[Feature] TensorDict.record_stream

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] TensorDict.record_stream #1016

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}33$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	57.8090μs	19.7476μs	50.6391 KOps/s	47.8349 KOps/s	$\textbf{\color{#35bf28}+5.86\%}$
test_plain_set_stack_nested	50.3040μs	19.8080μs	50.4847 KOps/s	46.9167 KOps/s	$\textbf{\color{#35bf28}+7.61\%}$
test_plain_set_nested_inplace	82.6070μs	21.5445μs	46.4156 KOps/s	43.5681 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_plain_set_stack_nested_inplace	72.1650μs	21.0420μs	47.5239 KOps/s	44.2032 KOps/s	$\textbf{\color{#35bf28}+7.51\%}$
test_items	22.1820μs	4.2157μs	237.2103 KOps/s	246.1842 KOps/s	$\color{#d91a1a}-3.65\%$
test_items_nested	0.5955ms	0.3732ms	2.6796 KOps/s	2.7582 KOps/s	$\color{#d91a1a}-2.85\%$
test_items_nested_locked	0.7623ms	0.3733ms	2.6790 KOps/s	2.7979 KOps/s	$\color{#d91a1a}-4.25\%$
test_items_nested_leaf	0.1215ms	68.1096μs	14.6822 KOps/s	14.5594 KOps/s	$\color{#35bf28}+0.84\%$
test_items_stack_nested	0.7047ms	0.3814ms	2.6221 KOps/s	2.7649 KOps/s	$\textbf{\color{#d91a1a}-5.16\%}$
test_items_stack_nested_leaf	0.1342ms	71.0043μs	14.0836 KOps/s	14.0226 KOps/s	$\color{#35bf28}+0.44\%$
test_items_stack_nested_locked	0.4623ms	0.3762ms	2.6579 KOps/s	2.7305 KOps/s	$\color{#d91a1a}-2.66\%$
test_keys	42.7200μs	3.6244μs	275.9080 KOps/s	282.9106 KOps/s	$\color{#d91a1a}-2.48\%$
test_keys_nested	0.1830ms	98.9406μs	10.1071 KOps/s	10.0450 KOps/s	$\color{#35bf28}+0.62\%$
test_keys_nested_locked	0.6957ms	0.1047ms	9.5533 KOps/s	9.5186 KOps/s	$\color{#35bf28}+0.36\%$
test_keys_nested_leaf	0.1413ms	82.0120μs	12.1933 KOps/s	12.1522 KOps/s	$\color{#35bf28}+0.34\%$
test_keys_stack_nested	0.1654ms	99.0379μs	10.0971 KOps/s	10.0016 KOps/s	$\color{#35bf28}+0.96\%$
test_keys_stack_nested_leaf	0.1494ms	80.7987μs	12.3764 KOps/s	12.1422 KOps/s	$\color{#35bf28}+1.93\%$
test_keys_stack_nested_locked	0.1895ms	0.1038ms	9.6312 KOps/s	9.2383 KOps/s	$\color{#35bf28}+4.25\%$
test_values	6.6904μs	1.0570μs	946.1029 KOps/s	947.0718 KOps/s	$\color{#d91a1a}-0.10\%$
test_values_nested	0.1382ms	74.0404μs	13.5061 KOps/s	13.2217 KOps/s	$\color{#35bf28}+2.15\%$
test_values_nested_locked	0.1350ms	74.9287μs	13.3460 KOps/s	13.2011 KOps/s	$\color{#35bf28}+1.10\%$
test_values_nested_leaf	0.1169ms	62.0084μs	16.1268 KOps/s	16.0685 KOps/s	$\color{#35bf28}+0.36\%$
test_values_stack_nested	0.1247ms	76.1907μs	13.1250 KOps/s	12.8613 KOps/s	$\color{#35bf28}+2.05\%$
