[Feature] dist_params_keys and dist_sample_keys

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] dist_params_keys and dist_sample_keys #1179

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}16$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	77.0240μs	19.5306μs	51.2018 KOps/s	51.0998 KOps/s	$\color{#35bf28}+0.20\%$
test_plain_set_stack_nested	65.3610μs	20.2114μs	49.4769 KOps/s	50.3589 KOps/s	$\color{#d91a1a}-1.75\%$
test_plain_set_nested_inplace	74.2380μs	21.4686μs	46.5797 KOps/s	46.1404 KOps/s	$\color{#35bf28}+0.95\%$
test_plain_set_stack_nested_inplace	60.3930μs	21.7115μs	46.0585 KOps/s	46.6502 KOps/s	$\color{#d91a1a}-1.27\%$
test_items	54.5210μs	4.1163μs	242.9366 KOps/s	234.2485 KOps/s	$\color{#35bf28}+3.71\%$
test_items_nested	0.5375ms	0.3956ms	2.5276 KOps/s	2.4140 KOps/s	$\color{#35bf28}+4.71\%$
test_items_nested_locked	0.9442ms	0.3945ms	2.5350 KOps/s	2.4254 KOps/s	$\color{#35bf28}+4.52\%$
test_items_nested_leaf	0.1438ms	76.4101μs	13.0873 KOps/s	13.0330 KOps/s	$\color{#35bf28}+0.42\%$
test_items_stack_nested	0.5783ms	0.3971ms	2.5183 KOps/s	2.4033 KOps/s	$\color{#35bf28}+4.78\%$
test_items_stack_nested_leaf	0.1505ms	77.7467μs	12.8623 KOps/s	12.6070 KOps/s	$\color{#35bf28}+2.02\%$
test_items_stack_nested_locked	0.5214ms	0.3928ms	2.5456 KOps/s	2.3518 KOps/s	$\textbf{\color{#35bf28}+8.24\%}$
test_keys	30.7870μs	3.4731μs	287.9246 KOps/s	286.3286 KOps/s	$\color{#35bf28}+0.56\%$
test_keys_nested	0.2635ms	0.1635ms	6.1145 KOps/s	5.9744 KOps/s	$\color{#35bf28}+2.35\%$
test_keys_nested_locked	2.1645ms	0.1714ms	5.8344 KOps/s	5.8088 KOps/s	$\color{#35bf28}+0.44\%$
test_keys_nested_leaf	0.2609ms	0.1460ms	6.8490 KOps/s	6.8259 KOps/s	$\color{#35bf28}+0.34\%$
test_keys_stack_nested	0.2705ms	0.1646ms	6.0770 KOps/s	6.0050 KOps/s	$\color{#35bf28}+1.20\%$
test_keys_stack_nested_leaf	0.2276ms	0.1450ms	6.8950 KOps/s	7.0201 KOps/s	$\color{#d91a1a}-1.78\%$
test_keys_stack_nested_locked	0.3177ms	0.1717ms	5.8257 KOps/s	5.8286 KOps/s	$\color{#d91a1a}-0.05\%$
test_values	9.1370μs	1.1260μs	888.0937 KOps/s	945.0227 KOps/s	$\textbf{\color{#d91a1a}-6.02\%}$
test_values_nested	0.3671ms	65.5121μs	15.2644 KOps/s	16.1774 KOps/s	$\textbf{\color{#d91a1a}-5.64\%}$
test_values_nested_locked	0.1247ms	62.1462μs	16.0911 KOps/s	15.9337 KOps/s	$\color{#35bf28}+0.99\%$
test_values_nested_leaf	0.1476ms	71.3493μs	14.0156 KOps/s	13.8993 KOps/s	$\color{#35bf28}+0.84\%$
test_values_stack_nested	0.1283ms	62.7663μs	15.9321 KOps/s	15.6357 KOps/s	$\color{#35bf28}+1.90\%$
test_values_stack_nested_leaf	0.1649ms	71.3454μs	14.0163 KOps/s	14.1682 KOps/s	$\color{#d91a1a}-1.07\%$
test_values_stack_nested_locked	0.1281ms	62.9489μs	15.8859 KOps/s	15.3892 KOps/s	$\color{#35bf28}+3.23\%$
test_membership	31.3180μs	0.8638μs	1.1576 MOps/s	1.1760 MOps/s	$\color{#d91a1a}-1.56\%$
test_membership_nested	27.6310μs	2.8896μs	346.0742 KOps/s	335.1465 KOps/s	$\color{#35bf28}+3.26\%$
test_membership_nested_leaf	0.1043ms	2.8911μs	345.8884 KOps/s	331.8641 KOps/s	$\color{#35bf28}+4.23\%$
test_membership_stacked_nested	29.4650μs	2.9037μs	344.3835 KOps/s	331.6313 KOps/s	$\color{#35bf28}+3.85\%$
test_membership_stacked_nested_leaf	63.5880μs	2.8773μs	347.5526 KOps/s	337.3847 KOps/s	$\color{#35bf28}+3.01\%$
test_membership_nested_last	36.3680μs	4.4379μs	225.3323 KOps/s	226.7926 KOps/s	$\color{#d91a1a}-0.64\%$
test_membership_nested_leaf_last	30.5870μs	4.4701μs	223.7081 KOps/s	223.9099 KOps/s	$\color{#d91a1a}-0.09\%$
test_membership_stacked_nested_last	50.6540μs	4.4134μs	226.5841 KOps/s	159.0741 KOps/s	$\textbf{\color{#35bf28}+42.44\%}$
test_membership_stacked_nested_leaf_last	49.4920μs	4.4817μs	223.1277 KOps/s	162.0865 KOps/s	$\textbf{\color{#35bf28}+37.66\%}$
test_nested_getleaf	53.9900μs	10.8734μs	91.9678 KOps/s	90.1668 KOps/s	$\color{#35bf28}+2.00\%$
test_nested_get	62.2360μs	10.4157μs	96.0087 KOps/s	94.9724 KOps/s	$\color{#35bf28}+1.09\%$
test_stacked_getleaf	36.2480μs	10.8775μs	91.9327 KOps/s	91.7479 KOps/s	$\color{#35bf28}+0.20\%$
test_stacked_get	64.1870μs	10.0369μs	99.6326 KOps/s	96.5908 KOps/s	$\color{#35bf28}+3.15\%$
test_nested_getitemleaf	33.3730μs	11.3373μs	88.2041 KOps/s	89.1791 KOps/s	$\color{#d91a1a}-1.09\%$
test_nested_getitem	0.1323ms	11.0002μs	90.9074 KOps/s	92.9425 KOps/s	$\color{#d91a1a}-2.19\%$
test_stacked_getitemleaf	84.4070μs	11.7769μs	84.9118 KOps/s	89.4261 KOps/s	$\textbf{\color{#d91a1a}-5.05\%}$
test_stacked_getitem	38.5010μs	10.7728μs	92.8263 KOps/s	93.3921 KOps/s	$\color{#d91a1a}-0.61\%$
test_lock_nested	4.8555ms	0.4712ms	2.1221 KOps/s	1.6904 KOps/s	$\textbf{\color{#35bf28}+25.54\%}$
test_lock_stack_nested	0.7330ms	0.4334ms	2.3075 KOps/s	2.3473 KOps/s	$\color{#d91a1a}-1.69\%$
test_unlock_nested	0.9661ms	0.3866ms	2.5866 KOps/s	2.5966 KOps/s	$\color{#d91a1a}-0.39\%$
test_unlock_stack_nested	0.6330ms	0.3527ms	2.8356 KOps/s	2.9214 KOps/s	$\color{#d91a1a}-2.94\%$
test_flatten_speed	0.2014ms	99.6332μs	10.0368 KOps/s	10.0252 KOps/s	$\color{#35bf28}+0.12\%$
test_unflatten_speed	0.6659ms	0.5256ms	1.9026 KOps/s	1.9091 KOps/s	$\color{#d91a1a}-0.34\%$
test_common_ops	5.8964ms	0.7744ms	1.2912 KOps/s	1.2647 KOps/s	$\color{#35bf28}+2.10\%$
test_creation	62.8170μs	2.4968μs	400.5123 KOps/s	401.2795 KOps/s	$\color{#d91a1a}-0.19\%$
test_creation_empty	38.8420μs	10.1092μs	98.9198 KOps/s	98.9460 KOps/s	$\color{#d91a1a}-0.03\%$
test_creation_nested_1	1.4387ms	13.0956μs	76.3613 KOps/s	77.5916 KOps/s	$\color{#d91a1a}-1.59\%$
test_creation_nested_2	69.0180μs	17.4206μs	57.4033 KOps/s	57.1303 KOps/s	$\color{#35bf28}+0.48\%$
test_clone	0.1769ms	13.5191μs	73.9694 KOps/s	73.0654 KOps/s	$\color{#35bf28}+1.24\%$
test_getitem[int]	1.0041ms	12.9001μs	77.5185 KOps/s	76.5195 KOps/s	$\color{#35bf28}+1.31\%$
test_getitem[slice_int]	0.1725ms	25.0804μs	39.8718 KOps/s	39.9405 KOps/s	$\color{#d91a1a}-0.17\%$
test_getitem[range]	0.3587ms	48.3783μs	20.6704 KOps/s	19.3248 KOps/s	$\textbf{\color{#35bf28}+6.96\%}$
