[Doc] Better doc for non-tensor data handling

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Doc] Better doc for non-tensor data handling #1173

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}27$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	46.7970μs	21.4237μs	46.6774 KOps/s	49.8425 KOps/s	$\textbf{\color{#d91a1a}-6.35\%}$
test_plain_set_stack_nested	50.0840μs	21.5035μs	46.5041 KOps/s	49.2354 KOps/s	$\textbf{\color{#d91a1a}-5.55\%}$
test_plain_set_nested_inplace	57.6880μs	23.0369μs	43.4087 KOps/s	45.1424 KOps/s	$\color{#d91a1a}-3.84\%$
test_plain_set_stack_nested_inplace	71.6640μs	23.0027μs	43.4732 KOps/s	45.3270 KOps/s	$\color{#d91a1a}-4.09\%$
test_items	80.7210μs	4.2487μs	235.3683 KOps/s	239.5594 KOps/s	$\color{#d91a1a}-1.75\%$
test_items_nested	0.8250ms	0.3975ms	2.5156 KOps/s	2.4856 KOps/s	$\color{#35bf28}+1.20\%$
test_items_nested_locked	0.6920ms	0.3896ms	2.5670 KOps/s	2.4788 KOps/s	$\color{#35bf28}+3.56\%$
test_items_nested_leaf	0.1582ms	76.6951μs	13.0386 KOps/s	12.4524 KOps/s	$\color{#35bf28}+4.71\%$
test_items_stack_nested	0.5096ms	0.3930ms	2.5446 KOps/s	2.4520 KOps/s	$\color{#35bf28}+3.78\%$
test_items_stack_nested_leaf	0.2530ms	80.4885μs	12.4241 KOps/s	12.3735 KOps/s	$\color{#35bf28}+0.41\%$
test_items_stack_nested_locked	0.8177ms	0.3940ms	2.5383 KOps/s	2.4741 KOps/s	$\color{#35bf28}+2.60\%$
test_keys	40.6160μs	3.5084μs	285.0270 KOps/s	282.8995 KOps/s	$\color{#35bf28}+0.75\%$
test_keys_nested	0.3255ms	0.1610ms	6.2117 KOps/s	5.9883 KOps/s	$\color{#35bf28}+3.73\%$
test_keys_nested_locked	0.6777ms	0.1676ms	5.9659 KOps/s	5.8165 KOps/s	$\color{#35bf28}+2.57\%$
test_keys_nested_leaf	0.2279ms	0.1411ms	7.0889 KOps/s	6.8940 KOps/s	$\color{#35bf28}+2.83\%$
test_keys_stack_nested	0.4137ms	0.1615ms	6.1908 KOps/s	6.1564 KOps/s	$\color{#35bf28}+0.56\%$
test_keys_stack_nested_leaf	0.2405ms	0.1380ms	7.2454 KOps/s	7.0923 KOps/s	$\color{#35bf28}+2.16\%$
test_keys_stack_nested_locked	0.3001ms	0.1668ms	5.9936 KOps/s	5.9621 KOps/s	$\color{#35bf28}+0.53\%$
test_values	9.0510μs	1.0355μs	965.6720 KOps/s	962.6341 KOps/s	$\color{#35bf28}+0.32\%$
test_values_nested	0.1199ms	60.7911μs	16.4498 KOps/s	16.1205 KOps/s	$\color{#35bf28}+2.04\%$
test_values_nested_locked	0.1205ms	60.6352μs	16.4921 KOps/s	16.1727 KOps/s	$\color{#35bf28}+1.97\%$
test_values_nested_leaf	0.1386ms	70.0570μs	14.2741 KOps/s	14.0269 KOps/s	$\color{#35bf28}+1.76\%$
test_values_stack_nested	0.1449ms	64.3419μs	15.5420 KOps/s	14.6876 KOps/s	$\textbf{\color{#35bf28}+5.82\%}$
test_values_stack_nested_leaf	0.1366ms	69.8833μs	14.3096 KOps/s	13.7918 KOps/s	$\color{#35bf28}+3.75\%$
test_values_stack_nested_locked	0.1255ms	62.0141μs	16.1254 KOps/s	15.9778 KOps/s	$\color{#35bf28}+0.92\%$
test_membership	17.0820μs	0.8807μs	1.1355 MOps/s	1.3557 MOps/s	$\textbf{\color{#d91a1a}-16.24\%}$
test_membership_nested	29.7850μs	2.8757μs	347.7389 KOps/s	344.2927 KOps/s	$\color{#35bf28}+1.00\%$
test_membership_nested_leaf	23.6940μs	2.9195μs	342.5257 KOps/s	341.4405 KOps/s	$\color{#35bf28}+0.32\%$
test_membership_stacked_nested	22.6030μs	2.8880μs	346.2663 KOps/s	344.5646 KOps/s	$\color{#35bf28}+0.49\%$
test_membership_stacked_nested_leaf	28.8130μs	2.8535μs	350.4493 KOps/s	344.0324 KOps/s	$\color{#35bf28}+1.87\%$
test_membership_nested_last	24.9070μs	4.3991μs	227.3200 KOps/s	216.9891 KOps/s	$\color{#35bf28}+4.76\%$
test_membership_nested_leaf_last	23.5540μs	4.4388μs	225.2841 KOps/s	226.0182 KOps/s	$\color{#d91a1a}-0.32\%$
test_membership_stacked_nested_last	29.1350μs	6.0747μs	164.6168 KOps/s	145.8382 KOps/s	$\textbf{\color{#35bf28}+12.88\%}$
test_membership_stacked_nested_leaf_last	33.7730μs	6.1176μs	163.4633 KOps/s	153.2492 KOps/s	$\textbf{\color{#35bf28}+6.67\%}$
test_nested_getleaf	34.4650μs	10.6369μs	94.0125 KOps/s	94.2991 KOps/s	$\color{#d91a1a}-0.30\%$
test_nested_get	33.4420μs	10.1757μs	98.2731 KOps/s	99.1399 KOps/s	$\color{#d91a1a}-0.87\%$
test_stacked_getleaf	0.1136ms	10.5883μs	94.4442 KOps/s	93.2327 KOps/s	$\color{#35bf28}+1.30\%$
test_stacked_get	47.9900μs	10.0374μs	99.6271 KOps/s	98.8724 KOps/s	$\color{#35bf28}+0.76\%$
test_nested_getitemleaf	44.0020μs	11.2835μs	88.6252 KOps/s	89.8902 KOps/s	$\color{#d91a1a}-1.41\%$
test_nested_getitem	37.5100μs	10.3035μs	97.0546 KOps/s	95.4111 KOps/s	$\color{#35bf28}+1.72\%$
test_stacked_getitemleaf	36.0070μs	10.9812μs	91.0647 KOps/s	88.5544 KOps/s	$\color{#35bf28}+2.83\%$
test_stacked_getitem	35.8870μs	10.4203μs	95.9664 KOps/s	96.5388 KOps/s	$\color{#d91a1a}-0.59\%$
test_lock_nested	0.8468ms	0.4481ms	2.2318 KOps/s	1.8082 KOps/s	$\textbf{\color{#35bf28}+23.43\%}$
test_lock_stack_nested	0.6535ms	0.4186ms	2.3890 KOps/s	2.3734 KOps/s	$\color{#35bf28}+0.66\%$
test_unlock_nested	0.6952ms	0.3711ms	2.6945 KOps/s	2.6727 KOps/s	$\color{#35bf28}+0.81\%$
test_unlock_stack_nested	0.7321ms	0.3403ms	2.9388 KOps/s	2.9440 KOps/s	$\color{#d91a1a}-0.18\%$
test_flatten_speed	0.2234ms	0.1003ms	9.9706 KOps/s	9.9529 KOps/s	$\color{#35bf28}+0.18\%$
test_unflatten_speed	0.6400ms	0.5104ms	1.9591 KOps/s	1.9544 KOps/s	$\color{#35bf28}+0.24\%$
test_common_ops	3.6181ms	0.8279ms	1.2079 KOps/s	1.2748 KOps/s	$\textbf{\color{#d91a1a}-5.25\%}$
test_creation	18.1940μs	2.4638μs	405.8853 KOps/s	402.7347 KOps/s	$\color{#35bf28}+0.78\%$
test_creation_empty	42.7100μs	13.4677μs	74.2518 KOps/s	91.5020 KOps/s	$\textbf{\color{#d91a1a}-18.85\%}$
test_creation_nested_1	53.7400μs	16.3060μs	61.3271 KOps/s	73.2016 KOps/s	$\textbf{\color{#d91a1a}-16.22\%}$
test_creation_nested_2	58.1890μs	21.0589μs	47.4860 KOps/s	53.9458 KOps/s	$\textbf{\color{#d91a1a}-11.97\%}$
test_clone	59.4910μs	13.5750μs	73.6646 KOps/s	73.6760 KOps/s	$\color{#d91a1a}-0.02\%$
test_getitem[int]	1.2040ms	13.4428μs	74.3893 KOps/s	78.4897 KOps/s	$\textbf{\color{#d91a1a}-5.22\%}$
test_getitem[slice_int]	0.1370ms	25.2764μs	39.5626 KOps/s	40.8642 KOps/s	$\color{#d91a1a}-3.19\%$
test_getitem[range]	0.1805ms	49.1289μs	20.3546 KOps/s	20.3226 KOps/s	$\color{#35bf28}+0.16\%$
