[BugFix] Deserializing a consolidated TD reproduces a consolidated TD

[vmoens]

Stack from ghstack (oldest at bottom):

• [Refactor] Put values, lengths and offsets of NJTs together in storage #1023
• [Feature] TD+NJT to(device) support #1022
• [Feature] NJT with lengths #1021
• [Feature] Non-blocking for consolidated TD #1020
• -> [BugFix] Deserializing a consolidated TD reproduces a consolidated TD #1019

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}18$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	59.0500μs	25.2437μs	39.6139 KOps/s	40.6596 KOps/s	$\color{#d91a1a}-2.57\%$
test_plain_set_stack_nested	84.5680μs	25.6750μs	38.9484 KOps/s	39.8994 KOps/s	$\color{#d91a1a}-2.38\%$
test_plain_set_nested_inplace	60.5030μs	27.7147μs	36.0819 KOps/s	37.1952 KOps/s	$\color{#d91a1a}-2.99\%$
test_plain_set_stack_nested_inplace	67.3350μs	27.8459μs	35.9119 KOps/s	37.0768 KOps/s	$\color{#d91a1a}-3.14\%$
test_items	24.8860μs	4.2378μs	235.9736 KOps/s	239.1530 KOps/s	$\color{#d91a1a}-1.33\%$
test_items_nested	0.4829ms	0.3810ms	2.6248 KOps/s	2.4793 KOps/s	$\textbf{\color{#35bf28}+5.87\%}$
test_items_nested_locked	0.7188ms	0.3848ms	2.5989 KOps/s	2.6149 KOps/s	$\color{#d91a1a}-0.61\%$
test_items_nested_leaf	0.1707ms	80.8330μs	12.3712 KOps/s	12.4399 KOps/s	$\color{#d91a1a}-0.55\%$
test_items_stack_nested	0.7212ms	0.3871ms	2.5835 KOps/s	2.4960 KOps/s	$\color{#35bf28}+3.50\%$
test_items_stack_nested_leaf	0.1806ms	84.6478μs	11.8137 KOps/s	11.7669 KOps/s	$\color{#35bf28}+0.40\%$
test_items_stack_nested_locked	0.8796ms	0.3947ms	2.5337 KOps/s	2.5621 KOps/s	$\color{#d91a1a}-1.11\%$
test_keys	22.6530μs	3.4920μs	286.3668 KOps/s	288.3002 KOps/s	$\color{#d91a1a}-0.67\%$
test_keys_nested	0.2544ms	0.1345ms	7.4348 KOps/s	7.3307 KOps/s	$\color{#35bf28}+1.42\%$
test_keys_nested_locked	1.7004ms	0.1390ms	7.1961 KOps/s	7.0542 KOps/s	$\color{#35bf28}+2.01\%$
test_keys_nested_leaf	0.2114ms	0.1176ms	8.5013 KOps/s	8.3779 KOps/s	$\color{#35bf28}+1.47\%$
test_keys_stack_nested	0.2621ms	0.1349ms	7.4143 KOps/s	7.4429 KOps/s	$\color{#d91a1a}-0.38\%$
test_keys_stack_nested_leaf	0.2548ms	0.1183ms	8.4534 KOps/s	8.5482 KOps/s	$\color{#d91a1a}-1.11\%$
test_keys_stack_nested_locked	0.2647ms	0.1397ms	7.1578 KOps/s	6.9916 KOps/s	$\color{#35bf28}+2.38\%$
test_values	8.2454μs	1.0495μs	952.8735 KOps/s	956.5693 KOps/s	$\color{#d91a1a}-0.39\%$
test_values_nested	0.1619ms	94.5524μs	10.5762 KOps/s	10.6399 KOps/s	$\color{#d91a1a}-0.60\%$
test_values_nested_locked	0.1773ms	94.5683μs	10.5744 KOps/s	10.6489 KOps/s	$\color{#d91a1a}-0.70\%$
test_values_nested_leaf	0.1331ms	81.0676μs	12.3354 KOps/s	12.5583 KOps/s	$\color{#d91a1a}-1.77\%$
test_values_stack_nested	0.2103ms	94.8277μs	10.5454 KOps/s	10.5714 KOps/s	$\color{#d91a1a}-0.25\%$
test_values_stack_nested_leaf	0.1528ms	82.0013μs	12.1949 KOps/s	12.6633 KOps/s	$\color{#d91a1a}-3.70\%$
test_values_stack_nested_locked	0.1834ms	94.9495μs	10.5319 KOps/s	10.6312 KOps/s	$\color{#d91a1a}-0.93\%$
test_membership	11.4920μs	0.8847μs	1.1303 MOps/s	1.1335 MOps/s	$\color{#d91a1a}-0.28\%$
test_membership_nested	24.2650μs	2.8039μs	356.6401 KOps/s	365.6378 KOps/s	$\color{#d91a1a}-2.46\%$
test_membership_nested_leaf	18.9850μs	2.7493μs	363.7224 KOps/s	361.8543 KOps/s	$\color{#35bf28}+0.52\%$
test_membership_stacked_nested	35.9180μs	2.7506μs	363.5526 KOps/s	372.0569 KOps/s	$\color{#d91a1a}-2.29\%$
test_membership_stacked_nested_leaf	19.6670μs	2.7674μs	361.3436 KOps/s	363.1602 KOps/s	$\color{#d91a1a}-0.50\%$
test_membership_nested_last	26.5700μs	4.2236μs	236.7667 KOps/s	237.6647 KOps/s	$\color{#d91a1a}-0.38\%$
test_membership_nested_leaf_last	22.8430μs	4.2362μs	236.0623 KOps/s	236.4647 KOps/s	$\color{#d91a1a}-0.17\%$
test_membership_stacked_nested_last	27.1400μs	4.2089μs	237.5913 KOps/s	202.1277 KOps/s	$\textbf{\color{#35bf28}+17.55\%}$
test_membership_stacked_nested_leaf_last	25.4870μs	4.1963μs	238.3056 KOps/s	200.7919 KOps/s	$\textbf{\color{#35bf28}+18.68\%}$
test_nested_getleaf	34.8950μs	10.5320μs	94.9483 KOps/s	92.4410 KOps/s	$\color{#35bf28}+2.71\%$
test_nested_get	31.1580μs	9.8692μs	101.3249 KOps/s	97.8560 KOps/s	$\color{#35bf28}+3.54\%$
test_stacked_getleaf	47.2580μs	10.3961μs	96.1902 KOps/s	93.1325 KOps/s	$\color{#35bf28}+3.28\%$
test_stacked_get	30.5270μs	9.9380μs	100.6235 KOps/s	98.4349 KOps/s	$\color{#35bf28}+2.22\%$
test_nested_getitemleaf	34.0940μs	10.9460μs	91.3572 KOps/s	87.8147 KOps/s	$\color{#35bf28}+4.03\%$
test_nested_getitem	42.1090μs	10.2798μs	97.2785 KOps/s	94.5554 KOps/s	$\color{#35bf28}+2.88\%$
test_stacked_getitemleaf	38.4720μs	11.0118μs	90.8119 KOps/s	89.2465 KOps/s	$\color{#35bf28}+1.75\%$
test_stacked_getitem	51.3590μs	10.0426μs	99.5760 KOps/s	96.6958 KOps/s	$\color{#35bf28}+2.98\%$
test_lock_nested	92.3414ms	0.6084ms	1.6438 KOps/s	1.9744 KOps/s	$\textbf{\color{#d91a1a}-16.75\%}$
test_lock_stack_nested	0.6091ms	0.4824ms	2.0731 KOps/s	2.1271 KOps/s	$\color{#d91a1a}-2.54\%$
test_unlock_nested	94.5816ms	0.5305ms	1.8850 KOps/s	2.3693 KOps/s	$\textbf{\color{#d91a1a}-20.44\%}$
test_unlock_stack_nested	0.6878ms	0.3961ms	2.5246 KOps/s	2.5963 KOps/s	$\color{#d91a1a}-2.76\%$
test_flatten_speed	0.2155ms	0.1028ms	9.7271 KOps/s	9.9148 KOps/s	$\color{#d91a1a}-1.89\%$
test_unflatten_speed	1.0408ms	0.5251ms	1.9042 KOps/s	1.9196 KOps/s	$\color{#d91a1a}-0.80\%$
test_common_ops	3.7642ms	1.2029ms	831.3564 Ops/s	883.3843 Ops/s	$\textbf{\color{#d91a1a}-5.89\%}$
test_creation	21.8900μs	2.0905μs	478.3546 KOps/s	468.6534 KOps/s	$\color{#35bf28}+2.07\%$
test_creation_empty	59.3700μs	20.6645μs	48.3923 KOps/s	54.3064 KOps/s	$\textbf{\color{#d91a1a}-10.89\%}$
test_creation_nested_1	67.6360μs	22.7572μs	43.9421 KOps/s	46.5487 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_creation_nested_2	69.3790μs	28.0882μs	35.6022 KOps/s	38.2196 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_clone	0.1288ms	16.7101μs	59.8442 KOps/s	59.2513 KOps/s	$\color{#35bf28}+1.00\%$