test_getitem[tuple]	0.1426ms	20.2340μs	49.4218 KOps/s	48.3110 KOps/s	$\color{#35bf28}+2.30\%$
test_getitem[list]	0.4922ms	44.1019μs	22.6748 KOps/s	21.9573 KOps/s	$\color{#35bf28}+3.27\%$
test_setitem_dim[int]	52.9190μs	27.3531μs	36.5589 KOps/s	38.6028 KOps/s	$\textbf{\color{#d91a1a}-5.29\%}$
test_setitem_dim[slice_int]	0.1076ms	54.6305μs	18.3048 KOps/s	18.9883 KOps/s	$\color{#d91a1a}-3.60\%$
test_setitem_dim[range]	0.1803ms	75.3264μs	13.2756 KOps/s	13.0540 KOps/s	$\color{#35bf28}+1.70\%$
test_setitem_dim[tuple]	91.5600μs	42.6077μs	23.4699 KOps/s	24.0395 KOps/s	$\color{#d91a1a}-2.37\%$
test_setitem	0.2032ms	20.2481μs	49.3873 KOps/s	49.5501 KOps/s	$\color{#d91a1a}-0.33\%$
test_set	0.1961ms	19.5557μs	51.1361 KOps/s	51.7889 KOps/s	$\color{#d91a1a}-1.26\%$
test_set_shared	4.3035ms	0.1794ms	5.5734 KOps/s	5.7468 KOps/s	$\color{#d91a1a}-3.02\%$
test_update	0.2573ms	21.4982μs	46.5156 KOps/s	46.5660 KOps/s	$\color{#d91a1a}-0.11\%$
test_update_nested	0.2435ms	31.6675μs	31.5781 KOps/s	30.7203 KOps/s	$\color{#35bf28}+2.79\%$
test_update__nested	0.6251ms	33.5973μs	29.7643 KOps/s	28.9561 KOps/s	$\color{#35bf28}+2.79\%$
test_set_nested	0.1806ms	21.9689μs	45.5189 KOps/s	44.8212 KOps/s	$\color{#35bf28}+1.56\%$
test_set_nested_new	0.2489ms	26.3998μs	37.8790 KOps/s	36.9477 KOps/s	$\color{#35bf28}+2.52\%$
test_select	0.2338ms	43.0620μs	23.2223 KOps/s	22.9644 KOps/s	$\color{#35bf28}+1.12\%$
test_select_nested	0.1398ms	63.8804μs	15.6542 KOps/s	15.8080 KOps/s	$\color{#d91a1a}-0.97\%$
test_exclude_nested	0.1704ms	82.9090μs	12.0614 KOps/s	12.1537 KOps/s	$\color{#d91a1a}-0.76\%$
test_empty[True]	0.6254ms	0.4149ms	2.4100 KOps/s	2.3839 KOps/s	$\color{#35bf28}+1.09\%$
test_empty[False]	14.7422μs	1.3797μs	724.7967 KOps/s	732.6409 KOps/s	$\color{#d91a1a}-1.07\%$
test_unbind_speed	0.4038ms	0.2712ms	3.6873 KOps/s	3.7443 KOps/s	$\color{#d91a1a}-1.52\%$
test_unbind_speed_stack0	0.4710ms	0.2687ms	3.7222 KOps/s	3.8476 KOps/s	$\color{#d91a1a}-3.26\%$
test_unbind_speed_stack1	0.1296s	0.8432ms	1.1860 KOps/s	1.3621 KOps/s	$\textbf{\color{#d91a1a}-12.93\%}$
test_split	1.9613ms	1.6319ms	612.7949 Ops/s	550.4267 Ops/s	$\textbf{\color{#35bf28}+11.33\%}$
test_chunk	0.1257s	1.8780ms	532.4877 Ops/s	542.4279 Ops/s	$\color{#d91a1a}-1.83\%$
test_consolidate_njt[False-None]	9.2872ms	8.5270ms	117.2748 Ops/s	118.8159 Ops/s	$\color{#d91a1a}-1.30\%$
test_creation[device0]	4.9837ms	96.0197μs	10.4145 KOps/s	10.4479 KOps/s	$\color{#d91a1a}-0.32\%$
test_creation_from_tensor	0.3604ms	97.9043μs	10.2141 KOps/s	10.4791 KOps/s	$\color{#d91a1a}-2.53\%$
test_add_one[memmap_tensor0]	0.3092ms	5.1009μs	196.0445 KOps/s	205.1919 KOps/s	$\color{#d91a1a}-4.46\%$
test_contiguous[memmap_tensor0]	0.1150ms	0.6189μs	1.6158 MOps/s	1.8426 MOps/s	$\textbf{\color{#d91a1a}-12.31\%}$
test_stack[memmap_tensor0]	0.1032ms	3.4819μs	287.2024 KOps/s	287.9906 KOps/s	$\color{#d91a1a}-0.27\%$
test_memmaptd_index	0.1291s	0.3140ms	3.1844 KOps/s	4.2836 KOps/s	$\textbf{\color{#d91a1a}-25.66\%}$
test_memmaptd_index_astensor	0.6628ms	0.3198ms	3.1270 KOps/s	3.1426 KOps/s	$\color{#d91a1a}-0.49\%$
test_memmaptd_index_op	1.2783ms	0.5691ms	1.7573 KOps/s	1.7707 KOps/s	$\color{#d91a1a}-0.76\%$
test_serialize_model	0.1346s	0.1240s	8.0661 Ops/s	8.4290 Ops/s	$\color{#d91a1a}-4.31\%$
test_serialize_model_pickle	0.4469s	0.3962s	2.5242 Ops/s	2.5385 Ops/s	$\color{#d91a1a}-0.56\%$
test_serialize_weights	0.1382s	0.1193s	8.3801 Ops/s	7.1432 Ops/s	$\textbf{\color{#35bf28}+17.32\%}$
test_serialize_weights_returnearly	0.2868s	0.1832s	5.4588 Ops/s	6.2388 Ops/s	$\textbf{\color{#d91a1a}-12.50\%}$
test_serialize_weights_pickle	1.2928s	0.7486s	1.3358 Ops/s	2.4539 Ops/s	$\textbf{\color{#d91a1a}-45.57\%}$
test_serialize_weights_filesystem	0.1593s	0.1503s	6.6520 Ops/s	6.5927 Ops/s	$\color{#35bf28}+0.90\%$
test_serialize_model_filesystem	0.1770s	0.1543s	6.4828 Ops/s	6.4052 Ops/s	$\color{#35bf28}+1.21\%$
test_reshape_pytree	58.8700μs	26.1471μs	38.2452 KOps/s	37.6735 KOps/s	$\color{#35bf28}+1.52\%$
test_reshape_td	87.2820μs	32.5149μs	30.7551 KOps/s	29.2007 KOps/s	$\textbf{\color{#35bf28}+5.32\%}$
test_view_pytree	75.8510μs	26.0114μs	38.4447 KOps/s	38.1571 KOps/s	$\color{#35bf28}+0.75\%$
test_view_td	0.1462ms	37.0951μs	26.9577 KOps/s	25.9047 KOps/s	$\color{#35bf28}+4.06\%$
test_unbind_pytree	69.7400μs	29.5432μs	33.8488 KOps/s	34.0868 KOps/s	$\color{#d91a1a}-0.70\%$
test_unbind_td	0.3984ms	39.9484μs	25.0323 KOps/s	23.3058 KOps/s	$\textbf{\color{#35bf28}+7.41\%}$
test_split_pytree	95.3670μs	29.5532μs	33.8373 KOps/s	33.5123 KOps/s	$\color{#35bf28}+0.97\%$
test_split_td	0.5750ms	46.1988μs	21.6456 KOps/s	20.8847 KOps/s	$\color{#35bf28}+3.64\%$
test_add_pytree	92.8130μs	35.1319μs	28.4642 KOps/s	28.2182 KOps/s	$\color{#35bf28}+0.87\%$
test_add_td	0.1712ms	57.1227μs	17.5062 KOps/s	18.8279 KOps/s	$\textbf{\color{#d91a1a}-7.02\%}$
test_compile_add_one_nested[tensordict-compile]	0.1465ms	62.2350μs	16.0681 KOps/s	15.1416 KOps/s	$\textbf{\color{#35bf28}+6.12\%}$
test_compile_add_one_nested[tensordict-eager]	0.4417ms	0.1722ms	5.8070 KOps/s	5.6795 KOps/s	$\color{#35bf28}+2.24\%$
test_compile_add_one_nested[pytree-compile]	0.1273ms	46.7733μs	21.3797 KOps/s	21.2581 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_add_one_nested[pytree-eager]	0.4624ms	0.1181ms	8.4684 KOps/s	8.5602 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_copy_nested[tensordict-compile]	71.6230μs	25.9205μs	38.5794 KOps/s	37.2353 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_copy_nested[tensordict-eager]	0.1405ms	59.2637μs	16.8737 KOps/s	16.8215 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_copy_nested[pytree-compile]	0.1532ms	76.1548μs	13.1312 KOps/s	12.9451 KOps/s	$\color{#35bf28}+1.44\%$
test_compile_copy_nested[pytree-eager]	0.1253ms	65.6723μs	15.2271 KOps/s	14.9714 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_add_one_flat[tensordict-compile]	0.1885ms	0.1033ms	9.6804 KOps/s	9.0719 KOps/s	$\textbf{\color{#35bf28}+6.71\%}$
