 Ops/s	244.4356 Ops/s	$\color{#35bf28}+0.19\%$
test_mod_wrap_and_backward[eager]	1.7722ms	1.3872ms	720.8765 Ops/s	677.7776 Ops/s	$\textbf{\color{#35bf28}+6.36\%}$
test_mod_wrap_and_backward[compile]	1.7679ms	1.3409ms	745.7887 Ops/s	687.8552 Ops/s	$\textbf{\color{#35bf28}+8.42\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3393ms	0.9089ms	1.1003 KOps/s	986.1842 Ops/s	$\textbf{\color{#35bf28}+11.57\%}$
test_seq_add[eager]	0.1702ms	0.1018ms	9.8186 KOps/s	10.0245 KOps/s	$\color{#d91a1a}-2.05\%$
test_seq_add[compile]	0.1392ms	91.0130μs	10.9874 KOps/s	10.9788 KOps/s	$\color{#35bf28}+0.08\%$
test_seq_add[compile-overhead]	0.1685ms	0.1245ms	8.0337 KOps/s	8.0065 KOps/s	$\color{#35bf28}+0.34\%$
test_seq_wrap[eager]	0.4517ms	0.3832ms	2.6099 KOps/s	2.4080 KOps/s	$\textbf{\color{#35bf28}+8.38\%}$
test_seq_wrap[compile]	0.3635ms	0.3150ms	3.1750 KOps/s	3.0712 KOps/s	$\color{#35bf28}+3.38\%$
test_seq_wrap[compile-overhead]	0.2637ms	0.2195ms	4.5561 KOps/s	4.4608 KOps/s	$\color{#35bf28}+2.14\%$
test_func_call_runtime[False-eager]	0.8499ms	0.7504ms	1.3326 KOps/s	1.3174 KOps/s	$\color{#35bf28}+1.15\%$
test_func_call_runtime[False-compile]	0.9698ms	0.7881ms	1.2689 KOps/s	1.2396 KOps/s	$\color{#35bf28}+2.36\%$
test_func_call_runtime[False-compile-overhead]	0.4178ms	0.3629ms	2.7554 KOps/s	2.7390 KOps/s	$\color{#35bf28}+0.60\%$
test_func_call_runtime[True-eager]	0.9675ms	0.9067ms	1.1029 KOps/s	1.0766 KOps/s	$\color{#35bf28}+2.45\%$
test_func_call_runtime[True-compile]	0.9460ms	0.8134ms	1.2294 KOps/s	1.1638 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_func_call_runtime[True-compile-overhead]	0.4321ms	0.3847ms	2.5992 KOps/s	2.6032 KOps/s	$\color{#d91a1a}-0.15\%$
test_func_call_cm_runtime[False-eager]	0.8175ms	0.7386ms	1.3538 KOps/s	1.3133 KOps/s	$\color{#35bf28}+3.09\%$
test_func_call_cm_runtime[False-compile]	0.8833ms	0.7911ms	1.2641 KOps/s	1.2408 KOps/s	$\color{#35bf28}+1.88\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4141ms	0.3637ms	2.7497 KOps/s	2.7413 KOps/s	$\color{#35bf28}+0.31\%$
test_func_call_cm_runtime[True-eager]	1.1194ms	1.0165ms	983.7746 Ops/s	959.3121 Ops/s	$\color{#35bf28}+2.55\%$
test_func_call_cm_runtime[True-compile]	0.8951ms	0.8382ms	1.1931 KOps/s	1.1559 KOps/s	$\color{#35bf28}+3.22\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4557ms	0.4093ms	2.4435 KOps/s	2.4247 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_func_call_cm_runtime[eager]	2.5795ms	2.1220ms	471.2546 Ops/s	463.3930 Ops/s	$\color{#35bf28}+1.70\%$
test_vmap_func_call_cm_runtime[compile]	0.9134ms	0.8552ms	1.1693 KOps/s	1.1600 KOps/s	$\color{#35bf28}+0.80\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4682ms	0.4137ms	2.4174 KOps/s	2.4096 KOps/s	$\color{#35bf28}+0.32\%$
test_distributed	0.8823ms	0.1581ms	6.3264 KOps/s	8.8268 KOps/s	$\textbf{\color{#d91a1a}-28.33\%}$
test_tdmodule	35.0110μs	14.8771μs	67.2174 KOps/s	64.6674 KOps/s	$\color{#35bf28}+3.94\%$
test_tdmodule_dispatch	49.6900μs	29.4736μs	33.9287 KOps/s	32.4082 KOps/s	$\color{#35bf28}+4.69\%$
test_tdseq	37.7400μs	16.2875μs	61.3967 KOps/s	62.7185 KOps/s	$\color{#d91a1a}-2.11\%$