test_getitem[int]	1.3485ms	16.9369μs	59.0426 KOps/s	60.3939 KOps/s	$\color{#d91a1a}-2.24\%$
test_getitem[slice_int]	0.1415ms	30.3537μs	32.9449 KOps/s	32.3013 KOps/s	$\color{#35bf28}+1.99\%$
test_getitem[range]	0.3005ms	58.2066μs	17.1802 KOps/s	16.7597 KOps/s	$\color{#35bf28}+2.51\%$
test_getitem[tuple]	0.1331ms	25.4545μs	39.2858 KOps/s	40.1834 KOps/s	$\color{#d91a1a}-2.23\%$
test_getitem[list]	0.2483ms	54.2026μs	18.4493 KOps/s	18.5125 KOps/s	$\color{#d91a1a}-0.34\%$
test_setitem_dim[int]	66.2030μs	33.6136μs	29.7498 KOps/s	31.3030 KOps/s	$\color{#d91a1a}-4.96\%$
test_setitem_dim[slice_int]	0.1211ms	61.6899μs	16.2101 KOps/s	16.1250 KOps/s	$\color{#35bf28}+0.53\%$
test_setitem_dim[range]	0.1245ms	85.6428μs	11.6764 KOps/s	11.6761 KOps/s	$+0.00\%$
test_setitem_dim[tuple]	0.1151ms	49.6002μs	20.1612 KOps/s	20.3469 KOps/s	$\color{#d91a1a}-0.91\%$
test_setitem	0.3345ms	31.1846μs	32.0671 KOps/s	33.6098 KOps/s	$\color{#d91a1a}-4.59\%$
test_set	0.2464ms	29.7124μs	33.6560 KOps/s	33.5810 KOps/s	$\color{#35bf28}+0.22\%$
test_set_shared	3.8888ms	0.2193ms	4.5589 KOps/s	4.6096 KOps/s	$\color{#d91a1a}-1.10\%$
test_update	0.1539ms	39.0935μs	25.5797 KOps/s	26.6731 KOps/s	$\color{#d91a1a}-4.10\%$
test_update_nested	0.1734ms	51.1274μs	19.5590 KOps/s	20.4796 KOps/s	$\color{#d91a1a}-4.50\%$
test_update__nested	0.1257ms	37.8373μs	26.4290 KOps/s	26.5791 KOps/s	$\color{#d91a1a}-0.57\%$
test_set_nested	0.1905ms	33.7479μs	29.6314 KOps/s	30.8000 KOps/s	$\color{#d91a1a}-3.79\%$
test_set_nested_new	94.5670μs	39.1500μs	25.5428 KOps/s	26.7531 KOps/s	$\color{#d91a1a}-4.52\%$
test_select	0.1138ms	55.0728μs	18.1578 KOps/s	17.9496 KOps/s	$\color{#35bf28}+1.16\%$
test_select_nested	0.1466ms	62.0866μs	16.1065 KOps/s	16.7524 KOps/s	$\color{#d91a1a}-3.86\%$
test_exclude_nested	0.1427ms	76.0495μs	13.1493 KOps/s	13.1429 KOps/s	$\color{#35bf28}+0.05\%$
test_empty[True]	0.5005ms	0.3502ms	2.8551 KOps/s	2.8382 KOps/s	$\color{#35bf28}+0.59\%$
test_empty[False]	6.7225μs	1.2577μs	795.1270 KOps/s	820.1906 KOps/s	$\color{#d91a1a}-3.06\%$
test_unbind_speed	0.4967ms	0.3007ms	3.3260 KOps/s	3.3170 KOps/s	$\color{#35bf28}+0.27\%$
test_unbind_speed_stack0	1.0614ms	0.3109ms	3.2161 KOps/s	3.3692 KOps/s	$\color{#d91a1a}-4.54\%$
test_unbind_speed_stack1	99.1491ms	0.8289ms	1.2065 KOps/s	1.3406 KOps/s	$\textbf{\color{#d91a1a}-10.00\%}$
test_split	3.1870ms	1.9782ms	505.5108 Ops/s	446.2051 Ops/s	$\textbf{\color{#35bf28}+13.29\%}$
test_chunk	0.1007s	2.1897ms	456.6854 Ops/s	445.3796 Ops/s	$\color{#35bf28}+2.54\%$
test_creation[device0]	0.2286ms	0.1170ms	8.5456 KOps/s	8.3852 KOps/s	$\color{#35bf28}+1.91\%$
test_creation_from_tensor	3.8654ms	0.1194ms	8.3741 KOps/s	8.5373 KOps/s	$\color{#d91a1a}-1.91\%$
test_add_one[memmap_tensor0]	0.3152ms	7.2997μs	136.9920 KOps/s	133.1742 KOps/s	$\color{#35bf28}+2.87\%$
test_contiguous[memmap_tensor0]	15.2580μs	1.8785μs	532.3255 KOps/s	502.0540 KOps/s	$\textbf{\color{#35bf28}+6.03\%}$
test_stack[memmap_tensor0]	79.0970μs	5.4358μs	183.9672 KOps/s	173.4378 KOps/s	$\textbf{\color{#35bf28}+6.07\%}$
test_memmaptd_index	1.1137ms	0.4261ms	2.3468 KOps/s	2.3987 KOps/s	$\color{#d91a1a}-2.16\%$
test_memmaptd_index_astensor	0.9965ms	0.5332ms	1.8754 KOps/s	1.9216 KOps/s	$\color{#d91a1a}-2.41\%$
test_memmaptd_index_op	3.4557ms	1.1142ms	897.5006 Ops/s	907.9353 Ops/s	$\color{#d91a1a}-1.15\%$
test_serialize_model	0.1284s	0.1252s	7.9894 Ops/s	8.3835 Ops/s	$\color{#d91a1a}-4.70\%$
test_serialize_model_pickle	0.4714s	0.4010s	2.4935 Ops/s	2.5575 Ops/s	$\color{#d91a1a}-2.50\%$
test_serialize_weights	0.1276s	0.1176s	8.5040 Ops/s	8.4288 Ops/s	$\color{#35bf28}+0.89\%$
test_serialize_weights_returnearly	0.1807s	0.1665s	6.0058 Ops/s	6.1218 Ops/s	$\color{#d91a1a}-1.90\%$
test_serialize_weights_pickle	0.4579s	0.4139s	2.4159 Ops/s	2.4854 Ops/s	$\color{#d91a1a}-2.79\%$
test_serialize_weights_filesystem	0.1547s	0.1457s	6.8619 Ops/s	6.9823 Ops/s	$\color{#d91a1a}-1.72\%$
test_serialize_model_filesystem	0.1665s	0.1524s	6.5633 Ops/s	5.6947 Ops/s	$\textbf{\color{#35bf28}+15.25\%}$
test_reshape_pytree	0.1072ms	38.3799μs	26.0553 KOps/s	25.3927 KOps/s	$\color{#35bf28}+2.61\%$
test_reshape_td	92.4630μs	46.0831μs	21.6999 KOps/s	21.0170 KOps/s	$\color{#35bf28}+3.25\%$
test_view_pytree	84.1670μs	38.0042μs	26.3129 KOps/s	25.5270 KOps/s	$\color{#35bf28}+3.08\%$
test_view_td	0.1075ms	52.2361μs	19.1438 KOps/s	18.8815 KOps/s	$\color{#35bf28}+1.39\%$
test_unbind_pytree	72.5360μs	35.8799μs	27.8707 KOps/s	27.3675 KOps/s	$\color{#35bf28}+1.84\%$
test_unbind_td	0.3351ms	45.7947μs	21.8366 KOps/s	22.4236 KOps/s	$\color{#d91a1a}-2.62\%$
test_split_pytree	87.7630μs	38.4455μs	26.0109 KOps/s	26.0980 KOps/s	$\color{#d91a1a}-0.33\%$
test_split_td	0.4959ms	59.1452μs	16.9076 KOps/s	17.1000 KOps/s	$\color{#d91a1a}-1.13\%$
test_add_pytree	0.1541ms	44.4951μs	22.4744 KOps/s	21.8177 KOps/s	$\color{#35bf28}+3.01\%$
test_add_td	0.1704ms	91.0156μs	10.9871 KOps/s	11.0734 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_add_one_nested[tensordict-compile]	0.1376ms	59.0796μs	16.9263 KOps/s	16.8796 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_add_one_nested[tensordict-eager]	0.4076ms	0.2018ms	4.9555 KOps/s	5.0168 KOps/s	$\color{#d91a1a}-1.22\%$
test_compile_add_one_nested[pytree-compile]	0.1180ms	58.3211μs	17.1464 KOps/s	17.4361 KOps/s	$\color{#d91a1a}-1.66\%$
test_compile_add_one_nested[pytree-eager]	0.2444ms	0.1395ms	7.1673 KOps/s	7.0115 KOps/s	$\color{#35bf28}+2.22\%$
test_compile_copy_nested[tensordict-compile]	59.1510μs	23.7441μs	42.1158 KOps/s	41.4916 KOps/s	$\color{#35bf28}+1.50\%$
test_compile_copy_nested[tensordict-eager]	0.1493ms	75.5966μs	13.2281 KOps/s	13.0299 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_copy_nested[pytree-compile]	0.1438ms	76.3003μs	13.1061 KOps/s	13.0167 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_copy_nested[pytree-eager]	0.1256ms	69.2352μs	14.4435 KOps/s	14.5789 KOps/s	$\color{#d91a1a}-0.93\%$
test_compile_add_one_flat[tensordict-compile]	0.2899ms	0.1830ms	5.4641 KOps/s	5.5051 KOps/s	$\color{#d91a1a}-0.74\%$