test_getitem[tuple]	0.1473ms	20.7876μs	48.1057 KOps/s	48.9545 KOps/s	$\color{#d91a1a}-1.73\%$
test_getitem[list]	0.1635ms	43.6875μs	22.8898 KOps/s	23.0951 KOps/s	$\color{#d91a1a}-0.89\%$
test_setitem_dim[int]	58.0480μs	24.6210μs	40.6157 KOps/s	39.8376 KOps/s	$\color{#35bf28}+1.95\%$
test_setitem_dim[slice_int]	91.2110μs	51.3681μs	19.4673 KOps/s	19.4704 KOps/s	$\color{#d91a1a}-0.02\%$
test_setitem_dim[range]	0.1352ms	73.0254μs	13.6939 KOps/s	13.7077 KOps/s	$\color{#d91a1a}-0.10\%$
test_setitem_dim[tuple]	81.0620μs	40.7455μs	24.5426 KOps/s	24.6576 KOps/s	$\color{#d91a1a}-0.47\%$
test_setitem	80.0500μs	21.6273μs	46.2378 KOps/s	49.8741 KOps/s	$\textbf{\color{#d91a1a}-7.29\%}$
test_set	84.3880μs	21.3260μs	46.8910 KOps/s	52.0785 KOps/s	$\textbf{\color{#d91a1a}-9.96\%}$
test_set_shared	1.2844ms	0.1720ms	5.8154 KOps/s	5.8107 KOps/s	$\color{#35bf28}+0.08\%$
test_update	0.1231ms	24.9543μs	40.0732 KOps/s	46.0702 KOps/s	$\textbf{\color{#d91a1a}-13.02\%}$
test_update_nested	0.1034ms	34.8443μs	28.6991 KOps/s	31.7068 KOps/s	$\textbf{\color{#d91a1a}-9.49\%}$
test_update__nested	1.0453ms	34.2323μs	29.2121 KOps/s	29.4505 KOps/s	$\color{#d91a1a}-0.81\%$
test_set_nested	76.7140μs	23.1285μs	43.2367 KOps/s	46.4178 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_set_nested_new	0.1232ms	27.6258μs	36.1981 KOps/s	37.8387 KOps/s	$\color{#d91a1a}-4.34\%$
test_select	81.5630μs	42.8229μs	23.3520 KOps/s	23.5858 KOps/s	$\color{#d91a1a}-0.99\%$
test_select_nested	0.1173ms	62.1359μs	16.0938 KOps/s	15.9556 KOps/s	$\color{#35bf28}+0.87\%$
test_exclude_nested	0.1396ms	79.9109μs	12.5139 KOps/s	12.3243 KOps/s	$\color{#35bf28}+1.54\%$
test_empty[True]	0.4933ms	0.4006ms	2.4965 KOps/s	2.4523 KOps/s	$\color{#35bf28}+1.80\%$
test_empty[False]	7.7870μs	1.3835μs	722.8066 KOps/s	720.2525 KOps/s	$\color{#35bf28}+0.35\%$
test_unbind_speed	0.3402ms	0.2704ms	3.6982 KOps/s	3.6935 KOps/s	$\color{#35bf28}+0.13\%$
test_unbind_speed_stack0	0.3642ms	0.2639ms	3.7889 KOps/s	3.8289 KOps/s	$\color{#d91a1a}-1.05\%$
test_unbind_speed_stack1	99.7246ms	0.7858ms	1.2726 KOps/s	1.4034 KOps/s	$\textbf{\color{#d91a1a}-9.32\%}$
test_split	95.4353ms	1.7934ms	557.5850 Ops/s	547.3445 Ops/s	$\color{#35bf28}+1.87\%$
test_chunk	1.8185ms	1.6330ms	612.3527 Ops/s	560.8081 Ops/s	$\textbf{\color{#35bf28}+9.19\%}$
test_consolidate_njt[False-None]	0.1067s	9.0117ms	110.9665 Ops/s	121.2736 Ops/s	$\textbf{\color{#d91a1a}-8.50\%}$
test_creation[device0]	0.2254ms	92.0190μs	10.8673 KOps/s	11.0358 KOps/s	$\color{#d91a1a}-1.53\%$
test_creation_from_tensor	3.4498ms	96.4119μs	10.3722 KOps/s	10.6639 KOps/s	$\color{#d91a1a}-2.74\%$
test_add_one[memmap_tensor0]	0.1772ms	5.1578μs	193.8798 KOps/s	207.9851 KOps/s	$\textbf{\color{#d91a1a}-6.78\%}$
test_contiguous[memmap_tensor0]	19.2960μs	0.5092μs	1.9640 MOps/s	1.9668 MOps/s	$\color{#d91a1a}-0.14\%$
test_stack[memmap_tensor0]	69.9610μs	3.6372μs	274.9359 KOps/s	292.6743 KOps/s	$\textbf{\color{#d91a1a}-6.06\%}$
test_memmaptd_index	1.0254ms	0.2420ms	4.1321 KOps/s	4.1634 KOps/s	$\color{#d91a1a}-0.75\%$
test_memmaptd_index_astensor	0.5866ms	0.3274ms	3.0541 KOps/s	3.0957 KOps/s	$\color{#d91a1a}-1.34\%$
test_memmaptd_index_op	0.9813ms	0.6221ms	1.6074 KOps/s	1.6977 KOps/s	$\textbf{\color{#d91a1a}-5.32\%}$
test_serialize_model	0.1216s	0.1149s	8.7003 Ops/s	8.0312 Ops/s	$\textbf{\color{#35bf28}+8.33\%}$
test_serialize_model_pickle	0.4764s	0.3944s	2.5357 Ops/s	2.4599 Ops/s	$\color{#35bf28}+3.08\%$
test_serialize_weights	0.1191s	0.1132s	8.8339 Ops/s	8.5503 Ops/s	$\color{#35bf28}+3.32\%$
test_serialize_weights_returnearly	0.1766s	0.1602s	6.2410 Ops/s	6.0953 Ops/s	$\color{#35bf28}+2.39\%$
test_serialize_weights_pickle	1.1519s	0.7148s	1.3991 Ops/s	2.4480 Ops/s	$\textbf{\color{#d91a1a}-42.85\%}$
test_serialize_weights_filesystem	0.1451s	0.1426s	7.0107 Ops/s	6.7919 Ops/s	$\color{#35bf28}+3.22\%$
test_serialize_model_filesystem	0.1516s	0.1424s	7.0215 Ops/s	6.0699 Ops/s	$\textbf{\color{#35bf28}+15.68\%}$
test_reshape_pytree	68.1480μs	26.9317μs	37.1310 KOps/s	37.6204 KOps/s	$\color{#d91a1a}-1.30\%$
test_reshape_td	71.9640μs	32.8954μs	30.3994 KOps/s	30.3701 KOps/s	$\color{#35bf28}+0.10\%$
test_view_pytree	87.7940μs	26.8334μs	37.2669 KOps/s	38.0114 KOps/s	$\color{#d91a1a}-1.96\%$
test_view_td	0.1053ms	37.6805μs	26.5389 KOps/s	26.2836 KOps/s	$\color{#35bf28}+0.97\%$
test_unbind_pytree	61.0350μs	29.7092μs	33.6596 KOps/s	33.8998 KOps/s	$\color{#d91a1a}-0.71\%$
test_unbind_td	0.3178ms	39.1655μs	25.5327 KOps/s	25.5899 KOps/s	$\color{#d91a1a}-0.22\%$
test_split_pytree	64.1600μs	30.2112μs	33.1003 KOps/s	34.4686 KOps/s	$\color{#d91a1a}-3.97\%$
test_split_td	0.5020ms	46.0917μs	21.6959 KOps/s	21.7806 KOps/s	$\color{#d91a1a}-0.39\%$
test_add_pytree	89.5070μs	36.5436μs	27.3646 KOps/s	28.5919 KOps/s	$\color{#d91a1a}-4.29\%$
test_add_td	0.1218ms	59.8938μs	16.6962 KOps/s	18.9819 KOps/s	$\textbf{\color{#d91a1a}-12.04\%}$
test_compile_add_one_nested[tensordict-compile]	0.1355ms	64.0597μs	15.6104 KOps/s	15.9566 KOps/s	$\color{#d91a1a}-2.17\%$
test_compile_add_one_nested[tensordict-eager]	1.3345ms	0.1733ms	5.7689 KOps/s	5.8085 KOps/s	$\color{#d91a1a}-0.68\%$
test_compile_add_one_nested[pytree-compile]	0.1406ms	46.2538μs	21.6199 KOps/s	21.8538 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_add_one_nested[pytree-eager]	0.2222ms	0.1201ms	8.3245 KOps/s	8.5593 KOps/s	$\color{#d91a1a}-2.74\%$