test_values_stack_nested_leaf	0.1209ms	61.6146μs	16.2299 KOps/s	15.9895 KOps/s	$\color{#35bf28}+1.50\%$
test_values_stack_nested_locked	0.1467ms	76.1302μs	13.1354 KOps/s	12.6032 KOps/s	$\color{#35bf28}+4.22\%$
test_membership	3.3433μs	0.7420μs	1.3477 MOps/s	1.1358 MOps/s	$\textbf{\color{#35bf28}+18.66\%}$
test_membership_nested	43.9320μs	2.8555μs	350.2073 KOps/s	362.1335 KOps/s	$\color{#d91a1a}-3.29\%$
test_membership_nested_leaf	35.1850μs	2.8450μs	351.4903 KOps/s	361.6725 KOps/s	$\color{#d91a1a}-2.82\%$
test_membership_stacked_nested	47.6390μs	2.8039μs	356.6459 KOps/s	362.0152 KOps/s	$\color{#d91a1a}-1.48\%$
test_membership_stacked_nested_leaf	29.0940μs	2.8472μs	351.2283 KOps/s	363.8387 KOps/s	$\color{#d91a1a}-3.47\%$
test_membership_nested_last	47.7700μs	4.1523μs	240.8303 KOps/s	242.3871 KOps/s	$\color{#d91a1a}-0.64\%$
test_membership_nested_leaf_last	35.5960μs	4.1715μs	239.7215 KOps/s	248.4352 KOps/s	$\color{#d91a1a}-3.51\%$
test_membership_stacked_nested_last	25.6080μs	4.1242μs	242.4686 KOps/s	250.0185 KOps/s	$\color{#d91a1a}-3.02\%$
test_membership_stacked_nested_leaf_last	28.2430μs	4.1463μs	241.1782 KOps/s	249.7452 KOps/s	$\color{#d91a1a}-3.43\%$
test_nested_getleaf	51.0660μs	10.5551μs	94.7413 KOps/s	93.2374 KOps/s	$\color{#35bf28}+1.61\%$
test_nested_get	49.7830μs	10.1341μs	98.6770 KOps/s	96.9506 KOps/s	$\color{#35bf28}+1.78\%$
test_stacked_getleaf	32.8010μs	10.6812μs	93.6223 KOps/s	91.3217 KOps/s	$\color{#35bf28}+2.52\%$
test_stacked_get	56.7160μs	10.0707μs	99.2975 KOps/s	99.0581 KOps/s	$\color{#35bf28}+0.24\%$
test_nested_getitemleaf	41.0970μs	11.1192μs	89.9349 KOps/s	89.8861 KOps/s	$\color{#35bf28}+0.05\%$
test_nested_getitem	56.5450μs	10.2781μs	97.2945 KOps/s	95.9438 KOps/s	$\color{#35bf28}+1.41\%$
test_stacked_getitemleaf	53.9810μs	11.2260μs	89.0792 KOps/s	91.0294 KOps/s	$\color{#d91a1a}-2.14\%$
test_stacked_getitem	60.9140μs	10.2335μs	97.7187 KOps/s	96.0904 KOps/s	$\color{#35bf28}+1.69\%$
test_lock_nested	83.1198ms	0.5697ms	1.7552 KOps/s	2.0065 KOps/s	$\textbf{\color{#d91a1a}-12.53\%}$
test_lock_stack_nested	0.9177ms	0.4557ms	2.1944 KOps/s	2.1726 KOps/s	$\color{#35bf28}+1.01\%$
test_unlock_nested	90.1293ms	0.4983ms	2.0068 KOps/s	2.3905 KOps/s	$\textbf{\color{#d91a1a}-16.05\%}$
test_unlock_stack_nested	0.8186ms	0.3880ms	2.5771 KOps/s	2.6399 KOps/s	$\color{#d91a1a}-2.38\%$
test_flatten_speed	0.1610ms	87.6151μs	11.4136 KOps/s	11.4397 KOps/s	$\color{#d91a1a}-0.23\%$
test_unflatten_speed	0.6144ms	0.4700ms	2.1276 KOps/s	2.1206 KOps/s	$\color{#35bf28}+0.33\%$