test_compile_add_one_flat[tensordict-eager]	0.4318ms	0.2401ms	4.1655 KOps/s	4.1420 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_add_one_flat[tensorclass-compile]	0.1162ms	50.0054μs	19.9979 KOps/s	21.1822 KOps/s	$\textbf{\color{#d91a1a}-5.59\%}$
test_compile_add_one_flat[tensorclass-eager]	0.1810ms	78.2313μs	12.7826 KOps/s	12.7067 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_add_one_flat[pytree-compile]	0.3634ms	0.1774ms	5.6372 KOps/s	5.6842 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_one_flat[pytree-eager]	1.1721ms	0.2881ms	3.4708 KOps/s	3.4349 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_add_self_flat[tensordict-eager]	0.4340ms	0.2767ms	3.6142 KOps/s	3.5999 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_add_self_flat[tensordict-compile]	0.7029ms	0.1859ms	5.3779 KOps/s	5.4839 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_add_self_flat[tensorclass-eager]	0.1640ms	73.6760μs	13.5729 KOps/s	13.8849 KOps/s	$\color{#d91a1a}-2.25\%$
test_compile_add_self_flat[tensorclass-compile]	0.1198ms	48.6884μs	20.5388 KOps/s	20.8385 KOps/s	$\color{#d91a1a}-1.44\%$
test_compile_add_self_flat[pytree-eager]	0.4876ms	0.2325ms	4.3019 KOps/s	4.3011 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_add_self_flat[pytree-compile]	0.3581ms	0.1775ms	5.6329 KOps/s	5.6927 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_copy_flat[tensordict-compile]	0.1961ms	0.1114ms	8.9736 KOps/s	9.0154 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_copy_flat[tensordict-eager]	0.1757ms	80.0509μs	12.4921 KOps/s	12.6223 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_copy_flat[pytree-compile]	0.1486ms	80.2075μs	12.4677 KOps/s	12.5683 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_copy_flat[pytree-eager]	0.1467ms	71.4195μs	14.0018 KOps/s	14.1988 KOps/s	$\color{#d91a1a}-1.39\%$
test_compile_assign_and_add[tensordict-compile]	0.2959ms	0.1962ms	5.0957 KOps/s	5.0471 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_assign_and_add[tensordict-eager]	2.8433ms	1.7779ms	562.4745 Ops/s	548.3042 Ops/s	$\color{#35bf28}+2.58\%$
test_compile_assign_and_add[pytree-compile]	0.2849ms	0.1940ms	5.1550 KOps/s	5.0354 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_assign_and_add[pytree-eager]	2.2772ms	1.0890ms	918.2552 Ops/s	900.5238 Ops/s	$\color{#35bf28}+1.97\%$
test_compile_assign_and_add_stack[compile]	0.7614ms	0.4238ms	2.3596 KOps/s	2.3926 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_assign_and_add_stack[eager]	5.8922ms	4.0792ms	245.1471 Ops/s	243.9153 Ops/s	$\color{#35bf28}+0.51\%$
test_compile_indexing[tensor-tensordict-compile]	86.6310μs	35.1167μs	28.4765 KOps/s	28.9887 KOps/s	$\color{#d91a1a}-1.77\%$
test_compile_indexing[tensor-tensordict-eager]	0.6502ms	49.0390μs	20.3919 KOps/s	20.7321 KOps/s	$\color{#d91a1a}-1.64\%$
test_compile_indexing[tensor-tensorclass-compile]	84.0460μs	30.8449μs	32.4202 KOps/s	34.3956 KOps/s	$\textbf{\color{#d91a1a}-5.74\%}$
test_compile_indexing[tensor-tensorclass-eager]	80.1190μs	29.4993μs	33.8991 KOps/s	35.6464 KOps/s	$\color{#d91a1a}-4.90\%$
test_compile_indexing[tensor-pytree-compile]	92.8730μs	30.6776μs	32.5971 KOps/s	35.0269 KOps/s	$\textbf{\color{#d91a1a}-6.94\%}$
test_compile_indexing[tensor-pytree-eager]	83.3760μs	29.1559μs	34.2984 KOps/s	35.3540 KOps/s	$\color{#d91a1a}-2.99\%$
test_compile_indexing[slice-tensordict-compile]	0.1831ms	75.2918μs	13.2817 KOps/s	13.7183 KOps/s	$\color{#d91a1a}-3.18\%$
test_compile_indexing[slice-tensordict-eager]	0.4707ms	27.6404μs	36.1789 KOps/s	35.7047 KOps/s	$\color{#35bf28}+1.33\%$
test_compile_indexing[slice-tensorclass-compile]	0.1479ms	69.0636μs	14.4794 KOps/s	14.4761 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_indexing[slice-tensorclass-eager]	67.9960μs	23.0358μs	43.4106 KOps/s	42.0676 KOps/s	$\color{#35bf28}+3.19\%$
test_compile_indexing[slice-pytree-compile]	0.1448ms	68.6744μs	14.5615 KOps/s	14.6405 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_indexing[slice-pytree-eager]	69.5500μs	22.4587μs	44.5261 KOps/s	41.8224 KOps/s	$\textbf{\color{#35bf28}+6.46\%}$
test_compile_indexing[int-tensordict-compile]	0.1696ms	74.2201μs	13.4734 KOps/s	13.5974 KOps/s	$\color{#d91a1a}-0.91\%$
test_compile_indexing[int-tensordict-eager]	0.8665ms	27.8863μs	35.8599 KOps/s	36.6668 KOps/s	$\color{#d91a1a}-2.20\%$
test_compile_indexing[int-tensorclass-compile]	0.1606ms	69.4711μs	14.3945 KOps/s	14.7371 KOps/s	$\color{#d91a1a}-2.32\%$
test_compile_indexing[int-tensorclass-eager]	78.6870μs	22.9848μs	43.5070 KOps/s	43.0328 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_indexing[int-pytree-compile]	0.1731ms	69.7711μs	14.3326 KOps/s	14.7068 KOps/s	$\color{#d91a1a}-2.54\%$
test_compile_indexing[int-pytree-eager]	63.7690μs	22.7694μs	43.9186 KOps/s	43.2730 KOps/s	$\color{#35bf28}+1.49\%$
test_mod_add[eager]	98.2540μs	27.4453μs	36.4361 KOps/s	38.7567 KOps/s	$\textbf{\color{#d91a1a}-5.99\%}$
test_mod_add[compile]	0.1251ms	40.8816μs	24.4609 KOps/s	26.1284 KOps/s	$\textbf{\color{#d91a1a}-6.38\%}$
test_mod_add[compile-overhead]	0.1087ms	39.8464μs	25.0964 KOps/s	26.4273 KOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_mod_wrap[eager]	0.4091ms	0.2117ms	4.7236 KOps/s	4.6189 KOps/s	$\color{#35bf28}+2.27\%$
test_mod_wrap[compile]	0.4601ms	0.2389ms	4.1858 KOps/s	4.3058 KOps/s	$\color{#d91a1a}-2.79\%$
test_mod_wrap[compile-overhead]	0.3184ms	0.2306ms	4.3363 KOps/s	4.3457 KOps/s	$\color{#d91a1a}-0.22\%$
test_mod_wrap_and_backward[eager]	18.1342ms	11.4303ms	87.4870 Ops/s	90.2938 Ops/s	$\color{#d91a1a}-3.11\%$
test_mod_wrap_and_backward[compile]	12.9822ms	11.2508ms	88.8826 Ops/s	82.6183 Ops/s	$\textbf{\color{#35bf28}+7.58\%}$
test_mod_wrap_and_backward[compile-overhead]	20.5894ms	11.9029ms	84.0128 Ops/s	79.6477 Ops/s	$\textbf{\color{#35bf28}+5.48\%}$
test_seq_add[eager]	0.1777ms	94.4210μs	10.5909 KOps/s	10.8725 KOps/s	$\color{#d91a1a}-2.59\%$