test_compile_copy_nested[tensordict-compile]	70.4410μs	25.9432μs	38.5457 KOps/s	37.6150 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_copy_nested[tensordict-eager]	0.1146ms	58.0611μs	17.2232 KOps/s	17.1776 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_copy_nested[pytree-compile]	0.1788ms	77.7829μs	12.8563 KOps/s	12.7773 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_copy_nested[pytree-eager]	0.1524ms	66.8428μs	14.9605 KOps/s	14.7859 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_add_one_flat[tensordict-compile]	0.2459ms	0.1046ms	9.5569 KOps/s	9.5132 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_add_one_flat[tensordict-eager]	0.4994ms	0.2141ms	4.6698 KOps/s	4.7035 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_add_one_flat[tensorclass-compile]	96.1400μs	46.1010μs	21.6915 KOps/s	21.8519 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_add_one_flat[tensorclass-eager]	0.5231ms	66.7087μs	14.9905 KOps/s	15.0260 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_add_one_flat[pytree-compile]	0.1741ms	0.1035ms	9.6603 KOps/s	9.7364 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_add_one_flat[pytree-eager]	0.4191ms	0.2069ms	4.8332 KOps/s	5.0180 KOps/s	$\color{#d91a1a}-3.68\%$
test_compile_add_self_flat[tensordict-eager]	0.5162ms	0.2306ms	4.3363 KOps/s	4.2778 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_add_self_flat[tensordict-compile]	0.3938ms	0.1070ms	9.3416 KOps/s	8.9987 KOps/s	$\color{#35bf28}+3.81\%$
test_compile_add_self_flat[tensorclass-eager]	0.1760ms	66.4857μs	15.0408 KOps/s	15.4717 KOps/s	$\color{#d91a1a}-2.78\%$
test_compile_add_self_flat[tensorclass-compile]	0.1035ms	48.3975μs	20.6622 KOps/s	21.0487 KOps/s	$\color{#d91a1a}-1.84\%$
test_compile_add_self_flat[pytree-eager]	0.3471ms	0.1593ms	6.2781 KOps/s	6.3508 KOps/s	$\color{#d91a1a}-1.14\%$
test_compile_add_self_flat[pytree-compile]	0.2517ms	0.1024ms	9.7665 KOps/s	9.7409 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_copy_flat[tensordict-compile]	57.4370μs	22.6702μs	44.1107 KOps/s	48.1853 KOps/s	$\textbf{\color{#d91a1a}-8.46\%}$
test_compile_copy_flat[tensordict-eager]	0.1465ms	67.6452μs	14.7830 KOps/s	14.9404 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_copy_flat[pytree-compile]	0.1694ms	79.5701μs	12.5675 KOps/s	12.5840 KOps/s	$\color{#d91a1a}-0.13\%$
test_compile_copy_flat[pytree-eager]	0.1459ms	68.5933μs	14.5787 KOps/s	14.7747 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_assign_and_add[tensordict-compile]	0.3065ms	0.2060ms	4.8532 KOps/s	4.6672 KOps/s	$\color{#35bf28}+3.99\%$
test_compile_assign_and_add[tensordict-eager]	2.1414ms	1.2890ms	775.7998 Ops/s	727.5346 Ops/s	$\textbf{\color{#35bf28}+6.63\%}$
test_compile_assign_and_add[pytree-compile]	0.2803ms	0.2063ms	4.8474 KOps/s	4.8099 KOps/s	$\color{#35bf28}+0.78\%$
test_compile_assign_and_add[pytree-eager]	1.3908ms	0.7952ms	1.2576 KOps/s	1.2697 KOps/s	$\color{#d91a1a}-0.95\%$
test_compile_assign_and_add_stack[compile]	0.6401ms	0.4544ms	2.2009 KOps/s	2.1950 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_assign_and_add_stack[eager]	3.0060ms	2.8007ms	357.0544 Ops/s	378.0656 Ops/s	$\textbf{\color{#d91a1a}-5.56\%}$
test_compile_indexing[tensor-tensordict-compile]	98.3440μs	36.7544μs	27.2076 KOps/s	28.3329 KOps/s	$\color{#d91a1a}-3.97\%$
test_compile_indexing[tensor-tensordict-eager]	0.4794ms	33.7221μs	29.6541 KOps/s	31.1803 KOps/s	$\color{#d91a1a}-4.89\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1649ms	29.2057μs	34.2399 KOps/s	32.8647 KOps/s	$\color{#35bf28}+4.18\%$
test_compile_indexing[tensor-tensorclass-eager]	85.1290μs	23.2771μs	42.9607 KOps/s	43.9820 KOps/s	$\color{#d91a1a}-2.32\%$
test_compile_indexing[tensor-pytree-compile]	93.1210μs	29.7827μs	33.5765 KOps/s	32.7216 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_indexing[tensor-pytree-eager]	0.1070ms	23.3107μs	42.8988 KOps/s	43.8849 KOps/s	$\color{#d91a1a}-2.25\%$
test_compile_indexing[slice-tensordict-compile]	0.1268ms	51.3114μs	19.4889 KOps/s	19.4860 KOps/s	$\color{#35bf28}+0.01\%$
test_compile_indexing[slice-tensordict-eager]	0.5828ms	21.8578μs	45.7503 KOps/s	47.6973 KOps/s	$\color{#d91a1a}-4.08\%$
test_compile_indexing[slice-tensorclass-compile]	0.1172ms	44.1419μs	22.6542 KOps/s	22.3420 KOps/s	$\color{#35bf28}+1.40\%$
test_compile_indexing[slice-tensorclass-eager]	78.6480μs	19.4441μs	51.4295 KOps/s	52.0466 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_indexing[slice-pytree-compile]	0.1221ms	45.5025μs	21.9768 KOps/s	22.0289 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[slice-pytree-eager]	54.5120μs	19.2183μs	52.0336 KOps/s	52.2151 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_indexing[int-tensordict-compile]	0.1372ms	52.2988μs	19.1209 KOps/s	18.9874 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_indexing[int-tensordict-eager]	0.9308ms	21.0907μs	47.4142 KOps/s	48.0510 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_indexing[int-tensorclass-compile]	0.1058ms	45.1211μs	22.1626 KOps/s	22.1339 KOps/s	$\color{#35bf28}+0.13\%$
test_compile_indexing[int-tensorclass-eager]	0.1001ms	19.3270μs	51.7412 KOps/s	52.5535 KOps/s	$\color{#d91a1a}-1.55\%$
test_compile_indexing[int-pytree-compile]	0.1379ms	45.0208μs	22.2120 KOps/s	21.9817 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_indexing[int-pytree-eager]	56.5160μs	19.3935μs	51.5636 KOps/s	52.5871 KOps/s	$\color{#d91a1a}-1.95\%$
test_mod_add[eager]	0.1031ms	35.4824μs	28.1830 KOps/s	29.6839 KOps/s	$\textbf{\color{#d91a1a}-5.06\%}$
test_mod_add[compile]	0.1132ms	48.3007μs	20.7037 KOps/s	20.7116 KOps/s	$\color{#d91a1a}-0.04\%$
test_mod_add[compile-overhead]	0.1367ms	48.1614μs	20.7635 KOps/s	20.4146 KOps/s	$\color{#35bf28}+1.71\%$
test_mod_wrap[eager]	0.4611ms	0.2266ms	4.4122 KOps/s	4.4540 KOps/s	$\color{#d91a1a}-0.94\%$
test_mod_wrap[compile]	0.3202ms	0.2052ms	4.8723 KOps/s	4.7814 KOps/s	$\color{#35bf28}+1.90\%$
test_mod_wrap[compile-overhead]	0.3090ms	0.2040ms	4.9031 KOps/s	4.8326 KOps/s	$\color{#35bf28}+1.46\%$
test_mod_wrap_and_backward[eager]	15.6451ms	11.8218ms	84.5892 Ops/s	85.7155 Ops/s	$\color{#d91a1a}-1.31\%$
test_mod_wrap_and_backward[compile]	16.8129ms	12.1804ms	82.0993 Ops/s	86.9697 Ops/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_mod_wrap_and_backward[compile-overhead]	17.3271ms	12.5942ms	79.4015 Ops/s	75.9167 Ops/s	$\color{#35bf28}+4.59\%$
test_seq_add[eager]	0.2728ms	0.1188ms	8.4160 KOps/s	8.5352 KOps/s	$\color{#d91a1a}-1.40\%$