test_compile_add_one_flat[tensordict-eager]	1.5166ms	0.2158ms	4.6347 KOps/s	4.5262 KOps/s	$\color{#35bf28}+2.40\%$
test_compile_add_one_flat[tensorclass-compile]	0.1465ms	45.2231μs	22.1126 KOps/s	21.4195 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_add_one_flat[tensorclass-eager]	0.5627ms	67.9997μs	14.7059 KOps/s	14.5617 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_add_one_flat[pytree-compile]	0.1934ms	0.1030ms	9.7096 KOps/s	9.5241 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_one_flat[pytree-eager]	0.6404ms	0.2047ms	4.8841 KOps/s	5.0169 KOps/s	$\color{#d91a1a}-2.65\%$
test_compile_add_self_flat[tensordict-eager]	0.3533ms	0.2301ms	4.3455 KOps/s	4.2060 KOps/s	$\color{#35bf28}+3.31\%$
test_compile_add_self_flat[tensordict-compile]	0.2345ms	0.1078ms	9.2762 KOps/s	9.0807 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_add_self_flat[tensorclass-eager]	0.1979ms	64.3823μs	15.5322 KOps/s	15.4680 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_add_self_flat[tensorclass-compile]	0.1363ms	47.2324μs	21.1719 KOps/s	20.5152 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_add_self_flat[pytree-eager]	0.2958ms	0.1594ms	6.2721 KOps/s	6.3154 KOps/s	$\color{#d91a1a}-0.69\%$
test_compile_add_self_flat[pytree-compile]	0.2560ms	0.1053ms	9.4966 KOps/s	9.5110 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_copy_flat[tensordict-compile]	72.0640μs	21.6352μs	46.2210 KOps/s	45.5809 KOps/s	$\color{#35bf28}+1.40\%$
test_compile_copy_flat[tensordict-eager]	0.1581ms	68.4247μs	14.6146 KOps/s	14.8756 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_copy_flat[pytree-compile]	0.1526ms	77.9562μs	12.8277 KOps/s	12.7591 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_copy_flat[pytree-eager]	0.1561ms	67.7048μs	14.7700 KOps/s	14.8977 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_assign_and_add[tensordict-compile]	0.3334ms	0.2127ms	4.7004 KOps/s	4.5570 KOps/s	$\color{#35bf28}+3.15\%$
test_compile_assign_and_add[tensordict-eager]	1.7520ms	1.3537ms	738.7425 Ops/s	744.4931 Ops/s	$\color{#d91a1a}-0.77\%$
test_compile_assign_and_add[pytree-compile]	0.4083ms	0.2097ms	4.7691 KOps/s	4.7086 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_assign_and_add[pytree-eager]	2.3328ms	0.8032ms	1.2450 KOps/s	1.2943 KOps/s	$\color{#d91a1a}-3.80\%$
test_compile_assign_and_add_stack[compile]	0.5996ms	0.4608ms	2.1702 KOps/s	2.1537 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_assign_and_add_stack[eager]	3.4484ms	2.5970ms	385.0622 Ops/s	375.1691 Ops/s	$\color{#35bf28}+2.64\%$
test_compile_indexing[tensor-tensordict-compile]	0.1162ms	36.5463μs	27.3626 KOps/s	26.7914 KOps/s	$\color{#35bf28}+2.13\%$
test_compile_indexing[tensor-tensordict-eager]	0.6852ms	32.3442μs	30.9174 KOps/s	28.9693 KOps/s	$\textbf{\color{#35bf28}+6.72\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.2570ms	30.4217μs	32.8712 KOps/s	33.1374 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1102ms	22.7118μs	44.0299 KOps/s	42.2884 KOps/s	$\color{#35bf28}+4.12\%$
test_compile_indexing[tensor-pytree-compile]	84.5170μs	30.4660μs	32.8235 KOps/s	31.9026 KOps/s	$\color{#35bf28}+2.89\%$
test_compile_indexing[tensor-pytree-eager]	66.0430μs	22.5887μs	44.2699 KOps/s	43.3258 KOps/s	$\color{#35bf28}+2.18\%$
test_compile_indexing[slice-tensordict-compile]	0.1430ms	52.4413μs	19.0689 KOps/s	18.7381 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_indexing[slice-tensordict-eager]	0.6374ms	20.5529μs	48.6549 KOps/s	47.2549 KOps/s	$\color{#35bf28}+2.96\%$
test_compile_indexing[slice-tensorclass-compile]	0.1041ms	45.0178μs	22.2135 KOps/s	21.0654 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_compile_indexing[slice-tensorclass-eager]	93.7450μs	18.4732μs	54.1324 KOps/s	54.0884 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[slice-pytree-compile]	0.1381ms	46.4916μs	21.5093 KOps/s	21.1938 KOps/s	$\color{#35bf28}+1.49\%$
test_compile_indexing[slice-pytree-eager]	56.6450μs	18.3881μs	54.3829 KOps/s	54.6263 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_indexing[int-tensordict-compile]	0.1512ms	54.3348μs	18.4044 KOps/s	18.6327 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_indexing[int-tensordict-eager]	1.0410ms	20.6122μs	48.5151 KOps/s	47.1515 KOps/s	$\color{#35bf28}+2.89\%$
test_compile_indexing[int-tensorclass-compile]	0.1023ms	45.7696μs	21.8486 KOps/s	21.7980 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_indexing[int-tensorclass-eager]	89.1060μs	18.2619μs	54.7588 KOps/s	54.2763 KOps/s	$\color{#35bf28}+0.89\%$
test_compile_indexing[int-pytree-compile]	0.1325ms	45.5128μs	21.9718 KOps/s	21.4105 KOps/s	$\color{#35bf28}+2.62\%$
test_compile_indexing[int-pytree-eager]	78.5660μs	18.2810μs	54.7016 KOps/s	54.7194 KOps/s	$\color{#d91a1a}-0.03\%$
test_mod_add[eager]	0.1202ms	34.8663μs	28.6810 KOps/s	28.9083 KOps/s	$\color{#d91a1a}-0.79\%$
test_mod_add[compile]	0.1085ms	49.1038μs	20.3650 KOps/s	20.1813 KOps/s	$\color{#35bf28}+0.91\%$
test_mod_add[compile-overhead]	0.1088ms	48.0203μs	20.8245 KOps/s	19.9876 KOps/s	$\color{#35bf28}+4.19\%$
test_mod_wrap[eager]	0.4448ms	0.2266ms	4.4130 KOps/s	4.3508 KOps/s	$\color{#35bf28}+1.43\%$
test_mod_wrap[compile]	0.3974ms	0.2107ms	4.7465 KOps/s	4.6969 KOps/s	$\color{#35bf28}+1.06\%$
test_mod_wrap[compile-overhead]	0.3657ms	0.2069ms	4.8342 KOps/s	4.7953 KOps/s	$\color{#35bf28}+0.81\%$
test_mod_wrap_and_backward[eager]	15.7376ms	11.8328ms	84.5105 Ops/s	80.1187 Ops/s	$\textbf{\color{#35bf28}+5.48\%}$
test_mod_wrap_and_backward[compile]	16.5967ms	12.3210ms	81.1619 Ops/s	74.9490 Ops/s	$\textbf{\color{#35bf28}+8.29\%}$
test_mod_wrap_and_backward[compile-overhead]	18.1805ms	13.5701ms	73.6914 Ops/s	71.4221 Ops/s	$\color{#35bf28}+3.18\%$
test_seq_add[eager]	0.2226ms	0.1136ms	8.8032 KOps/s	8.1771 KOps/s	$\textbf{\color{#35bf28}+7.66\%}$
test_seq_add[compile]	0.1440ms	64.8232μs	15.4266 KOps/s	15.5829 KOps/s	$\color{#d91a1a}-1.00\%$