test_seq_add[compile]	0.1738ms	65.5533μs	15.2548 KOps/s	15.1196 KOps/s	$\color{#35bf28}+0.89\%$
test_seq_add[compile-overhead]	0.1255ms	64.1788μs	15.5815 KOps/s	15.6294 KOps/s	$\color{#d91a1a}-0.31\%$
test_seq_wrap[eager]	0.5088ms	0.3905ms	2.5610 KOps/s	2.5482 KOps/s	$\color{#35bf28}+0.50\%$
test_seq_wrap[compile]	1.1726ms	0.2725ms	3.6703 KOps/s	3.6823 KOps/s	$\color{#d91a1a}-0.33\%$
test_seq_wrap[compile-overhead]	1.2923ms	0.2719ms	3.6774 KOps/s	3.7221 KOps/s	$\color{#d91a1a}-1.20\%$
test_func_call_runtime[False-eager]	1.2136ms	0.5249ms	1.9050 KOps/s	1.8915 KOps/s	$\color{#35bf28}+0.71\%$
test_func_call_runtime[False-compile]	0.6212ms	0.5016ms	1.9936 KOps/s	1.9838 KOps/s	$\color{#35bf28}+0.49\%$
test_func_call_runtime[False-compile-overhead]	0.9442ms	0.5026ms	1.9897 KOps/s	1.9491 KOps/s	$\color{#35bf28}+2.09\%$
test_func_call_runtime[True-eager]	0.9742ms	0.7370ms	1.3569 KOps/s	1.3168 KOps/s	$\color{#35bf28}+3.04\%$
test_func_call_runtime[True-compile]	1.0665ms	0.5166ms	1.9356 KOps/s	1.9358 KOps/s	$-0.01\%$
test_func_call_runtime[True-compile-overhead]	0.8900ms	0.5182ms	1.9299 KOps/s	1.9278 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[False-eager]	0.6635ms	0.5207ms	1.9205 KOps/s	1.8757 KOps/s	$\color{#35bf28}+2.39\%$
test_func_call_cm_runtime[False-compile]	0.6250ms	0.5024ms	1.9906 KOps/s	1.9928 KOps/s	$\color{#d91a1a}-0.11\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7084ms	0.5025ms	1.9902 KOps/s	2.0008 KOps/s	$\color{#d91a1a}-0.53\%$
test_func_call_cm_runtime[True-eager]	1.1428ms	0.8940ms	1.1186 KOps/s	1.1142 KOps/s	$\color{#35bf28}+0.40\%$
test_func_call_cm_runtime[True-compile]	0.8341ms	0.7277ms	1.3741 KOps/s	1.3176 KOps/s	$\color{#35bf28}+4.28\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8267ms	0.7278ms	1.3739 KOps/s	1.3057 KOps/s	$\textbf{\color{#35bf28}+5.22\%}$
test_vmap_func_call_cm_runtime[eager]	2.5954ms	1.9085ms	523.9823 Ops/s	513.6400 Ops/s	$\color{#35bf28}+2.01\%$
test_vmap_func_call_cm_runtime[compile]	2.8703ms	1.9824ms	504.4494 Ops/s	503.4518 Ops/s	$\color{#35bf28}+0.20\%$
test_vmap_func_call_cm_runtime[compile-overhead]	2.8193ms	1.9782ms	505.5135 Ops/s	497.4608 Ops/s	$\color{#35bf28}+1.62\%$
test_distributed	1.2800ms	0.1254ms	7.9776 KOps/s	7.8513 KOps/s	$\color{#35bf28}+1.61\%$
test_tdmodule	37.1090μs	19.0915μs	52.3794 KOps/s	55.3453 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_tdmodule_dispatch	77.5450μs	38.2396μs	26.1509 KOps/s	27.3383 KOps/s	$\color{#d91a1a}-4.34\%$
test_tdseq	55.2030μs	22.6183μs	44.2120 KOps/s	47.2381 KOps/s	$\textbf{\color{#d91a1a}-6.41\%}$
test_tdseq_dispatch	81.4220μs	43.5889μs	22.9416 KOps/s	22.6581 KOps/s	$\color{#35bf28}+1.25\%$
test_instantiation_functorch	2.0902ms	1.5742ms	635.2373 Ops/s	618.8433 Ops/s	$\color{#35bf28}+2.65\%$
test_instantiation_td	2.0978ms	1.1829ms	845.3772 Ops/s	839.5183 Ops/s	$\color{#35bf28}+0.70\%$
test_exec_functorch	0.3321ms	0.1858ms	5.3811 KOps/s	5.3246 KOps/s	$\color{#35bf28}+1.06\%$
test_exec_functional_call	0.4228ms	0.1777ms	5.6263 KOps/s	5.6700 KOps/s	$\color{#d91a1a}-0.77\%$
test_exec_td	0.3453ms	0.2025ms	4.9384 KOps/s	4.9148 KOps/s	$\color{#35bf28}+0.48\%$
test_exec_td_decorator	0.4370ms	0.2335ms	4.2830 KOps/s	4.1311 KOps/s	$\color{#35bf28}+3.68\%$
test_vmap_mlp_speed[True-True]	1.1484ms	0.7017ms	1.4251 KOps/s	1.4389 KOps/s	$\color{#d91a1a}-0.96\%$
test_vmap_mlp_speed[True-False]	1.0829ms	0.6940ms	1.4409 KOps/s	1.4628 KOps/s	$\color{#d91a1a}-1.50\%$
test_vmap_mlp_speed[False-True]	0.8396ms	0.5446ms	1.8363 KOps/s	1.8722 KOps/s	$\color{#d91a1a}-1.91\%$
test_vmap_mlp_speed[False-False]	0.7726ms	0.5418ms	1.8456 KOps/s	1.8618 KOps/s	$\color{#d91a1a}-0.87\%$
test_vmap_mlp_speed_decorator[True-True]	1.7103ms	0.6547ms	1.5274 KOps/s	1.5425 KOps/s	$\color{#d91a1a}-0.98\%$
test_vmap_mlp_speed_decorator[True-False]	0.9149ms	0.6585ms	1.5185 KOps/s	1.5610 KOps/s	$\color{#d91a1a}-2.73\%$
test_vmap_mlp_speed_decorator[False-True]	0.8546ms	0.5437ms	1.8391 KOps/s	1.8895 KOps/s	$\color{#d91a1a}-2.66\%$
test_vmap_mlp_speed_decorator[False-False]	0.7897ms	0.5414ms	1.8471 KOps/s	1.8632 KOps/s	$\color{#d91a1a}-0.87\%$
test_to_module_speed[True]	2.2808ms	1.4499ms	689.7108 Ops/s	718.5362 Ops/s	$\color{#d91a1a}-4.01\%$
test_to_module_speed[False]	2.1745ms	1.4010ms	713.7868 Ops/s	720.6253 Ops/s	$\color{#d91a1a}-0.95\%$
test_tc_init	83.2650μs	47.6458μs	20.9882 KOps/s	21.4992 KOps/s	$\color{#d91a1a}-2.38\%$
test_tc_init_nested	0.1756ms	94.5411μs	10.5774 KOps/s	10.6038 KOps/s	$\color{#d91a1a}-0.25\%$
test_tc_first_layer_tensor	39.3540μs	1.5077μs	663.2654 KOps/s	652.8907 KOps/s	$\color{#35bf28}+1.59\%$
test_tc_first_layer_nontensor	30.8770μs	4.7269μs	211.5540 KOps/s	211.4870 KOps/s	$\color{#35bf28}+0.03\%$
test_tc_second_layer_tensor	38.7320μs	2.8248μs	354.0030 KOps/s	352.1838 KOps/s	$\color{#35bf28}+0.52\%$
test_tc_second_layer_nontensor	44.4770μs	5.9916μs	166.9013 KOps/s	166.1793 KOps/s	$\color{#35bf28}+0.43\%$
test_unbind	0.5059s	13.3103ms	75.1299 Ops/s	73.5796 Ops/s	$\color{#35bf28}+2.11\%$
test_full_like	9.7996ms	8.7347ms	114.4856 Ops/s	78.9457 Ops/s	$\textbf{\color{#35bf28}+45.02\%}$
test_zeros_like	4.0593ms	3.3310ms	300.2085 Ops/s	148.6452 Ops/s	$\textbf{\color{#35bf28}+101.96\%}$
test_ones_like	4.9487ms	3.8708ms	258.3457 Ops/s	127.1683 Ops/s	$\textbf{\color{#35bf28}+103.15\%}$
test_clone	8.6125ms	6.4309ms	155.4988 Ops/s	100.2305 Ops/s	$\textbf{\color{#35bf28}+55.14\%}$
test_squeeze	70.7730μs	12.8876μs	77.5940 KOps/s	82.3584 KOps/s	$\textbf{\color{#d91a1a}-5.79\%}$
test_unsqueeze	0.1537ms	94.0566μs	10.6319 KOps/s	10.7513 KOps/s	$\color{#d91a1a}-1.11\%$
test_split	0.6148ms	0.1993ms	5.0166 KOps/s	5.0330 KOps/s	$\color{#d91a1a}-0.33\%$
test_permute	0.3685ms	0.2240ms	4.4634 KOps/s	4.3836 KOps/s	$\color{#35bf28}+1.82\%$
test_stack	35.7369ms	28.0963ms	35.5919 Ops/s	37.7997 Ops/s	$\textbf{\color{#d91a1a}-5.84\%}$
test_cat	31.9578ms	27.0063ms	37.0284 Ops/s	39.2863 Ops/s	$\textbf{\color{#d91a1a}-5.75\%}$

[github-actions]