test_seq_add[compile]	0.1614ms	64.2561μs	15.5627 KOps/s	16.1751 KOps/s	$\color{#d91a1a}-3.79\%$
test_seq_add[compile-overhead]	0.1290ms	61.9700μs	16.1368 KOps/s	15.9490 KOps/s	$\color{#35bf28}+1.18\%$
test_seq_wrap[eager]	0.7140ms	0.4547ms	2.1992 KOps/s	2.2364 KOps/s	$\color{#d91a1a}-1.66\%$
test_seq_wrap[compile]	0.3235ms	0.2305ms	4.3377 KOps/s	4.4092 KOps/s	$\color{#d91a1a}-1.62\%$
test_seq_wrap[compile-overhead]	0.3430ms	0.2314ms	4.3213 KOps/s	4.4056 KOps/s	$\color{#d91a1a}-1.91\%$
test_func_call_runtime[False-eager]	1.0793ms	0.5582ms	1.7913 KOps/s	1.8440 KOps/s	$\color{#d91a1a}-2.86\%$
test_func_call_runtime[False-compile]	0.7910ms	0.4235ms	2.3614 KOps/s	2.3417 KOps/s	$\color{#35bf28}+0.84\%$
test_func_call_runtime[False-compile-overhead]	0.5497ms	0.4226ms	2.3664 KOps/s	2.3497 KOps/s	$\color{#35bf28}+0.71\%$
test_func_call_runtime[True-eager]	1.5283ms	0.7735ms	1.2928 KOps/s	1.3102 KOps/s	$\color{#d91a1a}-1.33\%$
test_func_call_runtime[True-compile]	0.7119ms	0.4648ms	2.1516 KOps/s	2.1677 KOps/s	$\color{#d91a1a}-0.74\%$
test_func_call_runtime[True-compile-overhead]	0.5604ms	0.4629ms	2.1604 KOps/s	2.1460 KOps/s	$\color{#35bf28}+0.67\%$
test_func_call_cm_runtime[False-eager]	0.8044ms	0.5555ms	1.8000 KOps/s	1.8542 KOps/s	$\color{#d91a1a}-2.92\%$
test_func_call_cm_runtime[False-compile]	0.5407ms	0.4193ms	2.3851 KOps/s	2.3681 KOps/s	$\color{#35bf28}+0.72\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5838ms	0.4191ms	2.3860 KOps/s	2.3451 KOps/s	$\color{#35bf28}+1.75\%$
test_func_call_cm_runtime[True-eager]	1.0624ms	0.9096ms	1.0994 KOps/s	1.0939 KOps/s	$\color{#35bf28}+0.50\%$
test_func_call_cm_runtime[True-compile]	0.9093ms	0.4889ms	2.0453 KOps/s	2.0527 KOps/s	$\color{#d91a1a}-0.36\%$
test_func_call_cm_runtime[True-compile-overhead]	1.2201ms	0.4895ms	2.0428 KOps/s	2.0480 KOps/s	$\color{#d91a1a}-0.25\%$
test_vmap_func_call_cm_runtime[eager]	2.7167ms	1.9222ms	520.2244 Ops/s	521.7142 Ops/s	$\color{#d91a1a}-0.29\%$
test_vmap_func_call_cm_runtime[compile]	0.8928ms	0.5191ms	1.9266 KOps/s	1.9269 KOps/s	$\color{#d91a1a}-0.02\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8213ms	0.5205ms	1.9214 KOps/s	1.9333 KOps/s	$\color{#d91a1a}-0.61\%$
test_distributed	0.3094ms	0.1286ms	7.7757 KOps/s	7.8748 KOps/s	$\color{#d91a1a}-1.26\%$
test_tdmodule	89.5270μs	27.6462μs	36.1713 KOps/s	38.6341 KOps/s	$\textbf{\color{#d91a1a}-6.37\%}$
test_tdmodule_dispatch	81.0020μs	50.3189μs	19.8732 KOps/s	21.4433 KOps/s	$\textbf{\color{#d91a1a}-7.32\%}$
test_tdseq	58.4890μs	29.6920μs	33.6791 KOps/s	34.6120 KOps/s	$\color{#d91a1a}-2.70\%$
test_tdseq_dispatch	95.2780μs	55.0043μs	18.1804 KOps/s	18.6784 KOps/s	$\color{#d91a1a}-2.67\%$
test_instantiation_functorch	2.1648ms	1.5337ms	652.0203 Ops/s	656.9923 Ops/s	$\color{#d91a1a}-0.76\%$
test_exec_functorch	0.2768ms	0.1810ms	5.5259 KOps/s	5.5657 KOps/s	$\color{#d91a1a}-0.72\%$
test_exec_functional_call	0.4021ms	0.1742ms	5.7397 KOps/s	5.7367 KOps/s	$\color{#35bf28}+0.05\%$
test_exec_td_decorator	0.4342ms	0.2314ms	4.3213 KOps/s	4.2690 KOps/s	$\color{#35bf28}+1.23\%$
test_vmap_mlp_speed_decorator[True-True]	1.0490ms	0.6644ms	1.5051 KOps/s	1.5407 KOps/s	$\color{#d91a1a}-2.32\%$
test_vmap_mlp_speed_decorator[True-False]	1.1245ms	0.6656ms	1.5024 KOps/s	1.5433 KOps/s	$\color{#d91a1a}-2.65\%$
test_vmap_mlp_speed_decorator[False-True]	0.7948ms	0.5332ms	1.8755 KOps/s	1.9113 KOps/s	$\color{#d91a1a}-1.87\%$
test_vmap_mlp_speed_decorator[False-False]	0.7637ms	0.5331ms	1.8759 KOps/s	1.8987 KOps/s	$\color{#d91a1a}-1.20\%$
test_to_module_speed[True]	1.7477ms	1.3320ms	750.7602 Ops/s	728.2591 Ops/s	$\color{#35bf28}+3.09\%$
test_to_module_speed[False]	1.6574ms	1.2786ms	782.0847 Ops/s	759.2485 Ops/s	$\color{#35bf28}+3.01\%$
test_tc_init	98.0640μs	49.2029μs	20.3240 KOps/s	21.3888 KOps/s	$\color{#d91a1a}-4.98\%$
test_tc_init_nested	0.1509ms	95.0080μs	10.5254 KOps/s	10.7940 KOps/s	$\color{#d91a1a}-2.49\%$
test_tc_first_layer_tensor	23.3430μs	1.5248μs	655.8323 KOps/s	653.5225 KOps/s	$\color{#35bf28}+0.35\%$
test_tc_first_layer_nontensor	44.9040μs	4.6681μs	214.2219 KOps/s	211.5344 KOps/s	$\color{#35bf28}+1.27\%$
test_tc_second_layer_tensor	48.4140μs	2.8179μs	354.8747 KOps/s	349.5185 KOps/s	$\color{#35bf28}+1.53\%$
test_tc_second_layer_nontensor	47.8660μs	5.9795μs	167.2391 KOps/s	164.4000 KOps/s	$\color{#35bf28}+1.73\%$
test_unbind	0.2145s	14.4785ms	69.0678 Ops/s	77.4799 Ops/s	$\textbf{\color{#d91a1a}-10.86\%}$
test_full_like	8.1791ms	7.5282ms	132.8338 Ops/s	130.9416 Ops/s	$\color{#35bf28}+1.45\%$
test_zeros_like	3.3821ms	2.9505ms	338.9212 Ops/s	331.5403 Ops/s	$\color{#35bf28}+2.23\%$
test_ones_like	3.7106ms	3.2747ms	305.3746 Ops/s	300.1923 Ops/s	$\color{#35bf28}+1.73\%$
test_clone	6.0240ms	5.3467ms	187.0319 Ops/s	186.2018 Ops/s	$\color{#35bf28}+0.45\%$
test_squeeze	69.0490μs	11.9810μs	83.4656 KOps/s	80.3728 KOps/s	$\color{#35bf28}+3.85\%$
test_unsqueeze	0.2917ms	91.8441μs	10.8880 KOps/s	10.9705 KOps/s	$\color{#d91a1a}-0.75\%$
test_split	0.3456ms	0.1950ms	5.1270 KOps/s	4.9373 KOps/s	$\color{#35bf28}+3.84\%$
test_permute	0.4153ms	0.2094ms	4.7762 KOps/s	4.6828 KOps/s	$\color{#35bf28}+1.99\%$
test_stack	33.7307ms	26.7324ms	37.4077 Ops/s	37.7844 Ops/s	$\color{#d91a1a}-1.00\%$
test_cat	30.6869ms	26.4532ms	37.8026 Ops/s	37.2095 Ops/s	$\color{#35bf28}+1.59\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}31$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	31.8810μs	11.5438μs	86.6265 KOps/s	76.2026 KOps/s	$\textbf{\color{#35bf28}+13.68\%}$
test_plain_set_stack_nested	41.0610μs	11.7253μs	85.2860 KOps/s	74.7472 KOps/s	$\textbf{\color{#35bf28}+14.10\%}$
test_plain_set_nested_inplace	46.0210μs	12.5794μs	79.4949 KOps/s	69.5700 KOps/s	$\textbf{\color{#35bf28}+14.27\%}$
test_plain_set_stack_nested_inplace	53.0510μs	12.6526μs	79.0349 KOps/s	69.5730 KOps/s	$\textbf{\color{#35bf28}+13.60\%}$