test_common_ops	4.2299ms	1.1089ms	901.8237 Ops/s	868.5351 Ops/s	$\color{#35bf28}+3.83\%$
test_creation	31.0690μs	2.0809μs	480.5581 KOps/s	486.9312 KOps/s	$\color{#d91a1a}-1.31\%$
test_creation_empty	53.9510μs	16.3422μs	61.1914 KOps/s	51.5454 KOps/s	$\textbf{\color{#35bf28}+18.71\%}$
test_creation_nested_1	67.7460μs	19.4653μs	51.3735 KOps/s	43.9063 KOps/s	$\textbf{\color{#35bf28}+17.01\%}$
test_creation_nested_2	63.5990μs	23.8593μs	41.9123 KOps/s	37.2786 KOps/s	$\textbf{\color{#35bf28}+12.43\%}$
test_clone	0.1867ms	17.5382μs	57.0185 KOps/s	58.0304 KOps/s	$\color{#d91a1a}-1.74\%$
test_getitem[int]	1.1681ms	16.7077μs	59.8525 KOps/s	58.2499 KOps/s	$\color{#35bf28}+2.75\%$
test_getitem[slice_int]	0.1441ms	30.7387μs	32.5323 KOps/s	31.9232 KOps/s	$\color{#35bf28}+1.91\%$
test_getitem[range]	0.8185ms	59.7807μs	16.7278 KOps/s	17.1368 KOps/s	$\color{#d91a1a}-2.39\%$
test_getitem[tuple]	0.1525ms	24.7450μs	40.4122 KOps/s	39.2698 KOps/s	$\color{#35bf28}+2.91\%$
test_getitem[list]	0.1650ms	54.2236μs	18.4422 KOps/s	18.5126 KOps/s	$\color{#d91a1a}-0.38\%$
test_setitem_dim[int]	62.8370μs	33.4983μs	29.8522 KOps/s	29.8697 KOps/s	$\color{#d91a1a}-0.06\%$
test_setitem_dim[slice_int]	0.1122ms	62.4561μs	16.0112 KOps/s	16.5342 KOps/s	$\color{#d91a1a}-3.16\%$
test_setitem_dim[range]	0.1449ms	84.7425μs	11.8005 KOps/s	11.8681 KOps/s	$\color{#d91a1a}-0.57\%$
test_setitem_dim[tuple]	98.3840μs	50.1206μs	19.9519 KOps/s	20.1242 KOps/s	$\color{#d91a1a}-0.86\%$
test_setitem	73.6870μs	28.9742μs	34.5134 KOps/s	32.0359 KOps/s	$\textbf{\color{#35bf28}+7.73\%}$
test_set	0.2518ms	28.1426μs	35.5333 KOps/s	32.6689 KOps/s	$\textbf{\color{#35bf28}+8.77\%}$
test_set_shared	6.1110ms	0.2168ms	4.6128 KOps/s	4.6008 KOps/s	$\color{#35bf28}+0.26\%$
test_update	0.2757ms	34.6408μs	28.8677 KOps/s	25.8322 KOps/s	$\textbf{\color{#35bf28}+11.75\%}$
test_update_nested	1.8965ms	45.6021μs	21.9288 KOps/s	20.3513 KOps/s	$\textbf{\color{#35bf28}+7.75\%}$
test_update__nested	81.2820μs	35.2984μs	28.3299 KOps/s	27.1728 KOps/s	$\color{#35bf28}+4.26\%$
test_set_nested	0.3550ms	31.2292μs	32.0213 KOps/s	30.5810 KOps/s	$\color{#35bf28}+4.71\%$
test_set_nested_new	0.1934ms	35.9136μs	27.8446 KOps/s	26.0412 KOps/s	$\textbf{\color{#35bf28}+6.92\%}$
test_select	0.2517ms	52.9539μs	18.8843 KOps/s	18.0132 KOps/s	$\color{#35bf28}+4.84\%$
test_select_nested	0.1288ms	59.7454μs	16.7377 KOps/s	16.5404 KOps/s	$\color{#35bf28}+1.19\%$
test_exclude_nested	0.1751ms	75.0117μs	13.3313 KOps/s	13.2631 KOps/s	$\color{#35bf28}+0.51\%$