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

Total Benchmarks: 228. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1574ms	16.5035μs	60.5933 KOps/s	60.4300 KOps/s	$\color{#35bf28}+0.27\%$
test_plain_set_stack_nested	51.2330μs	16.5163μs	60.5462 KOps/s	60.1937 KOps/s	$\color{#35bf28}+0.59\%$
test_plain_set_nested_inplace	58.5630μs	18.0171μs	55.5029 KOps/s	57.2009 KOps/s	$\color{#d91a1a}-2.97\%$
test_plain_set_stack_nested_inplace	55.2740μs	17.6303μs	56.7206 KOps/s	56.8657 KOps/s	$\color{#d91a1a}-0.26\%$
test_items	38.3720μs	2.8842μs	346.7115 KOps/s	346.6003 KOps/s	$\color{#35bf28}+0.03\%$
test_items_nested	0.3927ms	0.3427ms	2.9180 KOps/s	2.9787 KOps/s	$\color{#d91a1a}-2.04\%$
test_items_nested_locked	0.4399ms	0.3436ms	2.9106 KOps/s	2.9522 KOps/s	$\color{#d91a1a}-1.41\%$
test_items_nested_leaf	92.5360μs	62.8090μs	15.9213 KOps/s	15.9128 KOps/s	$\color{#35bf28}+0.05\%$
test_items_stack_nested	0.5212ms	0.3443ms	2.9047 KOps/s	2.9635 KOps/s	$\color{#d91a1a}-1.98\%$
test_items_stack_nested_leaf	0.1168ms	63.8798μs	15.6544 KOps/s	15.7399 KOps/s	$\color{#d91a1a}-0.54\%$
test_items_stack_nested_locked	0.3969ms	0.3423ms	2.9210 KOps/s	2.9486 KOps/s	$\color{#d91a1a}-0.94\%$
test_keys	36.2820μs	3.4704μs	288.1492 KOps/s	290.8778 KOps/s	$\color{#d91a1a}-0.94\%$
test_keys_nested	0.1586ms	70.6595μs	14.1524 KOps/s	14.0699 KOps/s	$\color{#35bf28}+0.59\%$
test_keys_nested_locked	2.2858ms	77.1024μs	12.9698 KOps/s	12.9642 KOps/s	$\color{#35bf28}+0.04\%$
test_keys_nested_leaf	98.2160μs	61.6988μs	16.2078 KOps/s	15.9574 KOps/s	$\color{#35bf28}+1.57\%$
test_keys_stack_nested	0.1082ms	71.7801μs	13.9314 KOps/s	13.8388 KOps/s	$\color{#35bf28}+0.67\%$
test_keys_stack_nested_leaf	0.1001ms	62.5616μs	15.9842 KOps/s	15.7766 KOps/s	$\color{#35bf28}+1.32\%$
test_keys_stack_nested_locked	0.1257ms	78.4122μs	12.7531 KOps/s	12.8192 KOps/s	$\color{#d91a1a}-0.52\%$
test_values	7.3822μs	0.8732μs	1.1452 MOps/s	1.1690 MOps/s	$\color{#d91a1a}-2.04\%$
test_values_nested	0.1084ms	48.6051μs	20.5740 KOps/s	20.6216 KOps/s	$\color{#d91a1a}-0.23\%$
test_values_nested_locked	0.1313ms	50.9203μs	19.6385 KOps/s	19.9324 KOps/s	$\color{#d91a1a}-1.47\%$
test_values_nested_leaf	72.8650μs	42.7116μs	23.4129 KOps/s	23.5625 KOps/s	$\color{#d91a1a}-0.64\%$
test_values_stack_nested	75.5950μs	49.5490μs	20.1820 KOps/s	20.2433 KOps/s	$\color{#d91a1a}-0.30\%$
test_values_stack_nested_leaf	0.1852ms	43.4629μs	23.0082 KOps/s	22.8352 KOps/s	$\color{#35bf28}+0.76\%$
test_values_stack_nested_locked	0.2081ms	51.5410μs	19.4020 KOps/s	19.4863 KOps/s	$\color{#d91a1a}-0.43\%$
test_membership	1.7511μs	0.5005μs	1.9979 MOps/s	1.9881 MOps/s	$\color{#35bf28}+0.49\%$
test_membership_nested	15.5660μs	1.9034μs	525.3891 KOps/s	534.3527 KOps/s	$\color{#d91a1a}-1.68\%$
test_membership_nested_leaf	15.0810μs	1.8817μs	531.4382 KOps/s	531.3718 KOps/s	$\color{#35bf28}+0.01\%$
test_membership_stacked_nested	27.8720μs	1.9361μs	516.5069 KOps/s	523.3216 KOps/s	$\color{#d91a1a}-1.30\%$
test_membership_stacked_nested_leaf	28.0320μs	1.9258μs	519.2764 KOps/s	525.4838 KOps/s	$\color{#d91a1a}-1.18\%$
test_membership_nested_last	36.2420μs	3.0027μs	333.0342 KOps/s	336.7693 KOps/s	$\color{#d91a1a}-1.11\%$
test_membership_nested_leaf_last	27.3720μs	3.0093μs	332.3084 KOps/s	336.5411 KOps/s	$\color{#d91a1a}-1.26\%$
test_membership_stacked_nested_last	27.4420μs	3.0045μs	332.8362 KOps/s	335.8854 KOps/s	$\color{#d91a1a}-0.91\%$
test_membership_stacked_nested_leaf_last	30.3620μs	2.9736μs	336.2918 KOps/s	337.8819 KOps/s	$\color{#d91a1a}-0.47\%$
test_nested_getleaf	40.6830μs	6.0761μs	164.5779 KOps/s	164.5934 KOps/s	$-0.01\%$
test_nested_get	52.4530μs	5.8024μs	172.3417 KOps/s	175.8022 KOps/s	$\color{#d91a1a}-1.97\%$
test_stacked_getleaf	32.3820μs	6.0391μs	165.5867 KOps/s	165.1358 KOps/s	$\color{#35bf28}+0.27\%$
test_stacked_get	38.8320μs	5.7046μs	175.2985 KOps/s	176.6232 KOps/s	$\color{#d91a1a}-0.75\%$
test_nested_getitemleaf	71.4350μs	6.1201μs	163.3952 KOps/s	161.7896 KOps/s	$\color{#35bf28}+0.99\%$
test_nested_getitem	36.3820μs	5.8524μs	170.8700 KOps/s	173.6520 KOps/s	$\color{#d91a1a}-1.60\%$
test_stacked_getitemleaf	43.6430μs	6.1477μs	162.6615 KOps/s	164.6066 KOps/s	$\color{#d91a1a}-1.18\%$
test_stacked_getitem	31.5120μs	5.7719μs	173.2518 KOps/s	175.1236 KOps/s	$\color{#d91a1a}-1.07\%$
test_lock_nested	4.6138ms	0.4323ms	2.3132 KOps/s	2.3310 KOps/s	$\color{#d91a1a}-0.76\%$
test_lock_stack_nested	0.5956ms	0.3975ms	2.5154 KOps/s	2.5384 KOps/s	$\color{#d91a1a}-0.91\%$
test_unlock_nested	0.7655ms	0.3655ms	2.7362 KOps/s	2.7637 KOps/s	$\color{#d91a1a}-0.99\%$
test_unlock_stack_nested	0.3783ms	0.3335ms	2.9982 KOps/s	3.0008 KOps/s	$\color{#d91a1a}-0.09\%$
test_flatten_speed	0.1298ms	76.9179μs	13.0009 KOps/s	13.0597 KOps/s	$\color{#d91a1a}-0.45\%$
test_unflatten_speed	0.3671ms	0.3204ms	3.1210 KOps/s	3.1084 KOps/s	$\color{#35bf28}+0.41\%$
test_common_ops	1.5268ms	1.2389ms	807.1687 Ops/s	798.6372 Ops/s	$\color{#35bf28}+1.07\%$
test_creation	33.6120μs	1.4831μs	674.2700 KOps/s	666.8217 KOps/s	$\color{#35bf28}+1.12\%$
test_creation_empty	45.7330μs	15.0754μs	66.3332 KOps/s	65.5202 KOps/s	$\color{#35bf28}+1.24\%$
test_creation_nested_1	44.9730μs	17.4614μs	57.2692 KOps/s	59.1297 KOps/s	$\color{#d91a1a}-3.15\%$
test_creation_nested_2	81.8050μs	19.5703μs	51.0978 KOps/s	51.1729 KOps/s	$\color{#d91a1a}-0.15\%$
test_clone	0.1990ms	28.2926μs	35.3449 KOps/s	35.9490 KOps/s	$\color{#d91a1a}-1.68\%$
test_getitem[int]	96.4474ms	24.3319μs	41.0984 KOps/s	59.7055 KOps/s	$\textbf{\color{#d91a1a}-31.16\%}$
test_getitem[slice_int]	0.1171ms	27.5460μs	36.3029 KOps/s	35.4500 KOps/s	$\color{#35bf28}+2.41\%$
test_getitem[range]	0.2239ms	0.1075ms	9.3010 KOps/s	9.0819 KOps/s	$\color{#35bf28}+2.41\%$
test_getitem[tuple]	0.1988ms	23.9279μs	41.7922 KOps/s	41.3427 KOps/s	$\color{#35bf28}+1.09\%$
test_getitem[list]	0.2533ms	96.9149μs	10.3183 KOps/s	10.3381 KOps/s	$\color{#d91a1a}-0.19\%$
test_setitem_dim[int]	60.3740μs	43.8625μs	22.7985 KOps/s	22.2879 KOps/s	$\color{#35bf28}+2.29\%$