test_items	24.4300μs	2.9516μs	338.8033 KOps/s	335.8910 KOps/s	$\color{#35bf28}+0.87\%$
test_items_nested	0.4547ms	0.3595ms	2.7815 KOps/s	2.7496 KOps/s	$\color{#35bf28}+1.16\%$
test_items_nested_locked	0.4278ms	0.3613ms	2.7681 KOps/s	2.7494 KOps/s	$\color{#35bf28}+0.68\%$
test_items_nested_leaf	92.0810μs	58.9975μs	16.9499 KOps/s	17.1623 KOps/s	$\color{#d91a1a}-1.24\%$
test_items_stack_nested	0.4160ms	0.3616ms	2.7656 KOps/s	2.7618 KOps/s	$\color{#35bf28}+0.14\%$
test_items_stack_nested_leaf	99.2520μs	60.2601μs	16.5947 KOps/s	16.5898 KOps/s	$\color{#35bf28}+0.03\%$
test_items_stack_nested_locked	0.4242ms	0.3616ms	2.7658 KOps/s	2.7472 KOps/s	$\color{#35bf28}+0.67\%$
test_keys	27.4810μs	3.4806μs	287.3095 KOps/s	287.4800 KOps/s	$\color{#d91a1a}-0.06\%$
test_keys_nested	0.1226ms	81.8181μs	12.2222 KOps/s	12.2390 KOps/s	$\color{#d91a1a}-0.14\%$
test_keys_nested_locked	0.7472ms	88.2113μs	11.3364 KOps/s	11.4035 KOps/s	$\color{#d91a1a}-0.59\%$
test_keys_nested_leaf	0.1262ms	72.7692μs	13.7421 KOps/s	13.7673 KOps/s	$\color{#d91a1a}-0.18\%$
test_keys_stack_nested	0.1193ms	82.8439μs	12.0709 KOps/s	11.9646 KOps/s	$\color{#35bf28}+0.89\%$
test_keys_stack_nested_leaf	0.1121ms	73.7755μs	13.5546 KOps/s	13.6486 KOps/s	$\color{#d91a1a}-0.69\%$
test_keys_stack_nested_locked	0.1320ms	88.8227μs	11.2584 KOps/s	11.3252 KOps/s	$\color{#d91a1a}-0.59\%$
test_values	5.6767μs	0.8563μs	1.1677 MOps/s	1.1789 MOps/s	$\color{#d91a1a}-0.94\%$
test_values_nested	66.9010μs	34.8499μs	28.6945 KOps/s	29.0663 KOps/s	$\color{#d91a1a}-1.28\%$
test_values_nested_locked	64.7710μs	36.6033μs	27.3200 KOps/s	27.6966 KOps/s	$\color{#d91a1a}-1.36\%$
test_values_nested_leaf	84.8110μs	39.0567μs	25.6038 KOps/s	25.7300 KOps/s	$\color{#d91a1a}-0.49\%$
test_values_stack_nested	0.1383ms	34.7722μs	28.7586 KOps/s	28.7848 KOps/s	$\color{#d91a1a}-0.09\%$
test_values_stack_nested_leaf	0.1202ms	39.3841μs	25.3910 KOps/s	25.5317 KOps/s	$\color{#d91a1a}-0.55\%$
test_values_stack_nested_locked	97.8720μs	36.3699μs	27.4952 KOps/s	27.4087 KOps/s	$\color{#35bf28}+0.32\%$
test_membership	1.8190μs	0.5023μs	1.9907 MOps/s	1.9916 MOps/s	$\color{#d91a1a}-0.05\%$
test_membership_nested	17.9900μs	2.0081μs	497.9708 KOps/s	507.2659 KOps/s	$\color{#d91a1a}-1.83\%$
test_membership_nested_leaf	16.6300μs	2.0188μs	495.3499 KOps/s	499.8157 KOps/s	$\color{#d91a1a}-0.89\%$
test_membership_stacked_nested	30.8900μs	2.1245μs	470.6904 KOps/s	471.2052 KOps/s	$\color{#d91a1a}-0.11\%$
test_membership_stacked_nested_leaf	32.9900μs	2.1049μs	475.0832 KOps/s	472.6863 KOps/s	$\color{#35bf28}+0.51\%$
test_membership_nested_last	34.6710μs	3.0712μs	325.6096 KOps/s	318.0978 KOps/s	$\color{#35bf28}+2.36\%$
test_membership_nested_leaf_last	25.8810μs	3.0586μs	326.9488 KOps/s	315.4708 KOps/s	$\color{#35bf28}+3.64\%$
test_membership_stacked_nested_last	30.1710μs	4.9240μs	203.0876 KOps/s	219.8936 KOps/s	$\textbf{\color{#d91a1a}-7.64\%}$
test_membership_stacked_nested_leaf_last	34.7110μs	4.9109μs	203.6269 KOps/s	221.3865 KOps/s	$\textbf{\color{#d91a1a}-8.02\%}$
test_nested_getleaf	32.0410μs	6.1235μs	163.3044 KOps/s	161.6539 KOps/s	$\color{#35bf28}+1.02\%$
test_nested_get	32.6600μs	5.7779μs	173.0728 KOps/s	170.3179 KOps/s	$\color{#35bf28}+1.62\%$
test_stacked_getleaf	41.2910μs	6.1491μs	162.6252 KOps/s	161.9305 KOps/s	$\color{#35bf28}+0.43\%$
test_stacked_get	28.5100μs	5.7863μs	172.8211 KOps/s	170.5761 KOps/s	$\color{#35bf28}+1.32\%$
test_nested_getitemleaf	26.0100μs	6.2511μs	159.9729 KOps/s	161.3618 KOps/s	$\color{#d91a1a}-0.86\%$
test_nested_getitem	33.8800μs	5.9861μs	167.0530 KOps/s	170.0402 KOps/s	$\color{#d91a1a}-1.76\%$
test_stacked_getitemleaf	40.3110μs	6.2489μs	160.0285 KOps/s	159.4251 KOps/s	$\color{#35bf28}+0.38\%$
test_stacked_getitem	28.7400μs	5.9435μs	168.2519 KOps/s	169.0908 KOps/s	$\color{#d91a1a}-0.50\%$
test_lock_nested	0.7325ms	0.3881ms	2.5768 KOps/s	2.6365 KOps/s	$\color{#d91a1a}-2.26\%$
test_lock_stack_nested	0.4204ms	0.3486ms	2.8686 KOps/s	2.8550 KOps/s	$\color{#35bf28}+0.48\%$
test_unlock_nested	0.7476ms	0.3253ms	3.0741 KOps/s	3.1218 KOps/s	$\color{#d91a1a}-1.53\%$
test_unlock_stack_nested	0.3294ms	0.2878ms	3.4750 KOps/s	3.4666 KOps/s	$\color{#35bf28}+0.24\%$
test_flatten_speed	0.1292ms	77.5698μs	12.8916 KOps/s	13.1480 KOps/s	$\color{#d91a1a}-1.95\%$
test_unflatten_speed	0.3968ms	0.3254ms	3.0735 KOps/s	3.0765 KOps/s	$\color{#d91a1a}-0.10\%$
test_common_ops	94.0552ms	0.6770ms	1.4771 KOps/s	1.5585 KOps/s	$\textbf{\color{#d91a1a}-5.23\%}$
test_creation	38.2910μs	1.7268μs	579.1027 KOps/s	567.1246 KOps/s	$\color{#35bf28}+2.11\%$
test_creation_empty	55.2810μs	6.8718μs	145.5217 KOps/s	99.9904 KOps/s	$\textbf{\color{#35bf28}+45.54\%}$
test_creation_nested_1	45.1810μs	8.5653μs	116.7508 KOps/s	85.6500 KOps/s	$\textbf{\color{#35bf28}+36.31\%}$
test_creation_nested_2	43.5410μs	11.5010μs	86.9491 KOps/s	68.9366 KOps/s	$\textbf{\color{#35bf28}+26.13\%}$
test_clone	75.6920μs	11.3238μs	88.3099 KOps/s	92.8466 KOps/s	$\color{#d91a1a}-4.89\%$
test_getitem[int]	1.8426ms	11.0803μs	90.2500 KOps/s	91.2621 KOps/s	$\color{#d91a1a}-1.11\%$
test_getitem[slice_int]	0.1070ms	21.8329μs	45.8023 KOps/s	47.4929 KOps/s	$\color{#d91a1a}-3.56\%$
test_getitem[range]	0.1313ms	39.2036μs	25.5079 KOps/s	26.1735 KOps/s	$\color{#d91a1a}-2.54\%$
test_getitem[tuple]	0.1401ms	18.8814μs	52.9622 KOps/s	53.0905 KOps/s	$\color{#d91a1a}-0.24\%$
test_getitem[list]	0.1704ms	35.0485μs	28.5319 KOps/s	29.7078 KOps/s	$\color{#d91a1a}-3.96\%$
test_setitem_dim[int]	41.7210μs	20.0763μs	49.8101 KOps/s	51.7311 KOps/s	$\color{#d91a1a}-3.71\%$
test_setitem_dim[slice_int]	63.5810μs	39.5894μs	25.2593 KOps/s	25.6835 KOps/s	$\color{#d91a1a}-1.65\%$
test_setitem_dim[range]	92.0510μs	54.5309μs	18.3382 KOps/s	17.6643 KOps/s	$\color{#35bf28}+3.82\%$
test_setitem_dim[tuple]	56.4210μs	33.7076μs	29.6669 KOps/s	28.5984 KOps/s	$\color{#35bf28}+3.74\%$
test_setitem	64.4710μs	15.1499μs	66.0069 KOps/s	61.9503 KOps/s	$\textbf{\color{#35bf28}+6.55\%}$