test_empty[True]	0.6608ms	0.3152ms	3.1724 KOps/s	3.1370 KOps/s	$\color{#35bf28}+1.13\%$
test_empty[False]	22.9478μs	1.1958μs	836.2627 KOps/s	832.1309 KOps/s	$\color{#35bf28}+0.50\%$
test_unbind_speed	0.5216ms	0.2999ms	3.3343 KOps/s	3.2468 KOps/s	$\color{#35bf28}+2.70\%$
test_unbind_speed_stack0	0.4290ms	0.3004ms	3.3290 KOps/s	3.3582 KOps/s	$\color{#d91a1a}-0.87\%$
test_unbind_speed_stack1	98.1837ms	0.8248ms	1.2125 KOps/s	1.4532 KOps/s	$\textbf{\color{#d91a1a}-16.56\%}$
test_split	92.9165ms	2.1574ms	463.5170 Ops/s	452.7703 Ops/s	$\color{#35bf28}+2.37\%$
test_chunk	3.2902ms	1.9859ms	503.5394 Ops/s	450.5945 Ops/s	$\textbf{\color{#35bf28}+11.75\%}$
test_creation[device0]	3.8242ms	0.1203ms	8.3107 KOps/s	8.4631 KOps/s	$\color{#d91a1a}-1.80\%$
test_creation_from_tensor	0.2467ms	0.1174ms	8.5153 KOps/s	8.4477 KOps/s	$\color{#35bf28}+0.80\%$
test_add_one[memmap_tensor0]	0.2535ms	7.1436μs	139.9857 KOps/s	136.1869 KOps/s	$\color{#35bf28}+2.79\%$
test_contiguous[memmap_tensor0]	20.5080μs	1.9772μs	505.7571 KOps/s	536.1249 KOps/s	$\textbf{\color{#d91a1a}-5.66\%}$
test_stack[memmap_tensor0]	0.1000ms	5.7674μs	173.3884 KOps/s	177.8316 KOps/s	$\color{#d91a1a}-2.50\%$
test_memmaptd_index	1.1715ms	0.4004ms	2.4976 KOps/s	2.5303 KOps/s	$\color{#d91a1a}-1.29\%$
test_memmaptd_index_astensor	0.7345ms	0.4766ms	2.0981 KOps/s	2.0901 KOps/s	$\color{#35bf28}+0.38\%$
test_memmaptd_index_op	93.6502ms	1.0890ms	918.2857 Ops/s	955.2303 Ops/s	$\color{#d91a1a}-3.87\%$
test_serialize_model	0.1297s	0.1213s	8.2442 Ops/s	8.1407 Ops/s	$\color{#35bf28}+1.27\%$
test_serialize_model_pickle	0.5012s	0.4089s	2.4455 Ops/s	2.5316 Ops/s	$\color{#d91a1a}-3.40\%$
test_serialize_weights	0.1232s	0.1162s	8.6023 Ops/s	7.4643 Ops/s	$\textbf{\color{#35bf28}+15.25\%}$
test_serialize_weights_returnearly	0.2744s	0.1749s	5.7186 Ops/s	6.1440 Ops/s	$\textbf{\color{#d91a1a}-6.92\%}$
test_serialize_weights_pickle	0.5774s	0.4299s	2.3263 Ops/s	2.4863 Ops/s	$\textbf{\color{#d91a1a}-6.44\%}$
test_serialize_weights_filesystem	0.1438s	0.1413s	7.0763 Ops/s	6.7427 Ops/s	$\color{#35bf28}+4.95\%$
test_serialize_model_filesystem	0.1598s	0.1479s	6.7603 Ops/s	6.0689 Ops/s	$\textbf{\color{#35bf28}+11.39\%}$
test_reshape_pytree	79.1080μs	40.3629μs	24.7752 KOps/s	25.7797 KOps/s	$\color{#d91a1a}-3.90\%$
test_reshape_td	0.1354ms	47.2044μs	21.1845 KOps/s	20.4412 KOps/s	$\color{#35bf28}+3.64\%$
test_view_pytree	87.8940μs	39.3747μs	25.3970 KOps/s	25.6677 KOps/s	$\color{#d91a1a}-1.05\%$
test_view_td	0.1459ms	52.6707μs	18.9859 KOps/s	18.2688 KOps/s	$\color{#35bf28}+3.93\%$