test_seq_add[compile-overhead]	0.4492ms	64.2007μs	15.5761 KOps/s	15.8595 KOps/s	$\color{#d91a1a}-1.79\%$
test_seq_wrap[eager]	0.6872ms	0.4469ms	2.2374 KOps/s	2.1792 KOps/s	$\color{#35bf28}+2.67\%$
test_seq_wrap[compile]	0.3346ms	0.2309ms	4.3314 KOps/s	4.2717 KOps/s	$\color{#35bf28}+1.40\%$
test_seq_wrap[compile-overhead]	0.3199ms	0.2266ms	4.4126 KOps/s	4.3157 KOps/s	$\color{#35bf28}+2.24\%$
test_func_call_runtime[False-eager]	1.2083ms	0.5510ms	1.8148 KOps/s	1.7657 KOps/s	$\color{#35bf28}+2.78\%$
test_func_call_runtime[False-compile]	0.5883ms	0.4234ms	2.3618 KOps/s	2.3261 KOps/s	$\color{#35bf28}+1.54\%$
test_func_call_runtime[False-compile-overhead]	1.0161ms	0.4261ms	2.3471 KOps/s	2.3181 KOps/s	$\color{#35bf28}+1.25\%$
test_func_call_runtime[True-eager]	1.2699ms	0.7567ms	1.3216 KOps/s	1.2870 KOps/s	$\color{#35bf28}+2.69\%$
test_func_call_runtime[True-compile]	0.6089ms	0.4634ms	2.1581 KOps/s	2.1275 KOps/s	$\color{#35bf28}+1.44\%$
test_func_call_runtime[True-compile-overhead]	0.6192ms	0.4670ms	2.1412 KOps/s	2.1124 KOps/s	$\color{#35bf28}+1.36\%$
test_func_call_cm_runtime[False-eager]	0.7765ms	0.5395ms	1.8534 KOps/s	1.8098 KOps/s	$\color{#35bf28}+2.41\%$
test_func_call_cm_runtime[False-compile]	0.5760ms	0.4232ms	2.3631 KOps/s	2.2618 KOps/s	$\color{#35bf28}+4.48\%$
test_func_call_cm_runtime[False-compile-overhead]	0.8754ms	0.4318ms	2.3156 KOps/s	2.3105 KOps/s	$\color{#35bf28}+0.22\%$
test_func_call_cm_runtime[True-eager]	1.4646ms	0.9053ms	1.1046 KOps/s	1.0678 KOps/s	$\color{#35bf28}+3.45\%$
test_func_call_cm_runtime[True-compile]	0.8647ms	0.4944ms	2.0225 KOps/s	1.9986 KOps/s	$\color{#35bf28}+1.19\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6765ms	0.4914ms	2.0348 KOps/s	2.0073 KOps/s	$\color{#35bf28}+1.37\%$
test_vmap_func_call_cm_runtime[eager]	2.6199ms	1.9101ms	523.5245 Ops/s	493.2721 Ops/s	$\textbf{\color{#35bf28}+6.13\%}$
test_vmap_func_call_cm_runtime[compile]	0.8988ms	0.5224ms	1.9144 KOps/s	1.9030 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.7922ms	0.5426ms	1.8430 KOps/s	1.8902 KOps/s	$\color{#d91a1a}-2.50\%$
test_distributed	0.2773ms	0.1244ms	8.0357 KOps/s	7.6206 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_tdmodule	59.5210μs	25.3100μs	39.5100 KOps/s	37.6781 KOps/s	$\color{#35bf28}+4.86\%$
test_tdmodule_dispatch	71.8840μs	45.8435μs	21.8134 KOps/s	21.3671 KOps/s	$\color{#35bf28}+2.09\%$
test_tdseq	53.6900μs	27.0292μs	36.9971 KOps/s	34.0959 KOps/s	$\textbf{\color{#35bf28}+8.51\%}$
test_tdseq_dispatch	0.2536ms	56.1094μs	17.8223 KOps/s	18.8315 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_instantiation_functorch	1.6646ms	1.5185ms	658.5397 Ops/s	643.2113 Ops/s	$\color{#35bf28}+2.38\%$
test_exec_functorch	0.2725ms	0.1769ms	5.6516 KOps/s	5.5839 KOps/s	$\color{#35bf28}+1.21\%$
test_exec_functional_call	0.4292ms	0.1737ms	5.7561 KOps/s	5.7074 KOps/s	$\color{#35bf28}+0.85\%$
test_exec_td_decorator	0.5098ms	0.2305ms	4.3381 KOps/s	4.1929 KOps/s	$\color{#35bf28}+3.46\%$
test_vmap_mlp_speed_decorator[True-True]	1.0263ms	0.6430ms	1.5553 KOps/s	1.4897 KOps/s	$\color{#35bf28}+4.41\%$
test_vmap_mlp_speed_decorator[True-False]	1.2823ms	0.6458ms	1.5485 KOps/s	1.5092 KOps/s	$\color{#35bf28}+2.60\%$
test_vmap_mlp_speed_decorator[False-True]	0.7721ms	0.5220ms	1.9157 KOps/s	1.8646 KOps/s	$\color{#35bf28}+2.74\%$
test_vmap_mlp_speed_decorator[False-False]	0.9350ms	0.5258ms	1.9018 KOps/s	1.8705 KOps/s	$\color{#35bf28}+1.67\%$
test_to_module_speed[True]	3.4624ms	1.3789ms	725.2149 Ops/s	739.3728 Ops/s	$\color{#d91a1a}-1.91\%$
test_to_module_speed[False]	2.4883ms	1.3227ms	756.0391 Ops/s	772.1770 Ops/s	$\color{#d91a1a}-2.09\%$
test_tc_init	80.9600μs	44.0899μs	22.6809 KOps/s	22.8421 KOps/s	$\color{#d91a1a}-0.71\%$
test_tc_init_nested	0.1681ms	87.5027μs	11.4282 KOps/s	11.1830 KOps/s	$\color{#35bf28}+2.19\%$
test_tc_first_layer_tensor	26.3290μs	1.5597μs	641.1519 KOps/s	647.3472 KOps/s	$\color{#d91a1a}-0.96\%$
test_tc_first_layer_nontensor	23.9240μs	4.9252μs	203.0394 KOps/s	210.9157 KOps/s	$\color{#d91a1a}-3.73\%$
test_tc_second_layer_tensor	56.7780μs	2.9520μs	338.7567 KOps/s	346.7443 KOps/s	$\color{#d91a1a}-2.30\%$
test_tc_second_layer_nontensor	40.1540μs	6.2022μs	161.2319 KOps/s	164.7776 KOps/s	$\color{#d91a1a}-2.15\%$
test_unbind	0.2596s	14.8540ms	67.3218 Ops/s	76.9575 Ops/s	$\textbf{\color{#d91a1a}-12.52\%}$
test_full_like	22.3542ms	15.5618ms	64.2600 Ops/s	104.2200 Ops/s	$\textbf{\color{#d91a1a}-38.34\%}$
test_zeros_like	13.6061ms	8.7499ms	114.2876 Ops/s	261.5769 Ops/s	$\textbf{\color{#d91a1a}-56.31\%}$
test_ones_like	18.3728ms	8.4506ms	118.3350 Ops/s	237.3582 Ops/s	$\textbf{\color{#d91a1a}-50.14\%}$
test_clone	15.3638ms	11.0814ms	90.2416 Ops/s	154.3623 Ops/s	$\textbf{\color{#d91a1a}-41.54\%}$
test_squeeze	75.5410μs	12.4341μs	80.4243 KOps/s	80.1478 KOps/s	$\color{#35bf28}+0.34\%$
test_unsqueeze	0.3576ms	92.3933μs	10.8233 KOps/s	10.5880 KOps/s	$\color{#35bf28}+2.22\%$
test_split	0.3534ms	0.1972ms	5.0717 KOps/s	5.0724 KOps/s	$\color{#d91a1a}-0.01\%$
test_permute	0.4059ms	0.2028ms	4.9310 KOps/s	4.8591 KOps/s	$\color{#35bf28}+1.48\%$
test_stack	34.7807ms	29.9564ms	33.3819 Ops/s	34.6802 Ops/s	$\color{#d91a1a}-3.74\%$
test_cat	37.0327ms	29.6439ms	33.7338 Ops/s	33.2671 Ops/s	$\color{#35bf28}+1.40\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}57$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	31.8510μs	11.3157μs	88.3725 KOps/s	75.3863 KOps/s	$\textbf{\color{#35bf28}+17.23\%}$
test_plain_set_stack_nested	30.4920μs	11.4672μs	87.2056 KOps/s	74.5022 KOps/s	$\textbf{\color{#35bf28}+17.05\%}$
test_plain_set_nested_inplace	39.2120μs	12.3749μs	80.8085 KOps/s	69.7996 KOps/s	$\textbf{\color{#35bf28}+15.77\%}$
test_plain_set_stack_nested_inplace	41.9220μs	12.4228μs	80.4972 KOps/s	69.8702 KOps/s	$\textbf{\color{#35bf28}+15.21\%}$
test_items	29.6110μs	2.8938μs	345.5625 KOps/s	340.8484 KOps/s	$\color{#35bf28}+1.38\%$
test_items_nested	0.4754ms	0.3698ms	2.7040 KOps/s	2.7919 KOps/s	$\color{#d91a1a}-3.15\%$