test_tdseq_dispatch	55.1910μs	32.6905μs	30.5900 KOps/s	31.2895 KOps/s	$\color{#d91a1a}-2.24\%$
test_instantiation_functorch	2.0362ms	1.8941ms	527.9653 Ops/s	522.6955 Ops/s	$\color{#35bf28}+1.01\%$
test_exec_functorch	0.2582ms	0.2120ms	4.7170 KOps/s	4.6857 KOps/s	$\color{#35bf28}+0.67\%$
test_exec_functional_call	0.2894ms	0.2075ms	4.8193 KOps/s	4.6728 KOps/s	$\color{#35bf28}+3.14\%$
test_exec_td_decorator	0.4476ms	0.2629ms	3.8039 KOps/s	3.7677 KOps/s	$\color{#35bf28}+0.96\%$
test_vmap_mlp_speed_decorator[True-True]	0.8220ms	0.6920ms	1.4451 KOps/s	1.4228 KOps/s	$\color{#35bf28}+1.57\%$
test_vmap_mlp_speed_decorator[True-False]	0.8375ms	0.6951ms	1.4386 KOps/s	1.3882 KOps/s	$\color{#35bf28}+3.63\%$
test_vmap_mlp_speed_decorator[False-True]	0.7909ms	0.6287ms	1.5906 KOps/s	1.5921 KOps/s	$\color{#d91a1a}-0.10\%$
test_vmap_mlp_speed_decorator[False-False]	0.7496ms	0.6189ms	1.6157 KOps/s	1.6178 KOps/s	$\color{#d91a1a}-0.13\%$
test_vmap_transformer_speed_decorator[True-True]	20.6934ms	19.9301ms	50.1752 Ops/s	49.9762 Ops/s	$\color{#35bf28}+0.40\%$
test_vmap_transformer_speed_decorator[True-False]	20.1204ms	19.9235ms	50.1919 Ops/s	49.8920 Ops/s	$\color{#35bf28}+0.60\%$
test_vmap_transformer_speed_decorator[False-True]	20.6135ms	19.8065ms	50.4884 Ops/s	50.2952 Ops/s	$\color{#35bf28}+0.38\%$
test_vmap_transformer_speed_decorator[False-False]	20.1576ms	19.8101ms	50.4794 Ops/s	50.2582 Ops/s	$\color{#35bf28}+0.44\%$
test_to_module_speed[True]	1.3865ms	0.9945ms	1.0055 KOps/s	981.5413 Ops/s	$\color{#35bf28}+2.44\%$
test_to_module_speed[False]	1.4250ms	0.9870ms	1.0131 KOps/s	1.0040 KOps/s	$\color{#35bf28}+0.91\%$
test_tc_init	71.3310μs	35.4989μs	28.1699 KOps/s	28.5094 KOps/s	$\color{#d91a1a}-1.19\%$
test_tc_init_nested	0.1119ms	70.3091μs	14.2229 KOps/s	14.4859 KOps/s	$\color{#d91a1a}-1.82\%$
test_tc_first_layer_tensor	5.6757μs	0.6759μs	1.4796 MOps/s	1.4945 MOps/s	$\color{#d91a1a}-1.00\%$
test_tc_first_layer_nontensor	31.8300μs	2.2672μs	441.0716 KOps/s	451.2286 KOps/s	$\color{#d91a1a}-2.25\%$
test_tc_second_layer_tensor	7.1275μs	1.3699μs	729.9750 KOps/s	723.9189 KOps/s	$\color{#35bf28}+0.84\%$
test_tc_second_layer_nontensor	27.9900μs	2.9766μs	335.9504 KOps/s	337.3669 KOps/s	$\color{#d91a1a}-0.42\%$
test_unbind	0.1913s	9.6159ms	103.9942 Ops/s	92.4687 Ops/s	$\textbf{\color{#35bf28}+12.46\%}$
test_full_like	0.6570ms	0.5744ms	1.7410 KOps/s	1.7429 KOps/s	$\color{#d91a1a}-0.11\%$
test_zeros_like	0.2611ms	0.1980ms	5.0511 KOps/s	5.0481 KOps/s	$\color{#35bf28}+0.06\%$
test_ones_like	0.2421ms	0.1977ms	5.0569 KOps/s	5.0530 KOps/s	$\color{#35bf28}+0.08\%$
test_clone	0.4489ms	0.4148ms	2.4106 KOps/s	2.4096 KOps/s	$\color{#35bf28}+0.04\%$
test_squeeze	39.7200μs	9.7707μs	102.3469 KOps/s	99.7941 KOps/s	$\color{#35bf28}+2.56\%$
test_unsqueeze	0.2202ms	73.9952μs	13.5144 KOps/s	13.3027 KOps/s	$\color{#35bf28}+1.59\%$
test_split	0.3929ms	0.1578ms	6.3364 KOps/s	6.2583 KOps/s	$\color{#35bf28}+1.25\%$
test_permute	0.2327ms	0.1875ms	5.3331 KOps/s	5.6208 KOps/s	$\textbf{\color{#d91a1a}-5.12\%}$
test_stack	1.2855ms	0.8591ms	1.1640 KOps/s	1.1712 KOps/s	$\color{#d91a1a}-0.62\%$
test_cat	1.2556ms	1.2313ms	812.1371 Ops/s	811.9746 Ops/s	$\color{#35bf28}+0.02\%$