test_unbind_pytree	76.5030μs	36.1711μs	27.6464 KOps/s	27.6723 KOps/s	$\color{#d91a1a}-0.09\%$
test_unbind_td	0.3160ms	43.5398μs	22.9675 KOps/s	21.3990 KOps/s	$\textbf{\color{#35bf28}+7.33\%}$
test_split_pytree	0.1039ms	38.0172μs	26.3039 KOps/s	26.1244 KOps/s	$\color{#35bf28}+0.69\%$
test_split_td	0.5229ms	57.6761μs	17.3382 KOps/s	17.4134 KOps/s	$\color{#d91a1a}-0.43\%$
test_add_pytree	0.1175ms	45.7848μs	21.8413 KOps/s	22.1389 KOps/s	$\color{#d91a1a}-1.34\%$
test_add_td	0.1589ms	78.9896μs	12.6599 KOps/s	11.7198 KOps/s	$\textbf{\color{#35bf28}+8.02\%}$
test_compile_add_one_nested[tensordict-compile]	0.1384ms	59.6206μs	16.7727 KOps/s	17.5171 KOps/s	$\color{#d91a1a}-4.25\%$
test_compile_add_one_nested[tensordict-eager]	0.4267ms	0.1796ms	5.5693 KOps/s	5.6105 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_add_one_nested[pytree-compile]	0.1351ms	58.3802μs	17.1291 KOps/s	17.4221 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_add_one_nested[pytree-eager]	0.3381ms	0.1436ms	6.9629 KOps/s	7.0508 KOps/s	$\color{#d91a1a}-1.25\%$
test_compile_copy_nested[tensordict-compile]	72.4050μs	21.6438μs	46.2026 KOps/s	46.3840 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_copy_nested[tensordict-eager]	0.1430ms	69.1630μs	14.4586 KOps/s	15.0714 KOps/s	$\color{#d91a1a}-4.07\%$
test_compile_copy_nested[pytree-compile]	0.1471ms	77.5081μs	12.9019 KOps/s	13.0819 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_copy_nested[pytree-eager]	0.1841ms	70.6892μs	14.1464 KOps/s	14.4154 KOps/s	$\color{#d91a1a}-1.87\%$
test_compile_add_one_flat[tensordict-compile]	0.2638ms	0.1793ms	5.5767 KOps/s	5.7635 KOps/s	$\color{#d91a1a}-3.24\%$
test_compile_add_one_flat[tensordict-eager]	0.4261ms	0.1905ms	5.2493 KOps/s	5.1849 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_add_one_flat[tensorclass-compile]	0.1230ms	48.6074μs	20.5730 KOps/s	20.6239 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_one_flat[tensorclass-eager]	0.1726ms	69.8220μs	14.3221 KOps/s	14.6700 KOps/s	$\color{#d91a1a}-2.37\%$
test_compile_add_one_flat[pytree-compile]	0.4593ms	0.1831ms	5.4613 KOps/s	5.7282 KOps/s	$\color{#d91a1a}-4.66\%$
test_compile_add_one_flat[pytree-eager]	0.7172ms	0.2945ms	3.3954 KOps/s	3.4733 KOps/s	$\color{#d91a1a}-2.24\%$
test_compile_add_self_flat[tensordict-eager]	0.4203ms	0.2028ms	4.9317 KOps/s	4.9163 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_add_self_flat[tensordict-compile]	0.6256ms	0.1813ms	5.5143 KOps/s	5.7970 KOps/s	$\color{#d91a1a}-4.88\%$