test_setitem_dim[slice_int]	94.6060μs	66.6385μs	15.0063 KOps/s	14.7300 KOps/s	$\color{#35bf28}+1.88\%$
test_setitem_dim[range]	0.2295ms	0.1257ms	7.9575 KOps/s	7.8848 KOps/s	$\color{#35bf28}+0.92\%$
test_setitem_dim[tuple]	84.0850μs	60.8776μs	16.4264 KOps/s	16.2911 KOps/s	$\color{#35bf28}+0.83\%$
test_setitem	0.1882ms	41.2275μs	24.2557 KOps/s	24.1939 KOps/s	$\color{#35bf28}+0.26\%$
test_set	0.1915ms	40.0987μs	24.9385 KOps/s	25.0719 KOps/s	$\color{#d91a1a}-0.53\%$
test_set_shared	0.3568ms	53.0335μs	18.8560 KOps/s	19.1163 KOps/s	$\color{#d91a1a}-1.36\%$
test_update	0.2477ms	49.4928μs	20.2050 KOps/s	20.2311 KOps/s	$\color{#d91a1a}-0.13\%$
test_update_nested	0.2064ms	56.2155μs	17.7887 KOps/s	17.5493 KOps/s	$\color{#35bf28}+1.36\%$
test_update__nested	0.2101ms	59.8458μs	16.7096 KOps/s	16.4026 KOps/s	$\color{#35bf28}+1.87\%$
test_set_nested	0.1987ms	42.3409μs	23.6178 KOps/s	23.2307 KOps/s	$\color{#35bf28}+1.67\%$
test_set_nested_new	0.1974ms	45.8563μs	21.8072 KOps/s	21.3626 KOps/s	$\color{#35bf28}+2.08\%$
test_select	0.2445ms	59.1663μs	16.9015 KOps/s	16.6819 KOps/s	$\color{#35bf28}+1.32\%$
test_select_nested	0.3146ms	41.9903μs	23.8150 KOps/s	24.2433 KOps/s	$\color{#d91a1a}-1.77\%$
test_exclude_nested	0.2142ms	58.5848μs	17.0693 KOps/s	16.9499 KOps/s	$\color{#35bf28}+0.70\%$
test_empty[True]	0.2947ms	0.2576ms	3.8813 KOps/s	3.8257 KOps/s	$\color{#35bf28}+1.45\%$
test_empty[False]	3.2072μs	0.7463μs	1.3400 MOps/s	1.3363 MOps/s	$\color{#35bf28}+0.28\%$
test_to	0.1362ms	26.9491μs	37.1070 KOps/s	38.4378 KOps/s	$\color{#d91a1a}-3.46\%$
test_to_nonblocking	0.1627ms	25.7806μs	38.7889 KOps/s	39.5012 KOps/s	$\color{#d91a1a}-1.80\%$
test_unbind_speed	1.6495ms	0.2843ms	3.5175 KOps/s	3.5366 KOps/s	$\color{#d91a1a}-0.54\%$
test_unbind_speed_stack0	0.3449ms	0.2814ms	3.5537 KOps/s	3.5675 KOps/s	$\color{#d91a1a}-0.39\%$
test_unbind_speed_stack1	95.3655ms	0.7145ms	1.3995 KOps/s	1.3949 KOps/s	$\color{#35bf28}+0.33\%$
test_split	97.8299ms	2.1675ms	461.3667 Ops/s	446.7551 Ops/s	$\color{#35bf28}+3.27\%$
test_chunk	96.9350ms	2.1671ms	461.4559 Ops/s	447.1218 Ops/s	$\color{#35bf28}+3.21\%$
test_creation[device0]	0.2829ms	0.1280ms	7.8105 KOps/s	7.8200 KOps/s	$\color{#d91a1a}-0.12\%$
test_creation_from_tensor	0.3916ms	0.1301ms	7.6860 KOps/s	7.7163 KOps/s	$\color{#d91a1a}-0.39\%$
test_add_one[memmap_tensor0]	0.2387ms	9.1080μs	109.7933 KOps/s	110.5957 KOps/s	$\color{#d91a1a}-0.73\%$
test_contiguous[memmap_tensor0]	39.1820μs	2.2408μs	446.2677 KOps/s	449.2236 KOps/s	$\color{#d91a1a}-0.66\%$
test_stack[memmap_tensor0]	35.5920μs	6.6982μs	149.2938 KOps/s	148.2179 KOps/s	$\color{#35bf28}+0.73\%$
test_memmaptd_index	1.0948ms	0.4267ms	2.3437 KOps/s	2.2946 KOps/s	$\color{#35bf28}+2.14\%$
test_memmaptd_index_astensor	1.0387ms	0.4976ms	2.0095 KOps/s	2.0061 KOps/s	$\color{#35bf28}+0.17\%$
test_memmaptd_index_op	1.4528ms	1.0333ms	967.7491 Ops/s	958.6338 Ops/s	$\color{#35bf28}+0.95\%$
test_serialize_model	0.1315s	0.1302s	7.6781 Ops/s	7.6437 Ops/s	$\color{#35bf28}+0.45\%$
test_serialize_model_pickle	1.3596s	1.2166s	0.8219 Ops/s	0.8217 Ops/s	$\color{#35bf28}+0.03\%$
test_serialize_weights	0.2293s	0.1442s	6.9349 Ops/s	6.9388 Ops/s	$\color{#d91a1a}-0.06\%$
test_serialize_weights_returnearly	0.2300s	56.3455ms	17.7476 Ops/s	17.7640 Ops/s	$\color{#d91a1a}-0.09\%$
test_serialize_weights_pickle	1.3740s	1.2174s	0.8214 Ops/s	0.8203 Ops/s	$\color{#35bf28}+0.13\%$
test_reshape_pytree	0.1723ms	35.7485μs	27.9732 KOps/s	28.7097 KOps/s	$\color{#d91a1a}-2.57\%$
test_reshape_td	0.1610ms	41.9335μs	23.8473 KOps/s	24.0061 KOps/s	$\color{#d91a1a}-0.66\%$
test_view_pytree	0.1664ms	34.6919μs	28.8252 KOps/s	28.8179 KOps/s	$\color{#35bf28}+0.03\%$
test_view_td	0.1808ms	45.4267μs	22.0135 KOps/s	21.7021 KOps/s	$\color{#35bf28}+1.43\%$
test_unbind_pytree	0.1018ms	34.3050μs	29.1503 KOps/s	28.9452 KOps/s	$\color{#35bf28}+0.71\%$
test_unbind_td	0.5295ms	43.2058μs	23.1451 KOps/s	23.4760 KOps/s	$\color{#d91a1a}-1.41\%$
test_split_pytree	0.5282ms	45.4002μs	22.0263 KOps/s	22.0306 KOps/s	$\color{#d91a1a}-0.02\%$
test_split_td	0.1996ms	55.2410μs	18.1025 KOps/s	14.9312 KOps/s	$\textbf{\color{#35bf28}+21.24\%}$
test_add_pytree	0.2137ms	56.4759μs	17.7067 KOps/s	17.0981 KOps/s	$\color{#35bf28}+3.56\%$
test_add_td	0.2538ms	94.4780μs	10.5845 KOps/s	10.8259 KOps/s	$\color{#d91a1a}-2.23\%$
test_compile_add_one_nested[tensordict-compile]	0.3043ms	0.1594ms	6.2748 KOps/s	6.2506 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_add_one_nested[tensordict-eager]	0.3961ms	0.1638ms	6.1049 KOps/s	6.2875 KOps/s	$\color{#d91a1a}-2.90\%$
test_compile_add_one_nested[pytree-compile]	0.3106ms	0.1431ms	6.9889 KOps/s	6.9319 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_add_one_nested[pytree-eager]	0.5844ms	0.1813ms	5.5150 KOps/s	5.5300 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_copy_nested[tensordict-compile]	0.1370ms	21.4800μs	46.5550 KOps/s	46.4625 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_copy_nested[tensordict-eager]	0.4504ms	49.2004μs	20.3251 KOps/s	19.8286 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_copy_nested[pytree-compile]	0.4618ms	63.9861μs	15.6284 KOps/s	15.5711 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_copy_nested[pytree-eager]	0.4558ms	49.2593μs	20.3007 KOps/s	20.3491 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_add_one_flat[tensordict-compile]	0.4424ms	0.3209ms	3.1158 KOps/s	3.0514 KOps/s	$\color{#35bf28}+2.11\%$
test_compile_add_one_flat[tensordict-eager]	0.3656ms	0.2349ms	4.2579 KOps/s	4.2334 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_add_one_flat[tensorclass-compile]	0.2733ms	0.1267ms	7.8913 KOps/s	7.7536 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_add_one_flat[tensorclass-eager]	0.2357ms	66.4283μs	15.0538 KOps/s	15.2105 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_add_one_flat[pytree-compile]	0.4554ms	0.3240ms	3.0860 KOps/s	3.1383 KOps/s	$\color{#d91a1a}-1.67\%$