test_set	73.1410μs	14.6505μs	68.2572 KOps/s	64.1520 KOps/s	$\textbf{\color{#35bf28}+6.40\%}$
test_set_shared	1.6223ms	0.1524ms	6.5627 KOps/s	6.5742 KOps/s	$\color{#d91a1a}-0.18\%$
test_update	0.4870ms	16.6333μs	60.1203 KOps/s	50.7224 KOps/s	$\textbf{\color{#35bf28}+18.53\%}$
test_update_nested	80.0210μs	21.9335μs	45.5923 KOps/s	39.4708 KOps/s	$\textbf{\color{#35bf28}+15.51\%}$
test_update__nested	0.4433ms	26.0066μs	38.4517 KOps/s	37.7959 KOps/s	$\color{#35bf28}+1.74\%$
test_set_nested	85.0820μs	15.8492μs	63.0946 KOps/s	58.8107 KOps/s	$\textbf{\color{#35bf28}+7.28\%}$
test_set_nested_new	83.3020μs	17.9540μs	55.6980 KOps/s	51.7252 KOps/s	$\textbf{\color{#35bf28}+7.68\%}$
test_select	82.8020μs	29.9924μs	33.3418 KOps/s	31.5719 KOps/s	$\textbf{\color{#35bf28}+5.61\%}$
test_select_nested	85.6820μs	44.5053μs	22.4692 KOps/s	22.5907 KOps/s	$\color{#d91a1a}-0.54\%$
test_exclude_nested	99.5220μs	65.2053μs	15.3362 KOps/s	15.9730 KOps/s	$\color{#d91a1a}-3.99\%$
test_empty[True]	0.3355ms	0.2912ms	3.4335 KOps/s	3.4746 KOps/s	$\color{#d91a1a}-1.18\%$
test_empty[False]	4.2180μs	0.8408μs	1.1893 MOps/s	1.2082 MOps/s	$\color{#d91a1a}-1.56\%$
test_to	86.6420μs	56.5361μs	17.6878 KOps/s	17.9152 KOps/s	$\color{#d91a1a}-1.27\%$
test_to_nonblocking	83.9220μs	49.2615μs	20.2998 KOps/s	20.3592 KOps/s	$\color{#d91a1a}-0.29\%$
test_unbind_speed	1.5826ms	0.2465ms	4.0562 KOps/s	4.1031 KOps/s	$\color{#d91a1a}-1.14\%$
test_unbind_speed_stack0	0.3102ms	0.2426ms	4.1217 KOps/s	4.1252 KOps/s	$\color{#d91a1a}-0.09\%$
test_unbind_speed_stack1	93.7305ms	0.6690ms	1.4947 KOps/s	1.4878 KOps/s	$\color{#35bf28}+0.47\%$
test_split	94.5893ms	1.6222ms	616.4312 Ops/s	627.0729 Ops/s	$\color{#d91a1a}-1.70\%$
test_chunk	95.0014ms	1.6304ms	613.3295 Ops/s	627.1027 Ops/s	$\color{#d91a1a}-2.20\%$
test_consolidate[False-None]	96.7915ms	3.0210ms	331.0136 Ops/s	337.6421 Ops/s	$\color{#d91a1a}-1.96\%$
test_consolidate[default-None]	1.7845ms	1.7052ms	586.4398 Ops/s	586.4683 Ops/s	$-0.00\%$
test_consolidate[reduce-overhead-None]	1.8175ms	1.7585ms	568.6622 Ops/s	575.3135 Ops/s	$\color{#d91a1a}-1.16\%$
test_consolidate_njt[False-None]	6.9874ms	6.6221ms	151.0085 Ops/s	109.4341 Ops/s	$\textbf{\color{#35bf28}+37.99\%}$
test_to[False-False-None]	1.8579ms	1.7668ms	566.0051 Ops/s	569.8011 Ops/s	$\color{#d91a1a}-0.67\%$
test_to[True-False-None]	1.6355ms	1.3858ms	721.6282 Ops/s	743.9470 Ops/s	$\color{#d91a1a}-3.00\%$
test_to[within-False-None]	4.5208ms	4.2200ms	236.9662 Ops/s	239.3348 Ops/s	$\color{#d91a1a}-0.99\%$
test_to[True-default-None]	5.6636ms	5.3348ms	187.4486 Ops/s	181.8287 Ops/s	$\color{#35bf28}+3.09\%$
test_to_njt[False-False-None]	7.9503ms	6.9711ms	143.4496 Ops/s	134.6322 Ops/s	$\textbf{\color{#35bf28}+6.55\%}$
test_to_njt[True-False-None]	5.7504ms	5.4702ms	182.8081 Ops/s	168.6398 Ops/s	$\textbf{\color{#35bf28}+8.40\%}$
test_to_njt[within-False-None]	12.9846ms	12.4582ms	80.2683 Ops/s	77.1273 Ops/s	$\color{#35bf28}+4.07\%$
test_creation[device0]	0.4742ms	80.9347μs	12.3556 KOps/s	11.8492 KOps/s	$\color{#35bf28}+4.27\%$
test_creation_from_tensor	0.8415ms	84.2733μs	11.8662 KOps/s	11.3301 KOps/s	$\color{#35bf28}+4.73\%$
test_add_one[memmap_tensor0]	0.2472ms	7.1651μs	139.5646 KOps/s	145.7344 KOps/s	$\color{#d91a1a}-4.23\%$
test_contiguous[memmap_tensor0]	21.3559μs	0.4180μs	2.3924 MOps/s	2.4241 MOps/s	$\color{#d91a1a}-1.31\%$
test_stack[memmap_tensor0]	46.7000μs	4.4905μs	222.6906 KOps/s	225.9731 KOps/s	$\color{#d91a1a}-1.45\%$
test_memmaptd_index	1.5871ms	0.2547ms	3.9258 KOps/s	3.9720 KOps/s	$\color{#d91a1a}-1.16\%$
test_memmaptd_index_astensor	0.6118ms	0.3181ms	3.1436 KOps/s	3.1455 KOps/s	$\color{#d91a1a}-0.06\%$
test_memmaptd_index_op	1.0418ms	0.5801ms	1.7238 KOps/s	1.6149 KOps/s	$\textbf{\color{#35bf28}+6.74\%}$
test_serialize_model	0.1318s	0.1309s	7.6400 Ops/s	7.6604 Ops/s	$\color{#d91a1a}-0.27\%$
test_serialize_model_pickle	1.3617s	1.2204s	0.8194 Ops/s	0.8234 Ops/s	$\color{#d91a1a}-0.49\%$
test_serialize_weights	0.1310s	0.1300s	7.6914 Ops/s	7.6771 Ops/s	$\color{#35bf28}+0.19\%$
test_serialize_weights_returnearly	0.3039s	54.5189ms	18.3423 Ops/s	14.5271 Ops/s	$\textbf{\color{#35bf28}+26.26\%}$
test_serialize_weights_pickle	1.3609s	1.2225s	0.8180 Ops/s	0.8197 Ops/s	$\color{#d91a1a}-0.20\%$
test_reshape_pytree	49.7310μs	22.5219μs	44.4012 KOps/s	45.1330 KOps/s	$\color{#d91a1a}-1.62\%$
test_reshape_td	57.2310μs	27.5622μs	36.2816 KOps/s	36.8866 KOps/s	$\color{#d91a1a}-1.64\%$
test_view_pytree	52.3910μs	22.3005μs	44.8421 KOps/s	45.3112 KOps/s	$\color{#d91a1a}-1.04\%$
test_view_td	65.4810μs	32.3764μs	30.8867 KOps/s	31.0543 KOps/s	$\color{#d91a1a}-0.54\%$
test_unbind_pytree	66.5910μs	28.4656μs	35.1301 KOps/s	35.5769 KOps/s	$\color{#d91a1a}-1.26\%$
test_unbind_td	0.7843ms	37.5605μs	26.6237 KOps/s	26.9830 KOps/s	$\color{#d91a1a}-1.33\%$
test_split_pytree	58.0810μs	30.7067μs	32.5662 KOps/s	32.5834 KOps/s	$\color{#d91a1a}-0.05\%$
test_split_td	0.9562ms	39.7770μs	25.1402 KOps/s	26.2719 KOps/s	$\color{#d91a1a}-4.31\%$
test_add_pytree	65.6110μs	35.9709μs	27.8003 KOps/s	28.9393 KOps/s	$\color{#d91a1a}-3.94\%$
test_add_td	98.8610μs	50.7710μs	19.6963 KOps/s	19.5832 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_add_one_nested[tensordict-compile]	0.1758ms	0.1232ms	8.1190 KOps/s	7.9482 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_add_one_nested[tensordict-eager]	0.2302ms	0.1322ms	7.5632 KOps/s	7.4712 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_add_one_nested[pytree-compile]	0.2110ms	97.0450μs	10.3045 KOps/s	10.0494 KOps/s	$\color{#35bf28}+2.54\%$
test_compile_add_one_nested[pytree-eager]	1.5107ms	0.1528ms	6.5427 KOps/s	6.6377 KOps/s	$\color{#d91a1a}-1.43\%$
test_compile_copy_nested[tensordict-compile]	59.6610μs	24.2704μs	41.2024 KOps/s	45.8353 KOps/s	$\textbf{\color{#d91a1a}-10.11\%}$
test_compile_copy_nested[tensordict-eager]	77.2110μs	29.6917μs	33.6795 KOps/s	33.4812 KOps/s	$\color{#35bf28}+0.59\%$