test_items_nested_locked	0.5514ms	0.3664ms	2.7295 KOps/s	2.7758 KOps/s	$\color{#d91a1a}-1.67\%$
test_items_nested_leaf	80.6740μs	58.6082μs	17.0625 KOps/s	17.2462 KOps/s	$\color{#d91a1a}-1.07\%$
test_items_stack_nested	0.5432ms	0.3694ms	2.7068 KOps/s	2.7519 KOps/s	$\color{#d91a1a}-1.64\%$
test_items_stack_nested_leaf	0.1421ms	59.5786μs	16.7845 KOps/s	16.6061 KOps/s	$\color{#35bf28}+1.07\%$
test_items_stack_nested_locked	0.4272ms	0.3732ms	2.6798 KOps/s	2.7569 KOps/s	$\color{#d91a1a}-2.80\%$
test_keys	31.7410μs	3.4551μs	289.4290 KOps/s	289.6850 KOps/s	$\color{#d91a1a}-0.09\%$
test_keys_nested	0.1880ms	89.7302μs	11.1445 KOps/s	11.2424 KOps/s	$\color{#d91a1a}-0.87\%$
test_keys_nested_locked	0.7197ms	95.4723μs	10.4742 KOps/s	10.6137 KOps/s	$\color{#d91a1a}-1.31\%$
test_keys_nested_leaf	0.1039ms	80.1503μs	12.4766 KOps/s	12.6067 KOps/s	$\color{#d91a1a}-1.03\%$
test_keys_stack_nested	0.1210ms	89.7763μs	11.1388 KOps/s	11.0615 KOps/s	$\color{#35bf28}+0.70\%$
test_keys_stack_nested_leaf	0.1120ms	81.5276μs	12.2658 KOps/s	12.1811 KOps/s	$\color{#35bf28}+0.70\%$
test_keys_stack_nested_locked	0.1401ms	96.1949μs	10.3956 KOps/s	10.4405 KOps/s	$\color{#d91a1a}-0.43\%$
test_values	4.8068μs	0.8510μs	1.1750 MOps/s	1.1715 MOps/s	$\color{#35bf28}+0.30\%$
test_values_nested	62.0330μs	38.4759μs	25.9903 KOps/s	26.3826 KOps/s	$\color{#d91a1a}-1.49\%$
test_values_nested_locked	68.9630μs	39.7072μs	25.1843 KOps/s	25.3937 KOps/s	$\color{#d91a1a}-0.82\%$
test_values_nested_leaf	65.5930μs	42.7004μs	23.4190 KOps/s	23.4367 KOps/s	$\color{#d91a1a}-0.08\%$
test_values_stack_nested	0.1141ms	38.7264μs	25.8222 KOps/s	25.8768 KOps/s	$\color{#d91a1a}-0.21\%$
test_values_stack_nested_leaf	71.7030μs	43.4220μs	23.0298 KOps/s	23.1424 KOps/s	$\color{#d91a1a}-0.49\%$
test_values_stack_nested_locked	74.1340μs	40.2220μs	24.8620 KOps/s	25.0413 KOps/s	$\color{#d91a1a}-0.72\%$
test_membership	1.8411μs	0.5050μs	1.9802 MOps/s	1.9508 MOps/s	$\color{#35bf28}+1.50\%$
test_membership_nested	82.2335μs	2.0124μs	496.9107 KOps/s	485.7523 KOps/s	$\color{#35bf28}+2.30\%$
test_membership_nested_leaf	91.7040μs	1.9863μs	503.4440 KOps/s	493.6802 KOps/s	$\color{#35bf28}+1.98\%$
test_membership_stacked_nested	0.1246ms	2.0631μs	484.7190 KOps/s	483.9719 KOps/s	$\color{#35bf28}+0.15\%$
test_membership_stacked_nested_leaf	32.2420μs	2.0709μs	482.8926 KOps/s	480.9361 KOps/s	$\color{#35bf28}+0.41\%$
test_membership_nested_last	87.9040μs	3.0943μs	323.1793 KOps/s	320.5690 KOps/s	$\color{#35bf28}+0.81\%$
test_membership_nested_leaf_last	32.9810μs	3.0468μs	328.2120 KOps/s	319.6751 KOps/s	$\color{#35bf28}+2.67\%$
test_membership_stacked_nested_last	32.4220μs	3.0706μs	325.6700 KOps/s	284.3074 KOps/s	$\textbf{\color{#35bf28}+14.55\%}$
test_membership_stacked_nested_leaf_last	27.2310μs	3.0975μs	322.8412 KOps/s	279.7878 KOps/s	$\textbf{\color{#35bf28}+15.39\%}$
test_nested_getleaf	31.7310μs	6.1847μs	161.6900 KOps/s	161.9635 KOps/s	$\color{#d91a1a}-0.17\%$
test_nested_get	34.5320μs	5.8895μs	169.7933 KOps/s	167.6757 KOps/s	$\color{#35bf28}+1.26\%$
test_stacked_getleaf	35.9620μs	6.2019μs	161.2399 KOps/s	161.9406 KOps/s	$\color{#d91a1a}-0.43\%$
test_stacked_get	25.7810μs	5.8477μs	171.0087 KOps/s	169.8844 KOps/s	$\color{#35bf28}+0.66\%$
test_nested_getitemleaf	30.3810μs	6.5107μs	153.5935 KOps/s	154.7012 KOps/s	$\color{#d91a1a}-0.72\%$
test_nested_getitem	38.2110μs	6.1740μs	161.9704 KOps/s	159.9489 KOps/s	$\color{#35bf28}+1.26\%$
test_stacked_getitemleaf	35.9310μs	6.4927μs	154.0197 KOps/s	154.4529 KOps/s	$\color{#d91a1a}-0.28\%$
test_stacked_getitem	31.7910μs	6.2033μs	161.2033 KOps/s	161.7786 KOps/s	$\color{#d91a1a}-0.36\%$
test_lock_nested	9.7246ms	0.3812ms	2.6232 KOps/s	2.5678 KOps/s	$\color{#35bf28}+2.16\%$
test_lock_stack_nested	0.4272ms	0.3443ms	2.9045 KOps/s	2.8380 KOps/s	$\color{#35bf28}+2.34\%$
test_unlock_nested	0.6141ms	0.3139ms	3.1856 KOps/s	3.0811 KOps/s	$\color{#35bf28}+3.39\%$
test_unlock_stack_nested	0.3669ms	0.2821ms	3.5451 KOps/s	3.4292 KOps/s	$\color{#35bf28}+3.38\%$
test_flatten_speed	0.1096ms	74.5580μs	13.4124 KOps/s	13.2903 KOps/s	$\color{#35bf28}+0.92\%$
test_unflatten_speed	0.5019ms	0.3182ms	3.1428 KOps/s	3.1086 KOps/s	$\color{#35bf28}+1.10\%$
test_common_ops	1.6677ms	0.5759ms	1.7365 KOps/s	1.5031 KOps/s	$\textbf{\color{#35bf28}+15.52\%}$
test_creation	0.1031ms	1.7934μs	557.6025 KOps/s	563.5323 KOps/s	$\color{#d91a1a}-1.05\%$
test_creation_empty	42.5220μs	6.4736μs	154.4737 KOps/s	98.8154 KOps/s	$\textbf{\color{#35bf28}+56.33\%}$
test_creation_nested_1	33.9220μs	8.2309μs	121.4927 KOps/s	85.1542 KOps/s	$\textbf{\color{#35bf28}+42.67\%}$
test_creation_nested_2	37.8520μs	10.9119μs	91.6429 KOps/s	69.2482 KOps/s	$\textbf{\color{#35bf28}+32.34\%}$
test_clone	0.1307ms	10.5735μs	94.5761 KOps/s	84.9186 KOps/s	$\textbf{\color{#35bf28}+11.37\%}$
test_getitem[int]	1.6507ms	10.7455μs	93.0624 KOps/s	88.5935 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_getitem[slice_int]	0.2272ms	20.7417μs	48.2120 KOps/s	45.3889 KOps/s	$\textbf{\color{#35bf28}+6.22\%}$
test_getitem[range]	0.1636ms	38.0423μs	26.2865 KOps/s	25.1052 KOps/s	$\color{#35bf28}+4.71\%$
test_getitem[tuple]	0.1145ms	18.1341μs	55.1449 KOps/s	52.5374 KOps/s	$\color{#35bf28}+4.96\%$
test_getitem[list]	0.2434ms	33.0416μs	30.2649 KOps/s	28.7907 KOps/s	$\textbf{\color{#35bf28}+5.12\%}$
test_setitem_dim[int]	40.1420μs	19.2126μs	52.0493 KOps/s	48.6195 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$
test_setitem_dim[slice_int]	0.1688ms	38.4552μs	26.0043 KOps/s	25.2913 KOps/s	$\color{#35bf28}+2.82\%$
test_setitem_dim[range]	89.9140μs	53.0266μs	18.8585 KOps/s	18.2555 KOps/s	$\color{#35bf28}+3.30\%$
test_setitem_dim[tuple]	60.0120μs	32.4620μs	30.8053 KOps/s	28.8669 KOps/s	$\textbf{\color{#35bf28}+6.71\%}$
test_setitem	86.6740μs	14.0773μs	71.0365 KOps/s	57.3608 KOps/s	$\textbf{\color{#35bf28}+23.84\%}$
test_set	88.2340μs	13.5182μs	73.9744 KOps/s	59.4278 KOps/s	$\textbf{\color{#35bf28}+24.48\%}$