test_compile_add_one_flat[pytree-eager]	0.8190ms	0.6151ms	1.6258 KOps/s	1.6521 KOps/s	$\color{#d91a1a}-1.59\%$
test_compile_add_self_flat[tensordict-eager]	0.4422ms	0.2883ms	3.4684 KOps/s	3.5769 KOps/s	$\color{#d91a1a}-3.03\%$
test_compile_add_self_flat[tensordict-compile]	0.5273ms	0.3293ms	3.0366 KOps/s	3.1205 KOps/s	$\color{#d91a1a}-2.69\%$
test_compile_add_self_flat[tensorclass-eager]	0.2527ms	80.7766μs	12.3798 KOps/s	13.3075 KOps/s	$\textbf{\color{#d91a1a}-6.97\%}$
test_compile_add_self_flat[tensorclass-compile]	0.3008ms	0.1331ms	7.5158 KOps/s	7.7012 KOps/s	$\color{#d91a1a}-2.41\%$
test_compile_add_self_flat[pytree-eager]	0.7042ms	0.5156ms	1.9394 KOps/s	1.9768 KOps/s	$\color{#d91a1a}-1.89\%$
test_compile_add_self_flat[pytree-compile]	0.4727ms	0.3238ms	3.0887 KOps/s	3.1158 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_copy_flat[tensordict-compile]	0.1563ms	19.0449μs	52.5076 KOps/s	51.6295 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_copy_flat[tensordict-eager]	0.1411ms	39.0448μs	25.6116 KOps/s	24.5470 KOps/s	$\color{#35bf28}+4.34\%$
test_compile_copy_flat[pytree-compile]	0.2104ms	69.2620μs	14.4379 KOps/s	14.3039 KOps/s	$\color{#35bf28}+0.94\%$
test_compile_copy_flat[pytree-eager]	0.1723ms	51.4381μs	19.4408 KOps/s	19.0918 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_assign_and_add[tensordict-compile]	2.3352ms	0.8202ms	1.2193 KOps/s	1.1167 KOps/s	$\textbf{\color{#35bf28}+9.19\%}$
test_compile_assign_and_add[tensordict-eager]	3.4187ms	3.1614ms	316.3179 Ops/s	317.4796 Ops/s	$\color{#d91a1a}-0.37\%$
test_compile_assign_and_add[pytree-compile]	2.3304ms	0.8220ms	1.2166 KOps/s	1.1259 KOps/s	$\textbf{\color{#35bf28}+8.05\%}$
test_compile_assign_and_add[pytree-eager]	3.5491ms	3.1099ms	321.5580 Ops/s	320.5473 Ops/s	$\color{#35bf28}+0.32\%$
test_compile_indexing[tensor-tensordict-compile]	0.2620ms	0.1068ms	9.3595 KOps/s	9.0542 KOps/s	$\color{#35bf28}+3.37\%$
test_compile_indexing[tensor-tensordict-eager]	0.2413ms	58.9026μs	16.9772 KOps/s	15.8150 KOps/s	$\textbf{\color{#35bf28}+7.35\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.2831ms	0.1053ms	9.4934 KOps/s	9.6735 KOps/s	$\color{#d91a1a}-1.86\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2396ms	44.7485μs	22.3471 KOps/s	23.5360 KOps/s	$\textbf{\color{#d91a1a}-5.05\%}$
test_compile_indexing[tensor-pytree-compile]	0.3081ms	0.1073ms	9.3214 KOps/s	9.6390 KOps/s	$\color{#d91a1a}-3.30\%$
test_compile_indexing[tensor-pytree-eager]	0.2271ms	42.3445μs	23.6158 KOps/s	23.3859 KOps/s	$\color{#35bf28}+0.98\%$
test_compile_indexing[slice-tensordict-compile]	0.3026ms	0.1406ms	7.1123 KOps/s	7.3184 KOps/s	$\color{#d91a1a}-2.82\%$
test_compile_indexing[slice-tensordict-eager]	0.1718ms	24.6355μs	40.5918 KOps/s	39.0941 KOps/s	$\color{#35bf28}+3.83\%$
test_compile_indexing[slice-tensorclass-compile]	0.2835ms	0.1341ms	7.4555 KOps/s	7.6218 KOps/s	$\color{#d91a1a}-2.18\%$
test_compile_indexing[slice-tensorclass-eager]	0.1216ms	20.5639μs	48.6290 KOps/s	46.9424 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_indexing[slice-pytree-compile]	0.2946ms	0.1345ms	7.4375 KOps/s	7.5935 KOps/s	$\color{#d91a1a}-2.05\%$
test_compile_indexing[slice-pytree-eager]	0.1192ms	22.0205μs	45.4122 KOps/s	46.8344 KOps/s	$\color{#d91a1a}-3.04\%$
test_compile_indexing[int-tensordict-compile]	0.2904ms	0.1421ms	7.0394 KOps/s	7.1375 KOps/s	$\color{#d91a1a}-1.37\%$
test_compile_indexing[int-tensordict-eager]	0.5046ms	24.1286μs	41.4446 KOps/s	39.3191 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_compile_indexing[int-tensorclass-compile]	0.2788ms	0.1307ms	7.6522 KOps/s	7.3059 KOps/s	$\color{#35bf28}+4.74\%$
test_compile_indexing[int-tensorclass-eager]	0.1577ms	21.2602μs	47.0362 KOps/s	47.1006 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_indexing[int-pytree-compile]	0.2759ms	0.1361ms	7.3490 KOps/s	7.5783 KOps/s	$\color{#d91a1a}-3.02\%$
test_compile_indexing[int-pytree-eager]	57.6140μs	20.8082μs	48.0579 KOps/s	46.8269 KOps/s	$\color{#35bf28}+2.63\%$
test_mod_add[eager]	0.1935ms	32.6272μs	30.6493 KOps/s	31.4792 KOps/s	$\color{#d91a1a}-2.64\%$
test_mod_add[compile]	0.3206ms	69.5365μs	14.3809 KOps/s	14.2451 KOps/s	$\color{#35bf28}+0.95\%$
test_mod_add[compile-overhead]	0.2609ms	0.1345ms	7.4370 KOps/s	6.7847 KOps/s	$\textbf{\color{#35bf28}+9.61\%}$
test_mod_wrap[eager]	0.9734ms	0.7689ms	1.3005 KOps/s	1.2862 KOps/s	$\color{#35bf28}+1.11\%$
test_mod_wrap[compile]	2.2001ms	0.8294ms	1.2057 KOps/s	1.2045 KOps/s	$\color{#35bf28}+0.09\%$
test_mod_wrap[compile-overhead]	4.8641ms	3.0801ms	324.6597 Ops/s	327.3484 Ops/s	$\color{#d91a1a}-0.82\%$
test_mod_wrap_and_backward[eager]	4.1617ms	4.0242ms	248.4960 Ops/s	244.9290 Ops/s	$\color{#35bf28}+1.46\%$
test_mod_wrap_and_backward[compile]	4.4421ms	3.9888ms	250.7043 Ops/s	245.5426 Ops/s	$\color{#35bf28}+2.10\%$
test_mod_wrap_and_backward[compile-overhead]	1.3599ms	0.9014ms	1.1094 KOps/s	993.0840 Ops/s	$\textbf{\color{#35bf28}+11.72\%}$
test_seq_add[eager]	0.4813ms	99.0435μs	10.0966 KOps/s	10.1996 KOps/s	$\color{#d91a1a}-1.01\%$
test_seq_add[compile]	0.2283ms	82.3332μs	12.1458 KOps/s	12.2373 KOps/s	$\color{#d91a1a}-0.75\%$
test_seq_add[compile-overhead]	0.2844ms	0.1121ms	8.9173 KOps/s	8.7710 KOps/s	$\color{#35bf28}+1.67\%$
test_seq_wrap[eager]	1.0799ms	0.9208ms	1.0860 KOps/s	1.0753 KOps/s	$\color{#35bf28}+0.99\%$
test_seq_wrap[compile]	1.2295ms	0.8468ms	1.1809 KOps/s	1.1790 KOps/s	$\color{#35bf28}+0.16\%$
test_seq_wrap[compile-overhead]	0.3706ms	0.2183ms	4.5799 KOps/s	4.5001 KOps/s	$\color{#35bf28}+1.77\%$
test_func_call_runtime[False-eager]	2.7477ms	2.3741ms	421.2076 Ops/s	432.2579 Ops/s	$\color{#d91a1a}-2.56\%$
test_func_call_runtime[False-compile]	2.7935ms	2.3813ms	419.9449 Ops/s	422.2358 Ops/s	$\color{#d91a1a}-0.54\%$
test_func_call_runtime[False-compile-overhead]	0.5202ms	0.3591ms	2.7850 KOps/s	2.7787 KOps/s	$\color{#35bf28}+0.23\%$
test_func_call_runtime[True-eager]	2.9741ms	2.5399ms	393.7215 Ops/s	402.2854 Ops/s	$\color{#d91a1a}-2.13\%$