test_compile_copy_nested[pytree-compile]	0.2740ms	64.7975μs	15.4327 KOps/s	15.1986 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_copy_nested[pytree-eager]	98.4920μs	48.9933μs	20.4109 KOps/s	19.9896 KOps/s	$\color{#35bf28}+2.11\%$
test_compile_add_one_flat[tensordict-compile]	0.1949ms	0.1441ms	6.9381 KOps/s	7.0548 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_add_one_flat[tensordict-eager]	0.3083ms	0.2187ms	4.5719 KOps/s	4.6064 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_add_one_flat[tensorclass-compile]	0.2152ms	0.1011ms	9.8907 KOps/s	10.2020 KOps/s	$\color{#d91a1a}-3.05\%$
test_compile_add_one_flat[tensorclass-eager]	0.1224ms	59.2045μs	16.8906 KOps/s	17.9204 KOps/s	$\textbf{\color{#d91a1a}-5.75\%}$
test_compile_add_one_flat[pytree-compile]	0.2011ms	0.1376ms	7.2680 KOps/s	7.3243 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_add_one_flat[pytree-eager]	0.6127ms	0.5020ms	1.9919 KOps/s	2.0699 KOps/s	$\color{#d91a1a}-3.77\%$
test_compile_add_self_flat[tensordict-eager]	0.3924ms	0.2710ms	3.6898 KOps/s	3.8096 KOps/s	$\color{#d91a1a}-3.14\%$
test_compile_add_self_flat[tensordict-compile]	0.1901ms	0.1450ms	6.8944 KOps/s	7.0009 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_add_self_flat[tensorclass-eager]	0.1609ms	70.0734μs	14.2708 KOps/s	14.2978 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_add_self_flat[tensorclass-compile]	0.2434ms	0.1071ms	9.3397 KOps/s	9.7250 KOps/s	$\color{#d91a1a}-3.96\%$
test_compile_add_self_flat[pytree-eager]	0.6086ms	0.4228ms	2.3653 KOps/s	2.4633 KOps/s	$\color{#d91a1a}-3.98\%$
test_compile_add_self_flat[pytree-compile]	0.1931ms	0.1418ms	7.0514 KOps/s	7.3647 KOps/s	$\color{#d91a1a}-4.25\%$
test_compile_copy_flat[tensordict-compile]	56.9110μs	19.5892μs	51.0486 KOps/s	50.7700 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_copy_flat[tensordict-eager]	57.6410μs	31.4274μs	31.8194 KOps/s	31.5690 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_copy_flat[pytree-compile]	0.1773ms	69.6380μs	14.3600 KOps/s	14.2363 KOps/s	$\color{#35bf28}+0.87\%$
test_compile_copy_flat[pytree-eager]	83.0420μs	50.6816μs	19.7310 KOps/s	19.4149 KOps/s	$\color{#35bf28}+1.63\%$
test_compile_assign_and_add[tensordict-compile]	1.6317ms	0.3933ms	2.5423 KOps/s	2.2067 KOps/s	$\textbf{\color{#35bf28}+15.21\%}$
test_compile_assign_and_add[tensordict-eager]	3.0968ms	2.7198ms	367.6802 Ops/s	383.2757 Ops/s	$\color{#d91a1a}-4.07\%$
test_compile_assign_and_add[pytree-compile]	1.5931ms	0.4323ms	2.3133 KOps/s	2.2408 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_assign_and_add[pytree-eager]	2.8515ms	2.7504ms	363.5814 Ops/s	376.3057 Ops/s	$\color{#d91a1a}-3.38\%$
test_compile_indexing[tensor-tensordict-compile]	0.6112ms	0.1191ms	8.3981 KOps/s	8.6781 KOps/s	$\color{#d91a1a}-3.23\%$
test_compile_indexing[tensor-tensordict-eager]	1.5024ms	81.2986μs	12.3003 KOps/s	12.2148 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_indexing[tensor-tensorclass-compile]	0.4453ms	0.1130ms	8.8496 KOps/s	9.2474 KOps/s	$\color{#d91a1a}-4.30\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1483ms	71.2171μs	14.0416 KOps/s	13.6068 KOps/s	$\color{#35bf28}+3.19\%$
test_compile_indexing[tensor-pytree-compile]	0.1941ms	0.1080ms	9.2596 KOps/s	9.1864 KOps/s	$\color{#35bf28}+0.80\%$
test_compile_indexing[tensor-pytree-eager]	0.1533ms	71.6997μs	13.9471 KOps/s	14.1901 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_indexing[slice-tensordict-compile]	0.1769ms	0.1021ms	9.7909 KOps/s	9.8197 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_indexing[slice-tensordict-eager]	0.1392ms	17.8336μs	56.0740 KOps/s	56.1020 KOps/s	$\color{#d91a1a}-0.05\%$
test_compile_indexing[slice-tensorclass-compile]	0.1910ms	97.2601μs	10.2817 KOps/s	10.2117 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_indexing[slice-tensorclass-eager]	62.4210μs	16.2315μs	61.6086 KOps/s	62.6598 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_indexing[slice-pytree-compile]	0.1773ms	98.7444μs	10.1272 KOps/s	10.1805 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_indexing[slice-pytree-eager]	73.3410μs	16.0757μs	62.2058 KOps/s	62.2267 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_indexing[int-tensordict-compile]	0.1864ms	0.1059ms	9.4398 KOps/s	9.7336 KOps/s	$\color{#d91a1a}-3.02\%$
test_compile_indexing[int-tensordict-eager]	0.5627ms	17.7556μs	56.3202 KOps/s	56.7823 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_indexing[int-tensorclass-compile]	0.1486ms	0.1013ms	9.8674 KOps/s	10.1915 KOps/s	$\color{#d91a1a}-3.18\%$
test_compile_indexing[int-tensorclass-eager]	48.2410μs	16.1863μs	61.7806 KOps/s	62.8433 KOps/s	$\color{#d91a1a}-1.69\%$
test_compile_indexing[int-pytree-compile]	0.1991ms	98.8019μs	10.1213 KOps/s	10.1676 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_indexing[int-pytree-eager]	94.8120μs	16.1990μs	61.7323 KOps/s	62.7841 KOps/s	$\color{#d91a1a}-1.68\%$
test_mod_add[eager]	93.2710μs	37.4791μs	26.6815 KOps/s	25.3414 KOps/s	$\textbf{\color{#35bf28}+5.29\%}$
test_mod_add[compile]	0.3314ms	84.3280μs	11.8585 KOps/s	12.2739 KOps/s	$\color{#d91a1a}-3.38\%$
test_mod_add[compile-overhead]	0.3301ms	0.1692ms	5.9091 KOps/s	5.7084 KOps/s	$\color{#35bf28}+3.51\%$
test_mod_wrap[eager]	0.3342ms	0.2614ms	3.8262 KOps/s	3.8528 KOps/s	$\color{#d91a1a}-0.69\%$
test_mod_wrap[compile]	0.7042ms	0.2875ms	3.4781 KOps/s	3.4519 KOps/s	$\color{#35bf28}+0.76\%$
test_mod_wrap[compile-overhead]	7.0937ms	3.6677ms	272.6470 Ops/s	271.1666 Ops/s	$\color{#35bf28}+0.55\%$
test_mod_wrap_and_backward[eager]	1.5401ms	1.4153ms	706.5411 Ops/s	660.5606 Ops/s	$\textbf{\color{#35bf28}+6.96\%}$
test_mod_wrap_and_backward[compile]	1.4776ms	1.2914ms	774.3345 Ops/s	705.0158 Ops/s	$\textbf{\color{#35bf28}+9.83\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3873ms	0.9472ms	1.0557 KOps/s	900.9079 Ops/s	$\textbf{\color{#35bf28}+17.18\%}$