test_set_shared	1.6510ms	0.1531ms	6.5323 KOps/s	6.4955 KOps/s	$\color{#35bf28}+0.57\%$
test_update	0.9476ms	15.5148μs	64.4546 KOps/s	48.3077 KOps/s	$\textbf{\color{#35bf28}+33.43\%}$
test_update_nested	97.8340μs	20.6720μs	48.3747 KOps/s	38.3572 KOps/s	$\textbf{\color{#35bf28}+26.12\%}$
test_update__nested	1.5303ms	26.3033μs	38.0180 KOps/s	36.7945 KOps/s	$\color{#35bf28}+3.33\%$
test_set_nested	0.1252ms	15.0658μs	66.3753 KOps/s	55.7674 KOps/s	$\textbf{\color{#35bf28}+19.02\%}$
test_set_nested_new	0.1012ms	17.2572μs	57.9467 KOps/s	49.0732 KOps/s	$\textbf{\color{#35bf28}+18.08\%}$
test_select	0.1050ms	29.0643μs	34.4065 KOps/s	30.4239 KOps/s	$\textbf{\color{#35bf28}+13.09\%}$
test_select_nested	73.4530μs	43.8608μs	22.7994 KOps/s	22.6793 KOps/s	$\color{#35bf28}+0.53\%$
test_exclude_nested	0.1020ms	63.5859μs	15.7267 KOps/s	15.6587 KOps/s	$\color{#35bf28}+0.43\%$
test_empty[True]	0.4155ms	0.2988ms	3.3468 KOps/s	3.3449 KOps/s	$\color{#35bf28}+0.06\%$
test_empty[False]	3.3881μs	0.8384μs	1.1928 MOps/s	1.2207 MOps/s	$\color{#d91a1a}-2.29\%$
test_to	87.5840μs	56.6539μs	17.6510 KOps/s	18.1212 KOps/s	$\color{#d91a1a}-2.59\%$
test_to_nonblocking	0.1946ms	48.0572μs	20.8086 KOps/s	20.9410 KOps/s	$\color{#d91a1a}-0.63\%$
test_unbind_speed	0.8098ms	0.2306ms	4.3357 KOps/s	4.0919 KOps/s	$\textbf{\color{#35bf28}+5.96\%}$
test_unbind_speed_stack0	0.3571ms	0.2365ms	4.2285 KOps/s	4.0284 KOps/s	$\color{#35bf28}+4.97\%$
test_unbind_speed_stack1	94.8264ms	0.6691ms	1.4945 KOps/s	1.4628 KOps/s	$\color{#35bf28}+2.16\%$
test_split	96.4935ms	1.5858ms	630.6122 Ops/s	607.3719 Ops/s	$\color{#35bf28}+3.83\%$
test_chunk	97.6251ms	1.5901ms	628.8998 Ops/s	607.6914 Ops/s	$\color{#35bf28}+3.49\%$
test_consolidate[False-None]	0.1003s	2.9628ms	337.5179 Ops/s	334.4361 Ops/s	$\color{#35bf28}+0.92\%$
test_consolidate[default-None]	1.8300ms	1.6968ms	589.3493 Ops/s	563.9114 Ops/s	$\color{#35bf28}+4.51\%$
test_consolidate[reduce-overhead-None]	1.9089ms	1.7383ms	575.2599 Ops/s	554.6691 Ops/s	$\color{#35bf28}+3.71\%$
test_consolidate_njt[False-None]	6.9571ms	6.5694ms	152.2217 Ops/s	151.1552 Ops/s	$\color{#35bf28}+0.71\%$
test_to[False-False-None]	1.8944ms	1.7374ms	575.5773 Ops/s	581.6273 Ops/s	$\color{#d91a1a}-1.04\%$
test_to[True-False-None]	1.5306ms	1.3167ms	759.4900 Ops/s	728.3904 Ops/s	$\color{#35bf28}+4.27\%$
test_to[within-False-None]	4.3139ms	4.1445ms	241.2854 Ops/s	172.8556 Ops/s	$\textbf{\color{#35bf28}+39.59\%}$
test_to[True-default-None]	5.6165ms	5.3311ms	187.5773 Ops/s	180.4422 Ops/s	$\color{#35bf28}+3.95\%$
test_to_njt[False-False-None]	7.1930ms	6.9669ms	143.5365 Ops/s	135.3720 Ops/s	$\textbf{\color{#35bf28}+6.03\%}$
test_to_njt[True-False-None]	5.7499ms	5.5006ms	181.7982 Ops/s	175.6928 Ops/s	$\color{#35bf28}+3.48\%$
test_to_njt[within-False-None]	12.5031ms	12.2968ms	81.3220 Ops/s	79.2058 Ops/s	$\color{#35bf28}+2.67\%$
test_creation[device0]	0.4565ms	81.0088μs	12.3443 KOps/s	12.1070 KOps/s	$\color{#35bf28}+1.96\%$
test_creation_from_tensor	0.4543ms	84.4569μs	11.8404 KOps/s	11.6555 KOps/s	$\color{#35bf28}+1.59\%$
test_add_one[memmap_tensor0]	0.2449ms	6.9343μs	144.2113 KOps/s	135.1120 KOps/s	$\textbf{\color{#35bf28}+6.73\%}$
test_contiguous[memmap_tensor0]	1.9611μs	0.4283μs	2.3348 MOps/s	2.3861 MOps/s	$\color{#d91a1a}-2.15\%$
test_stack[memmap_tensor0]	0.1366ms	4.4544μs	224.4991 KOps/s	214.0227 KOps/s	$\color{#35bf28}+4.90\%$
test_memmaptd_index	1.6213ms	0.2481ms	4.0311 KOps/s	3.7373 KOps/s	$\textbf{\color{#35bf28}+7.86\%}$
test_memmaptd_index_astensor	0.5719ms	0.3086ms	3.2402 KOps/s	3.0030 KOps/s	$\textbf{\color{#35bf28}+7.90\%}$
test_memmaptd_index_op	0.9681ms	0.5582ms	1.7915 KOps/s	1.5202 KOps/s	$\textbf{\color{#35bf28}+17.85\%}$
test_serialize_model	0.4228s	0.1731s	5.7766 Ops/s	7.6183 Ops/s	$\textbf{\color{#d91a1a}-24.17\%}$
test_serialize_model_pickle	1.3495s	1.2113s	0.8255 Ops/s	0.8222 Ops/s	$\color{#35bf28}+0.41\%$
test_serialize_weights	0.1313s	0.1300s	7.6904 Ops/s	7.6259 Ops/s	$\color{#35bf28}+0.85\%$
test_serialize_weights_returnearly	0.3256s	54.4506ms	18.3653 Ops/s	14.6641 Ops/s	$\textbf{\color{#35bf28}+25.24\%}$
test_serialize_weights_pickle	1.3759s	1.2202s	0.8196 Ops/s	0.8239 Ops/s	$\color{#d91a1a}-0.53\%$
test_reshape_pytree	0.1964ms	22.1436μs	45.1597 KOps/s	43.6208 KOps/s	$\color{#35bf28}+3.53\%$
test_reshape_td	0.1542ms	26.4020μs	37.8760 KOps/s	35.2343 KOps/s	$\textbf{\color{#35bf28}+7.50\%}$
test_view_pytree	0.1650ms	21.9874μs	45.4805 KOps/s	44.2272 KOps/s	$\color{#35bf28}+2.83\%$
test_view_td	83.7140μs	29.2082μs	34.2370 KOps/s	32.7808 KOps/s	$\color{#35bf28}+4.44\%$
test_unbind_pytree	0.1493ms	27.7435μs	36.0445 KOps/s	34.9062 KOps/s	$\color{#35bf28}+3.26\%$
test_unbind_td	0.7781ms	36.0558μs	27.7348 KOps/s	26.3362 KOps/s	$\textbf{\color{#35bf28}+5.31\%}$
test_split_pytree	0.1427ms	29.2243μs	34.2181 KOps/s	32.6320 KOps/s	$\color{#35bf28}+4.86\%$
test_split_td	0.9746ms	37.6252μs	26.5779 KOps/s	25.1527 KOps/s	$\textbf{\color{#35bf28}+5.67\%}$
test_add_pytree	0.1712ms	34.6304μs	28.8764 KOps/s	26.6952 KOps/s	$\textbf{\color{#35bf28}+8.17\%}$
test_add_td	0.2278ms	46.1149μs	21.6850 KOps/s	18.8283 KOps/s	$\textbf{\color{#35bf28}+15.17\%}$
test_compile_add_one_nested[tensordict-compile]	0.2712ms	0.1214ms	8.2367 KOps/s	7.8274 KOps/s	$\textbf{\color{#35bf28}+5.23\%}$
test_compile_add_one_nested[tensordict-eager]	0.3320ms	0.1327ms	7.5384 KOps/s	7.3935 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_add_one_nested[pytree-compile]	0.2512ms	97.4635μs	10.2602 KOps/s	10.0309 KOps/s	$\color{#35bf28}+2.29\%$
test_compile_add_one_nested[pytree-eager]	0.3188ms	0.1497ms	6.6816 KOps/s	6.5412 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_copy_nested[tensordict-compile]	0.2041ms	23.7634μs	42.0815 KOps/s	44.7345 KOps/s	$\textbf{\color{#d91a1a}-5.93\%}$
test_compile_copy_nested[tensordict-eager]	0.1506ms	29.3760μs	34.0414 KOps/s	33.1417 KOps/s	$\color{#35bf28}+2.71\%$