test_func_call_runtime[True-compile]	2.8117ms	2.4081ms	415.2723 Ops/s	417.1377 Ops/s	$\color{#d91a1a}-0.45\%$
test_func_call_runtime[True-compile-overhead]	0.7983ms	0.3835ms	2.6075 KOps/s	2.6350 KOps/s	$\color{#d91a1a}-1.04\%$
test_func_call_cm_runtime[False-eager]	2.7552ms	2.3590ms	423.9024 Ops/s	433.9498 Ops/s	$\color{#d91a1a}-2.32\%$
test_func_call_cm_runtime[False-compile]	2.8830ms	2.3950ms	417.5323 Ops/s	423.9553 Ops/s	$\color{#d91a1a}-1.52\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5115ms	0.3633ms	2.7523 KOps/s	2.7732 KOps/s	$\color{#d91a1a}-0.75\%$
test_func_call_cm_runtime[True-eager]	2.8431ms	2.6399ms	378.7979 Ops/s	383.6147 Ops/s	$\color{#d91a1a}-1.26\%$
test_func_call_cm_runtime[True-compile]	2.6854ms	2.4431ms	409.3128 Ops/s	411.8526 Ops/s	$\color{#d91a1a}-0.62\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5131ms	0.4090ms	2.4451 KOps/s	2.4597 KOps/s	$\color{#d91a1a}-0.60\%$
test_vmap_func_call_cm_runtime[eager]	4.1812ms	3.7514ms	266.5670 Ops/s	267.3659 Ops/s	$\color{#d91a1a}-0.30\%$
test_vmap_func_call_cm_runtime[compile]	2.5868ms	2.4454ms	408.9283 Ops/s	413.8238 Ops/s	$\color{#d91a1a}-1.18\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5305ms	0.4116ms	2.4296 KOps/s	2.4442 KOps/s	$\color{#d91a1a}-0.60\%$
test_distributed	1.9612ms	0.2202ms	4.5422 KOps/s	8.5399 KOps/s	$\textbf{\color{#d91a1a}-46.81\%}$
test_tdmodule	0.2365ms	14.7124μs	67.9701 KOps/s	67.3288 KOps/s	$\color{#35bf28}+0.95\%$
test_tdmodule_dispatch	50.7730μs	28.4415μs	35.1599 KOps/s	34.8380 KOps/s	$\color{#35bf28}+0.92\%$
test_tdseq	44.9930μs	15.5117μs	64.4674 KOps/s	63.8603 KOps/s	$\color{#35bf28}+0.95\%$
test_tdseq_dispatch	53.5740μs	31.0811μs	32.1739 KOps/s	31.3776 KOps/s	$\color{#35bf28}+2.54\%$
test_instantiation_functorch	1.9983ms	1.8184ms	549.9437 Ops/s	536.4621 Ops/s	$\color{#35bf28}+2.51\%$
test_instantiation_td	1.8349ms	1.1972ms	835.2845 Ops/s	821.8796 Ops/s	$\color{#35bf28}+1.63\%$
test_exec_functorch	1.1604ms	0.9823ms	1.0180 KOps/s	1.0121 KOps/s	$\color{#35bf28}+0.59\%$
test_exec_functional_call	1.1255ms	0.9968ms	1.0032 KOps/s	1.0085 KOps/s	$\color{#d91a1a}-0.53\%$
test_exec_td	1.1633ms	1.0140ms	986.2229 Ops/s	974.4189 Ops/s	$\color{#35bf28}+1.21\%$
test_exec_td_decorator	1.1834ms	1.0466ms	955.4635 Ops/s	955.7548 Ops/s	$\color{#d91a1a}-0.03\%$
test_vmap_mlp_speed[True-True]	1.9655ms	1.2495ms	800.2906 Ops/s	796.0017 Ops/s	$\color{#35bf28}+0.54\%$
test_vmap_mlp_speed[True-False]	1.4033ms	1.2444ms	803.5800 Ops/s	798.8533 Ops/s	$\color{#35bf28}+0.59\%$
test_vmap_mlp_speed[False-True]	1.3402ms	1.1433ms	874.6630 Ops/s	873.0842 Ops/s	$\color{#35bf28}+0.18\%$
test_vmap_mlp_speed[False-False]	1.2887ms	1.1417ms	875.8688 Ops/s	876.8408 Ops/s	$\color{#d91a1a}-0.11\%$
test_vmap_mlp_speed_decorator[True-True]	1.4189ms	1.2243ms	816.7794 Ops/s	820.6733 Ops/s	$\color{#d91a1a}-0.47\%$
test_vmap_mlp_speed_decorator[True-False]	1.7938ms	1.2299ms	813.0665 Ops/s	819.2043 Ops/s	$\color{#d91a1a}-0.75\%$
test_vmap_mlp_speed_decorator[False-True]	1.2832ms	1.1425ms	875.2619 Ops/s	878.0181 Ops/s	$\color{#d91a1a}-0.31\%$
test_vmap_mlp_speed_decorator[False-False]	1.3172ms	1.1417ms	875.9105 Ops/s	877.6248 Ops/s	$\color{#d91a1a}-0.20\%$
test_vmap_transformer_speed[True-True]	13.3198ms	12.9795ms	77.0448 Ops/s	76.6919 Ops/s	$\color{#35bf28}+0.46\%$
test_vmap_transformer_speed[True-False]	13.1228ms	12.9584ms	77.1698 Ops/s	76.8804 Ops/s	$\color{#35bf28}+0.38\%$
test_vmap_transformer_speed[False-True]	12.9937ms	12.7863ms	78.2089 Ops/s	77.9203 Ops/s	$\color{#35bf28}+0.37\%$
test_vmap_transformer_speed[False-False]	12.9204ms	12.7315ms	78.5451 Ops/s	77.8667 Ops/s	$\color{#35bf28}+0.87\%$
test_vmap_transformer_speed_decorator[True-True]	33.6697ms	33.4591ms	29.8872 Ops/s	29.9268 Ops/s	$\color{#d91a1a}-0.13\%$
test_vmap_transformer_speed_decorator[True-False]	33.7920ms	33.5517ms	29.8048 Ops/s	29.9202 Ops/s	$\color{#d91a1a}-0.39\%$
test_vmap_transformer_speed_decorator[False-True]	33.4846ms	33.2878ms	30.0411 Ops/s	30.0079 Ops/s	$\color{#35bf28}+0.11\%$
test_vmap_transformer_speed_decorator[False-False]	33.2702ms	33.1038ms	30.2080 Ops/s	29.9991 Ops/s	$\color{#35bf28}+0.70\%$
test_to_module_speed[True]	1.3706ms	1.0045ms	995.5324 Ops/s	998.3117 Ops/s	$\color{#d91a1a}-0.28\%$
test_to_module_speed[False]	1.3828ms	0.9762ms	1.0244 KOps/s	1.0297 KOps/s	$\color{#d91a1a}-0.51\%$
test_tc_init	57.0540μs	33.1318μs	30.1824 KOps/s	30.2185 KOps/s	$\color{#d91a1a}-0.12\%$
test_tc_init_nested	0.1083ms	71.5052μs	13.9850 KOps/s	14.5181 KOps/s	$\color{#d91a1a}-3.67\%$
test_tc_first_layer_tensor	4.2603μs	0.6771μs	1.4768 MOps/s	1.4728 MOps/s	$\color{#35bf28}+0.27\%$
test_tc_first_layer_nontensor	23.0910μs	2.2320μs	448.0270 KOps/s	441.9764 KOps/s	$\color{#35bf28}+1.37\%$
test_tc_second_layer_tensor	33.0245μs	1.3831μs	723.0055 KOps/s	731.5392 KOps/s	$\color{#d91a1a}-1.17\%$
test_tc_second_layer_nontensor	31.6020μs	2.9691μs	336.8074 KOps/s	342.1341 KOps/s	$\color{#d91a1a}-1.56\%$
test_unbind	0.2046s	12.7220ms	78.6041 Ops/s	88.2514 Ops/s	$\textbf{\color{#d91a1a}-10.93\%}$
test_full_like	0.7747ms	0.5771ms	1.7328 KOps/s	1.7384 KOps/s	$\color{#d91a1a}-0.32\%$
test_zeros_like	0.3881ms	0.1981ms	5.0471 KOps/s	5.0434 KOps/s	$\color{#35bf28}+0.07\%$
test_ones_like	0.3589ms	0.1979ms	5.0519 KOps/s	5.0503 KOps/s	$\color{#35bf28}+0.03\%$
test_clone	0.5760ms	0.4140ms	2.4152 KOps/s	2.4146 KOps/s	$\color{#35bf28}+0.03\%$
test_squeeze	0.1232ms	9.9328μs	100.6763 KOps/s	102.1743 KOps/s	$\color{#d91a1a}-1.47\%$
test_unsqueeze	0.2525ms	73.7529μs	13.5588 KOps/s	13.4526 KOps/s	$\color{#35bf28}+0.79\%$
test_split	0.4186ms	0.1600ms	6.2519 KOps/s	6.4251 KOps/s	$\color{#d91a1a}-2.70\%$
test_permute	0.2990ms	0.1759ms	5.6865 KOps/s	5.6218 KOps/s	$\color{#35bf28}+1.15\%$
test_stack	1.3734ms	0.8738ms	1.1444 KOps/s	1.1713 KOps/s	$\color{#d91a1a}-2.30\%$
test_cat	1.3833ms	1.2314ms	812.0840 Ops/s	811.7786 Ops/s	$\color{#35bf28}+0.04\%$