test_compile_add_self_flat[tensorclass-eager]	0.2228ms	62.8211μs	15.9182 KOps/s	15.9772 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_add_self_flat[tensorclass-compile]	0.1173ms	48.9850μs	20.4144 KOps/s	21.2364 KOps/s	$\color{#d91a1a}-3.87\%$
test_compile_add_self_flat[pytree-eager]	0.3750ms	0.2376ms	4.2094 KOps/s	4.2457 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_add_self_flat[pytree-compile]	0.2931ms	0.1804ms	5.5419 KOps/s	5.6779 KOps/s	$\color{#d91a1a}-2.39\%$
test_compile_copy_flat[tensordict-compile]	0.2016ms	0.1078ms	9.2790 KOps/s	9.6687 KOps/s	$\color{#d91a1a}-4.03\%$
test_compile_copy_flat[tensordict-eager]	0.1348ms	60.0907μs	16.6415 KOps/s	17.1801 KOps/s	$\color{#d91a1a}-3.13\%$
test_compile_copy_flat[pytree-compile]	0.1473ms	78.0182μs	12.8175 KOps/s	12.5326 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_copy_flat[pytree-eager]	0.1266ms	69.8018μs	14.3263 KOps/s	14.1830 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_assign_and_add[tensordict-compile]	0.3854ms	0.1987ms	5.0335 KOps/s	5.1441 KOps/s	$\color{#d91a1a}-2.15\%$
test_compile_assign_and_add[tensordict-eager]	1.8350ms	1.6865ms	592.9445 Ops/s	599.6238 Ops/s	$\color{#d91a1a}-1.11\%$
test_compile_assign_and_add[pytree-compile]	0.4370ms	0.1974ms	5.0657 KOps/s	5.1955 KOps/s	$\color{#d91a1a}-2.50\%$
test_compile_assign_and_add[pytree-eager]	1.3137ms	1.1321ms	883.3349 Ops/s	915.5837 Ops/s	$\color{#d91a1a}-3.52\%$
test_compile_assign_and_add_stack[compile]	0.5157ms	0.4212ms	2.3739 KOps/s	2.4062 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_assign_and_add_stack[eager]	3.9394ms	3.6982ms	270.4040 Ops/s	255.2251 Ops/s	$\textbf{\color{#35bf28}+5.95\%}$
test_compile_indexing[tensor-tensordict-compile]	0.1170ms	35.0286μs	28.5481 KOps/s	27.8832 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_indexing[tensor-tensordict-eager]	1.1510ms	50.0505μs	19.9798 KOps/s	20.4236 KOps/s	$\color{#d91a1a}-2.17\%$
test_compile_indexing[tensor-tensorclass-compile]	80.9720μs	29.7142μs	33.6539 KOps/s	31.6485 KOps/s	$\textbf{\color{#35bf28}+6.34\%}$
test_compile_indexing[tensor-tensorclass-eager]	90.1790μs	30.2068μs	33.1052 KOps/s	33.9651 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_indexing[tensor-pytree-compile]	77.9160μs	29.4951μs	33.9039 KOps/s	31.2885 KOps/s	$\textbf{\color{#35bf28}+8.36\%}$
test_compile_indexing[tensor-pytree-eager]	70.9630μs	29.8607μs	33.4888 KOps/s	33.9741 KOps/s	$\color{#d91a1a}-1.43\%$
test_compile_indexing[slice-tensordict-compile]	0.1474ms	74.3924μs	13.4422 KOps/s	13.1872 KOps/s	$\color{#35bf28}+1.93\%$