test_compile_copy_nested[pytree-compile]	0.4750ms	64.1107μs	15.5980 KOps/s	15.1087 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_copy_nested[pytree-eager]	0.1789ms	49.4303μs	20.2305 KOps/s	20.0907 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_add_one_flat[tensordict-compile]	0.2956ms	0.1432ms	6.9809 KOps/s	7.0552 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_add_one_flat[tensordict-eager]	0.3515ms	0.2171ms	4.6056 KOps/s	4.6263 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_add_one_flat[tensorclass-compile]	0.2528ms	98.7116μs	10.1305 KOps/s	10.0313 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_add_one_flat[tensorclass-eager]	0.2362ms	56.0620μs	17.8374 KOps/s	17.4935 KOps/s	$\color{#35bf28}+1.97\%$
test_compile_add_one_flat[pytree-compile]	0.2982ms	0.1364ms	7.3335 KOps/s	7.3983 KOps/s	$\color{#d91a1a}-0.88\%$
test_compile_add_one_flat[pytree-eager]	0.6615ms	0.4793ms	2.0865 KOps/s	2.0351 KOps/s	$\color{#35bf28}+2.53\%$
test_compile_add_self_flat[tensordict-eager]	0.4176ms	0.2615ms	3.8239 KOps/s	3.8674 KOps/s	$\color{#d91a1a}-1.13\%$
test_compile_add_self_flat[tensordict-compile]	0.2964ms	0.1447ms	6.9104 KOps/s	7.0046 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_add_self_flat[tensorclass-eager]	0.2323ms	68.0001μs	14.7059 KOps/s	14.7352 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_add_self_flat[tensorclass-compile]	0.2828ms	0.1005ms	9.9461 KOps/s	10.0343 KOps/s	$\color{#d91a1a}-0.88\%$
test_compile_add_self_flat[pytree-eager]	0.5886ms	0.4011ms	2.4932 KOps/s	2.4705 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_add_self_flat[pytree-compile]	0.2750ms	0.1368ms	7.3085 KOps/s	7.4292 KOps/s	$\color{#d91a1a}-1.63\%$
test_compile_copy_flat[tensordict-compile]	0.1685ms	18.9437μs	52.7880 KOps/s	55.0293 KOps/s	$\color{#d91a1a}-4.07\%$
test_compile_copy_flat[tensordict-eager]	57.7320μs	31.3901μs	31.8571 KOps/s	31.9784 KOps/s	$\color{#d91a1a}-0.38\%$
test_compile_copy_flat[pytree-compile]	0.2129ms	70.9265μs	14.0991 KOps/s	13.9613 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_copy_flat[pytree-eager]	74.8830μs	51.7770μs	19.3136 KOps/s	19.1485 KOps/s	$\color{#35bf28}+0.86\%$
test_compile_assign_and_add[tensordict-compile]	1.6230ms	0.3911ms	2.5568 KOps/s	2.1738 KOps/s	$\textbf{\color{#35bf28}+17.62\%}$
test_compile_assign_and_add[tensordict-eager]	2.9571ms	2.6681ms	374.8019 Ops/s	377.3664 Ops/s	$\color{#d91a1a}-0.68\%$
test_compile_assign_and_add[pytree-compile]	1.5864ms	0.4372ms	2.2873 KOps/s	2.2324 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_assign_and_add[pytree-eager]	2.9609ms	2.6811ms	372.9765 Ops/s	370.7875 Ops/s	$\color{#35bf28}+0.59\%$
test_compile_indexing[tensor-tensordict-compile]	0.3174ms	0.1222ms	8.1845 KOps/s	8.8013 KOps/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5966ms	85.1283μs	11.7470 KOps/s	11.9556 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2996ms	0.1156ms	8.6529 KOps/s	9.3179 KOps/s	$\textbf{\color{#d91a1a}-7.14\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.2742ms	73.0332μs	13.6924 KOps/s	14.6684 KOps/s	$\textbf{\color{#d91a1a}-6.65\%}$
test_compile_indexing[tensor-pytree-compile]	0.2789ms	0.1155ms	8.6544 KOps/s	9.3835 KOps/s	$\textbf{\color{#d91a1a}-7.77\%}$
test_compile_indexing[tensor-pytree-eager]	0.2333ms	73.0661μs	13.6862 KOps/s	14.6786 KOps/s	$\textbf{\color{#d91a1a}-6.76\%}$
test_compile_indexing[slice-tensordict-compile]	0.2571ms	0.1075ms	9.3037 KOps/s	9.8047 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_compile_indexing[slice-tensordict-eager]	0.1383ms	17.0527μs	58.6419 KOps/s	55.3203 KOps/s	$\textbf{\color{#35bf28}+6.00\%}$
test_compile_indexing[slice-tensorclass-compile]	0.2897ms	98.5185μs	10.1504 KOps/s	10.1550 KOps/s	$\color{#d91a1a}-0.05\%$
test_compile_indexing[slice-tensorclass-eager]	0.1264ms	15.9922μs	62.5306 KOps/s	60.4110 KOps/s	$\color{#35bf28}+3.51\%$
test_compile_indexing[slice-pytree-compile]	0.2734ms	0.1012ms	9.8783 KOps/s	10.1307 KOps/s	$\color{#d91a1a}-2.49\%$
test_compile_indexing[slice-pytree-eager]	0.1763ms	15.8867μs	62.9459 KOps/s	60.4709 KOps/s	$\color{#35bf28}+4.09\%$
test_compile_indexing[int-tensordict-compile]	0.2586ms	0.1045ms	9.5683 KOps/s	9.7016 KOps/s	$\color{#d91a1a}-1.37\%$
test_compile_indexing[int-tensordict-eager]	0.5747ms	17.2699μs	57.9043 KOps/s	56.2081 KOps/s	$\color{#35bf28}+3.02\%$
test_compile_indexing[int-tensorclass-compile]	0.2949ms	0.1025ms	9.7565 KOps/s	10.1175 KOps/s	$\color{#d91a1a}-3.57\%$
test_compile_indexing[int-tensorclass-eager]	0.1754ms	15.7938μs	63.3159 KOps/s	61.1667 KOps/s	$\color{#35bf28}+3.51\%$
test_compile_indexing[int-pytree-compile]	0.2814ms	0.1029ms	9.7175 KOps/s	9.8721 KOps/s	$\color{#d91a1a}-1.57\%$
test_compile_indexing[int-pytree-eager]	0.2192ms	16.5157μs	60.5484 KOps/s	60.8818 KOps/s	$\color{#d91a1a}-0.55\%$
test_mod_add[eager]	0.2261ms	36.3526μs	27.5084 KOps/s	25.0423 KOps/s	$\textbf{\color{#35bf28}+9.85\%}$
test_mod_add[compile]	0.2876ms	80.7007μs	12.3915 KOps/s	12.2190 KOps/s	$\color{#35bf28}+1.41\%$
test_mod_add[compile-overhead]	0.3389ms	0.1692ms	5.9087 KOps/s	5.5688 KOps/s	$\textbf{\color{#35bf28}+6.10\%}$
test_mod_wrap[eager]	0.4030ms	0.2504ms	3.9941 KOps/s	3.9122 KOps/s	$\color{#35bf28}+2.09\%$
test_mod_wrap[compile]	0.4194ms	0.2808ms	3.5614 KOps/s	3.4270 KOps/s	$\color{#35bf28}+3.92\%$
test_mod_wrap[compile-overhead]	7.0319ms	3.7238ms	268.5403 Ops/s	274.3183 Ops/s	$\color{#d91a1a}-2.11\%$
test_mod_wrap_and_backward[eager]	1.6870ms	1.3809ms	724.1893 Ops/s	677.7905 Ops/s	$\textbf{\color{#35bf28}+6.85\%}$
test_mod_wrap_and_backward[compile]	1.6017ms	1.3882ms	720.3731 Ops/s	719.5755 Ops/s	$\color{#35bf28}+0.11\%$
test_mod_wrap_and_backward[compile-overhead]	1.4774ms	0.9993ms	1.0007 KOps/s	945.3912 Ops/s	$\textbf{\color{#35bf28}+5.85\%}$
test_seq_add[eager]	0.3133ms	0.1122ms	8.9157 KOps/s	8.2473 KOps/s	$\textbf{\color{#35bf28}+8.10\%}$