test_seq_add[eager]	0.2077ms	0.1191ms	8.3960 KOps/s	8.3026 KOps/s	$\color{#35bf28}+1.13\%$
test_seq_add[compile]	0.2390ms	90.3318μs	11.0703 KOps/s	11.2626 KOps/s	$\color{#d91a1a}-1.71\%$
test_seq_add[compile-overhead]	0.1784ms	0.1308ms	7.6464 KOps/s	7.6519 KOps/s	$\color{#d91a1a}-0.07\%$
test_seq_wrap[eager]	0.5319ms	0.4292ms	2.3299 KOps/s	2.2700 KOps/s	$\color{#35bf28}+2.64\%$
test_seq_wrap[compile]	0.4263ms	0.3081ms	3.2461 KOps/s	3.2679 KOps/s	$\color{#d91a1a}-0.67\%$
test_seq_wrap[compile-overhead]	0.2897ms	0.2270ms	4.4061 KOps/s	4.3516 KOps/s	$\color{#35bf28}+1.25\%$
test_func_call_runtime[False-eager]	0.8512ms	0.7640ms	1.3088 KOps/s	1.3151 KOps/s	$\color{#d91a1a}-0.48\%$
test_func_call_runtime[False-compile]	0.8686ms	0.7624ms	1.3116 KOps/s	1.3005 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_runtime[False-compile-overhead]	0.4177ms	0.3691ms	2.7093 KOps/s	2.7002 KOps/s	$\color{#35bf28}+0.34\%$
test_func_call_runtime[True-eager]	1.0888ms	0.9283ms	1.0772 KOps/s	1.0867 KOps/s	$\color{#d91a1a}-0.88\%$
test_func_call_runtime[True-compile]	0.9946ms	0.7820ms	1.2788 KOps/s	1.2699 KOps/s	$\color{#35bf28}+0.70\%$
test_func_call_runtime[True-compile-overhead]	0.5343ms	0.3913ms	2.5557 KOps/s	2.4841 KOps/s	$\color{#35bf28}+2.88\%$
test_func_call_cm_runtime[False-eager]	0.8103ms	0.7599ms	1.3160 KOps/s	1.3209 KOps/s	$\color{#d91a1a}-0.37\%$
test_func_call_cm_runtime[False-compile]	1.3193ms	0.7638ms	1.3093 KOps/s	1.2346 KOps/s	$\textbf{\color{#35bf28}+6.05\%}$
test_func_call_cm_runtime[False-compile-overhead]	0.4689ms	0.3727ms	2.6832 KOps/s	2.6907 KOps/s	$\color{#d91a1a}-0.28\%$
test_func_call_cm_runtime[True-eager]	1.1517ms	1.0331ms	967.9273 Ops/s	968.1841 Ops/s	$\color{#d91a1a}-0.03\%$
test_func_call_cm_runtime[True-compile]	0.9465ms	0.8159ms	1.2257 KOps/s	1.2257 KOps/s	$+0.00\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4631ms	0.4149ms	2.4102 KOps/s	2.3958 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_func_call_cm_runtime[eager]	2.5959ms	2.1276ms	470.0050 Ops/s	462.9727 Ops/s	$\color{#35bf28}+1.52\%$
test_vmap_func_call_cm_runtime[compile]	0.9236ms	0.8234ms	1.2145 KOps/s	1.1969 KOps/s	$\color{#35bf28}+1.48\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4708ms	0.4176ms	2.3947 KOps/s	2.3721 KOps/s	$\color{#35bf28}+0.95\%$
test_distributed	2.3602ms	0.1338ms	7.4719 KOps/s	7.9925 KOps/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_tdmodule	31.3310μs	18.7402μs	53.3612 KOps/s	48.9263 KOps/s	$\textbf{\color{#35bf28}+9.06\%}$
test_tdmodule_dispatch	55.9210μs	34.1832μs	29.2541 KOps/s	26.6822 KOps/s	$\textbf{\color{#35bf28}+9.64\%}$
test_tdseq	42.4100μs	20.2777μs	49.3153 KOps/s	46.0008 KOps/s	$\textbf{\color{#35bf28}+7.21\%}$
test_tdseq_dispatch	57.6610μs	36.8570μs	27.1319 KOps/s	24.6153 KOps/s	$\textbf{\color{#35bf28}+10.22\%}$
test_instantiation_functorch	1.6586ms	1.5754ms	634.7432 Ops/s	626.6863 Ops/s	$\color{#35bf28}+1.29\%$
test_exec_functorch	0.2291ms	0.1528ms	6.5433 KOps/s	6.8882 KOps/s	$\textbf{\color{#d91a1a}-5.01\%}$
test_exec_functional_call	0.2138ms	0.1424ms	7.0227 KOps/s	7.0936 KOps/s	$\color{#d91a1a}-1.00\%$
test_exec_td_decorator	0.4169ms	0.1930ms	5.1815 KOps/s	5.2381 KOps/s	$\color{#d91a1a}-1.08\%$
test_vmap_mlp_speed_decorator[True-True]	0.7983ms	0.6992ms	1.4303 KOps/s	1.4259 KOps/s	$\color{#35bf28}+0.31\%$
test_vmap_mlp_speed_decorator[True-False]	0.8106ms	0.6969ms	1.4350 KOps/s	1.4233 KOps/s	$\color{#35bf28}+0.82\%$
test_vmap_mlp_speed_decorator[False-True]	0.7336ms	0.6116ms	1.6352 KOps/s	1.6484 KOps/s	$\color{#d91a1a}-0.80\%$
test_vmap_mlp_speed_decorator[False-False]	0.7660ms	0.6092ms	1.6416 KOps/s	1.6411 KOps/s	$\color{#35bf28}+0.03\%$
test_vmap_transformer_speed_decorator[True-True]	19.6848ms	19.5692ms	51.1006 Ops/s	51.1110 Ops/s	$\color{#d91a1a}-0.02\%$
test_vmap_transformer_speed_decorator[True-False]	19.6952ms	19.5239ms	51.2193 Ops/s	51.1217 Ops/s	$\color{#35bf28}+0.19\%$
test_vmap_transformer_speed_decorator[False-True]	19.6050ms	19.4579ms	51.3929 Ops/s	51.4107 Ops/s	$\color{#d91a1a}-0.03\%$
test_vmap_transformer_speed_decorator[False-False]	19.5161ms	19.3970ms	51.5544 Ops/s	51.4433 Ops/s	$\color{#35bf28}+0.22\%$
test_to_module_speed[True]	1.0764ms	0.9874ms	1.0128 KOps/s	1.0221 KOps/s	$\color{#d91a1a}-0.91\%$
test_to_module_speed[False]	1.5316ms	0.9572ms	1.0447 KOps/s	1.0306 KOps/s	$\color{#35bf28}+1.37\%$
test_tc_init	55.3710μs	33.7221μs	29.6541 KOps/s	27.2778 KOps/s	$\textbf{\color{#35bf28}+8.71\%}$
test_tc_init_nested	0.1672ms	67.5715μs	14.7991 KOps/s	13.6452 KOps/s	$\textbf{\color{#35bf28}+8.46\%}$
test_tc_first_layer_tensor	7.5557μs	0.6995μs	1.4297 MOps/s	1.3984 MOps/s	$\color{#35bf28}+2.24\%$
test_tc_first_layer_nontensor	25.5110μs	2.2752μs	439.5312 KOps/s	425.5230 KOps/s	$\color{#35bf28}+3.29\%$
test_tc_second_layer_tensor	7.5300μs	1.4551μs	687.2476 KOps/s	700.1098 KOps/s	$\color{#d91a1a}-1.84\%$
test_tc_second_layer_nontensor	29.1310μs	3.0251μs	330.5712 KOps/s	324.2155 KOps/s	$\color{#35bf28}+1.96\%$
test_unbind	0.2301s	10.2343ms	97.7104 Ops/s	142.8901 Ops/s	$\textbf{\color{#d91a1a}-31.62\%}$
test_full_like	10.2356ms	9.5043ms	105.2157 Ops/s	101.4691 Ops/s	$\color{#35bf28}+3.69\%$
test_zeros_like	6.8208ms	4.3949ms	227.5345 Ops/s	228.6243 Ops/s	$\color{#d91a1a}-0.48\%$
test_ones_like	4.8656ms	4.3738ms	228.6328 Ops/s	228.9806 Ops/s	$\color{#d91a1a}-0.15\%$
test_clone	7.7262ms	6.7444ms	148.2703 Ops/s	148.4143 Ops/s	$\color{#d91a1a}-0.10\%$
test_squeeze	54.4410μs	9.7511μs	102.5526 KOps/s	102.1737 KOps/s	$\color{#35bf28}+0.37\%$
test_unsqueeze	0.2172ms	74.3979μs	13.4412 KOps/s	13.5290 KOps/s	$\color{#d91a1a}-0.65\%$
test_split	0.3924ms	0.1687ms	5.9286 KOps/s	6.1879 KOps/s	$\color{#d91a1a}-4.19\%$
test_permute	0.2778ms	0.1931ms	5.1791 KOps/s	5.3031 KOps/s	$\color{#d91a1a}-2.34\%$
test_stack	51.9318ms	51.4245ms	19.4460 Ops/s	19.5330 Ops/s	$\color{#d91a1a}-0.45\%$
test_cat	52.6105ms	51.5841ms	19.3858 Ops/s	19.4470 Ops/s	$\color{#d91a1a}-0.31\%$