test_compile_indexing[slice-tensordict-eager]	0.5850ms	28.0190μs	35.6901 KOps/s	36.2430 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_indexing[slice-tensorclass-compile]	0.1501ms	69.0387μs	14.4846 KOps/s	14.0261 KOps/s	$\color{#35bf28}+3.27\%$
test_compile_indexing[slice-tensorclass-eager]	83.8770μs	24.3036μs	41.1461 KOps/s	42.5667 KOps/s	$\color{#d91a1a}-3.34\%$
test_compile_indexing[slice-pytree-compile]	0.1449ms	68.1576μs	14.6719 KOps/s	14.2486 KOps/s	$\color{#35bf28}+2.97\%$
test_compile_indexing[slice-pytree-eager]	77.3150μs	24.1646μs	41.3828 KOps/s	42.4649 KOps/s	$\color{#d91a1a}-2.55\%$
test_compile_indexing[int-tensordict-compile]	0.1535ms	74.1688μs	13.4828 KOps/s	13.1598 KOps/s	$\color{#35bf28}+2.45\%$
test_compile_indexing[int-tensordict-eager]	1.2897ms	27.0832μs	36.9233 KOps/s	35.7284 KOps/s	$\color{#35bf28}+3.34\%$
test_compile_indexing[int-tensorclass-compile]	0.1329ms	68.5750μs	14.5826 KOps/s	14.0574 KOps/s	$\color{#35bf28}+3.74\%$
test_compile_indexing[int-tensorclass-eager]	72.5150μs	24.0741μs	41.5385 KOps/s	42.9430 KOps/s	$\color{#d91a1a}-3.27\%$
test_compile_indexing[int-pytree-compile]	0.1973ms	70.0825μs	14.2689 KOps/s	14.2042 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_indexing[int-pytree-eager]	97.0920μs	24.0859μs	41.5181 KOps/s	42.7735 KOps/s	$\color{#d91a1a}-2.93\%$
test_mod_add[eager]	0.1168ms	25.1950μs	39.6903 KOps/s	36.4392 KOps/s	$\textbf{\color{#35bf28}+8.92\%}$
test_mod_add[compile]	0.1045ms	40.1812μs	24.8873 KOps/s	24.1569 KOps/s	$\color{#35bf28}+3.02\%$
test_mod_add[compile-overhead]	0.1057ms	39.7823μs	25.1368 KOps/s	24.8282 KOps/s	$\color{#35bf28}+1.24\%$
test_mod_wrap[eager]	0.3119ms	0.2066ms	4.8398 KOps/s	4.8533 KOps/s	$\color{#d91a1a}-0.28\%$
test_mod_wrap[compile]	0.3537ms	0.2312ms	4.3257 KOps/s	4.2738 KOps/s	$\color{#35bf28}+1.21\%$
test_mod_wrap[compile-overhead]	0.3550ms	0.2288ms	4.3707 KOps/s	4.3265 KOps/s	$\color{#35bf28}+1.02\%$
test_mod_wrap_and_backward[eager]	12.1286ms	10.8329ms	92.3117 Ops/s	92.6425 Ops/s	$\color{#d91a1a}-0.36\%$
test_mod_wrap_and_backward[compile]	13.8425ms	11.2454ms	88.9255 Ops/s	83.6687 Ops/s	$\textbf{\color{#35bf28}+6.28\%}$
test_mod_wrap_and_backward[compile-overhead]	17.5350ms	12.7258ms	78.5804 Ops/s	85.4875 Ops/s	$\textbf{\color{#d91a1a}-8.08\%}$
test_seq_add[eager]	0.1922ms	88.3181μs	11.3227 KOps/s	10.4263 KOps/s	$\textbf{\color{#35bf28}+8.60\%}$
test_seq_add[compile]	0.2237ms	66.9642μs	14.9333 KOps/s	15.0302 KOps/s	$\color{#d91a1a}-0.64\%$