test_seq_add[compile]	0.2864ms	88.9942μs	11.2367 KOps/s	10.7240 KOps/s	$\color{#35bf28}+4.78\%$
test_seq_add[compile-overhead]	0.2815ms	0.1304ms	7.6664 KOps/s	7.3741 KOps/s	$\color{#35bf28}+3.96\%$
test_seq_wrap[eager]	0.5946ms	0.4149ms	2.4104 KOps/s	2.2412 KOps/s	$\textbf{\color{#35bf28}+7.55\%}$
test_seq_wrap[compile]	0.4483ms	0.2979ms	3.3567 KOps/s	3.1987 KOps/s	$\color{#35bf28}+4.94\%$
test_seq_wrap[compile-overhead]	0.3992ms	0.2270ms	4.4062 KOps/s	4.2933 KOps/s	$\color{#35bf28}+2.63\%$
test_func_call_runtime[False-eager]	0.9568ms	0.7546ms	1.3251 KOps/s	1.3200 KOps/s	$\color{#35bf28}+0.39\%$
test_func_call_runtime[False-compile]	0.9312ms	0.7618ms	1.3127 KOps/s	1.3315 KOps/s	$\color{#d91a1a}-1.41\%$
test_func_call_runtime[False-compile-overhead]	0.5302ms	0.3663ms	2.7301 KOps/s	2.6936 KOps/s	$\color{#35bf28}+1.35\%$
test_func_call_runtime[True-eager]	1.0976ms	0.9089ms	1.1002 KOps/s	1.0909 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_runtime[True-compile]	0.9402ms	0.7571ms	1.3208 KOps/s	1.2981 KOps/s	$\color{#35bf28}+1.75\%$
test_func_call_runtime[True-compile-overhead]	0.5303ms	0.3881ms	2.5763 KOps/s	2.5547 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_cm_runtime[False-eager]	0.9030ms	0.7398ms	1.3517 KOps/s	1.2797 KOps/s	$\textbf{\color{#35bf28}+5.62\%}$
test_func_call_cm_runtime[False-compile]	1.0472ms	0.7798ms	1.2824 KOps/s	1.3228 KOps/s	$\color{#d91a1a}-3.05\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5050ms	0.3685ms	2.7135 KOps/s	2.6805 KOps/s	$\color{#35bf28}+1.23\%$
test_func_call_cm_runtime[True-eager]	1.2492ms	1.0310ms	969.9701 Ops/s	979.6320 Ops/s	$\color{#d91a1a}-0.99\%$
test_func_call_cm_runtime[True-compile]	0.9528ms	0.7856ms	1.2729 KOps/s	1.2408 KOps/s	$\color{#35bf28}+2.59\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6030ms	0.4166ms	2.4001 KOps/s	2.3945 KOps/s	$\color{#35bf28}+0.23\%$
test_vmap_func_call_cm_runtime[eager]	2.5862ms	2.1329ms	468.8490 Ops/s	468.7779 Ops/s	$\color{#35bf28}+0.02\%$
test_vmap_func_call_cm_runtime[compile]	0.9720ms	0.8032ms	1.2450 KOps/s	1.2099 KOps/s	$\color{#35bf28}+2.90\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5661ms	0.4179ms	2.3928 KOps/s	2.3739 KOps/s	$\color{#35bf28}+0.79\%$
test_distributed	3.0002ms	0.1828ms	5.4709 KOps/s	8.4904 KOps/s	$\textbf{\color{#d91a1a}-35.56\%}$
test_tdmodule	32.4010μs	18.5749μs	53.8362 KOps/s	46.4361 KOps/s	$\textbf{\color{#35bf28}+15.94\%}$
test_tdmodule_dispatch	55.9530μs	33.5259μs	29.8277 KOps/s	26.1586 KOps/s	$\textbf{\color{#35bf28}+14.03\%}$
test_tdseq	39.0010μs	19.3437μs	51.6965 KOps/s	44.8389 KOps/s	$\textbf{\color{#35bf28}+15.29\%}$
test_tdseq_dispatch	0.1165ms	36.3313μs	27.5245 KOps/s	23.9436 KOps/s	$\textbf{\color{#35bf28}+14.96\%}$
test_instantiation_functorch	1.9757ms	1.5627ms	639.9212 Ops/s	629.4835 Ops/s	$\color{#35bf28}+1.66\%$
test_exec_functorch	0.2602ms	0.1454ms	6.8770 KOps/s	6.5825 KOps/s	$\color{#35bf28}+4.47\%$
test_exec_functional_call	0.5508ms	0.1389ms	7.1998 KOps/s	6.9847 KOps/s	$\color{#35bf28}+3.08\%$
test_exec_td_decorator	0.6014ms	0.1896ms	5.2733 KOps/s	5.2206 KOps/s	$\color{#35bf28}+1.01\%$
test_vmap_mlp_speed_decorator[True-True]	0.8753ms	0.6884ms	1.4527 KOps/s	1.3789 KOps/s	$\textbf{\color{#35bf28}+5.35\%}$
test_vmap_mlp_speed_decorator[True-False]	1.1057ms	0.6864ms	1.4568 KOps/s	1.4273 KOps/s	$\color{#35bf28}+2.07\%$
test_vmap_mlp_speed_decorator[False-True]	1.0279ms	0.6026ms	1.6594 KOps/s	1.6113 KOps/s	$\color{#35bf28}+2.99\%$
test_vmap_mlp_speed_decorator[False-False]	1.0165ms	0.6045ms	1.6543 KOps/s	1.5833 KOps/s	$\color{#35bf28}+4.48\%$
test_vmap_transformer_speed_decorator[True-True]	19.9332ms	19.4430ms	51.4323 Ops/s	51.4767 Ops/s	$\color{#d91a1a}-0.09\%$
test_vmap_transformer_speed_decorator[True-False]	20.4896ms	19.5419ms	51.1722 Ops/s	51.6756 Ops/s	$\color{#d91a1a}-0.97\%$
test_vmap_transformer_speed_decorator[False-True]	19.6752ms	19.4321ms	51.4613 Ops/s	51.8705 Ops/s	$\color{#d91a1a}-0.79\%$
test_vmap_transformer_speed_decorator[False-False]	19.7756ms	19.4081ms	51.5248 Ops/s	51.8045 Ops/s	$\color{#d91a1a}-0.54\%$
test_to_module_speed[True]	1.3817ms	0.9607ms	1.0409 KOps/s	1.0348 KOps/s	$\color{#35bf28}+0.58\%$
test_to_module_speed[False]	1.3769ms	0.9538ms	1.0484 KOps/s	1.0485 KOps/s	$\color{#d91a1a}-0.01\%$
test_tc_init	0.1566ms	33.7907μs	29.5940 KOps/s	26.0457 KOps/s	$\textbf{\color{#35bf28}+13.62\%}$
test_tc_init_nested	0.1134ms	68.0980μs	14.6847 KOps/s	13.2756 KOps/s	$\textbf{\color{#35bf28}+10.61\%}$
test_tc_first_layer_tensor	0.4049ms	0.8156μs	1.2261 MOps/s	1.4614 MOps/s	$\textbf{\color{#d91a1a}-16.10\%}$
test_tc_first_layer_nontensor	0.1017ms	2.2412μs	446.1892 KOps/s	432.8250 KOps/s	$\color{#35bf28}+3.09\%$
test_tc_second_layer_tensor	23.3510μs	1.4921μs	670.2109 KOps/s	712.8577 KOps/s	$\textbf{\color{#d91a1a}-5.98\%}$
test_tc_second_layer_nontensor	0.4156ms	3.0091μs	332.3216 KOps/s	329.1568 KOps/s	$\color{#35bf28}+0.96\%$
test_unbind	0.2240s	11.2624ms	88.7907 Ops/s	141.1256 Ops/s	$\textbf{\color{#d91a1a}-37.08\%}$
test_full_like	11.4232ms	10.0634ms	99.3698 Ops/s	98.9624 Ops/s	$\color{#35bf28}+0.41\%$
test_zeros_like	5.8725ms	4.3690ms	228.8854 Ops/s	229.0482 Ops/s	$\color{#d91a1a}-0.07\%$
test_ones_like	5.8041ms	4.4558ms	224.4244 Ops/s	223.4283 Ops/s	$\color{#35bf28}+0.45\%$
test_clone	8.1707ms	6.9824ms	143.2171 Ops/s	103.0967 Ops/s	$\textbf{\color{#35bf28}+38.92\%}$
test_squeeze	0.1303ms	9.6911μs	103.1871 KOps/s	105.6238 KOps/s	$\color{#d91a1a}-2.31\%$
test_unsqueeze	0.2329ms	71.6196μs	13.9627 KOps/s	13.2709 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_split	0.3069ms	0.1570ms	6.3704 KOps/s	5.9938 KOps/s	$\textbf{\color{#35bf28}+6.28\%}$
test_permute	0.3262ms	0.1777ms	5.6264 KOps/s	5.2931 KOps/s	$\textbf{\color{#35bf28}+6.30\%}$
test_stack	53.2448ms	51.5248ms	19.4081 Ops/s	19.1990 Ops/s	$\color{#35bf28}+1.09\%$
test_cat	52.4337ms	51.5639ms	19.3934 Ops/s	19.2758 Ops/s	$\color{#35bf28}+0.61\%$