test_seq_add[compile-overhead]	0.1493ms	64.7723μs	15.4387 KOps/s	15.1796 KOps/s	$\color{#35bf28}+1.71\%$
test_seq_wrap[eager]	0.5945ms	0.3704ms	2.6997 KOps/s	2.5406 KOps/s	$\textbf{\color{#35bf28}+6.26\%}$
test_seq_wrap[compile]	1.2435ms	0.2742ms	3.6474 KOps/s	3.6318 KOps/s	$\color{#35bf28}+0.43\%$
test_seq_wrap[compile-overhead]	1.2234ms	0.2688ms	3.7198 KOps/s	3.6196 KOps/s	$\color{#35bf28}+2.77\%$
test_func_call_runtime[False-eager]	0.7695ms	0.5187ms	1.9279 KOps/s	1.8993 KOps/s	$\color{#35bf28}+1.51\%$
test_func_call_runtime[False-compile]	0.6710ms	0.5075ms	1.9704 KOps/s	2.0137 KOps/s	$\color{#d91a1a}-2.15\%$
test_func_call_runtime[False-compile-overhead]	0.6366ms	0.5026ms	1.9898 KOps/s	2.0167 KOps/s	$\color{#d91a1a}-1.34\%$
test_func_call_runtime[True-eager]	1.2276ms	0.7289ms	1.3719 KOps/s	1.3458 KOps/s	$\color{#35bf28}+1.94\%$
test_func_call_runtime[True-compile]	0.7101ms	0.5185ms	1.9285 KOps/s	1.9634 KOps/s	$\color{#d91a1a}-1.77\%$
test_func_call_runtime[True-compile-overhead]	0.6614ms	0.5160ms	1.9379 KOps/s	1.9700 KOps/s	$\color{#d91a1a}-1.63\%$
test_func_call_cm_runtime[False-eager]	1.0140ms	0.5250ms	1.9046 KOps/s	1.9231 KOps/s	$\color{#d91a1a}-0.96\%$
test_func_call_cm_runtime[False-compile]	0.6242ms	0.5064ms	1.9748 KOps/s	2.0186 KOps/s	$\color{#d91a1a}-2.17\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6590ms	0.5067ms	1.9735 KOps/s	2.0247 KOps/s	$\color{#d91a1a}-2.53\%$
test_func_call_cm_runtime[True-eager]	1.0123ms	0.8600ms	1.1628 KOps/s	1.1594 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[True-compile]	1.0015ms	0.7374ms	1.3561 KOps/s	1.3693 KOps/s	$\color{#d91a1a}-0.97\%$
test_func_call_cm_runtime[True-compile-overhead]	1.5220ms	0.7497ms	1.3338 KOps/s	1.3680 KOps/s	$\color{#d91a1a}-2.50\%$
test_vmap_func_call_cm_runtime[eager]	2.6595ms	1.8630ms	536.7731 Ops/s	539.3825 Ops/s	$\color{#d91a1a}-0.48\%$
test_vmap_func_call_cm_runtime[compile]	2.9442ms	1.9129ms	522.7541 Ops/s	517.4955 Ops/s	$\color{#35bf28}+1.02\%$
test_vmap_func_call_cm_runtime[compile-overhead]	3.1697ms	1.9148ms	522.2461 Ops/s	517.4568 Ops/s	$\color{#35bf28}+0.93\%$
test_distributed	0.2666ms	0.1284ms	7.7874 KOps/s	7.6841 KOps/s	$\color{#35bf28}+1.34\%$
test_tdmodule	47.7590μs	17.2314μs	58.0336 KOps/s	51.2797 KOps/s	$\textbf{\color{#35bf28}+13.17\%}$
test_tdmodule_dispatch	81.8340μs	34.4643μs	29.0155 KOps/s	25.9414 KOps/s	$\textbf{\color{#35bf28}+11.85\%}$
test_tdseq	39.6250μs	19.7698μs	50.5822 KOps/s	45.0715 KOps/s	$\textbf{\color{#35bf28}+12.23\%}$
test_tdseq_dispatch	69.6310μs	39.8608μs	25.0873 KOps/s	22.5395 KOps/s	$\textbf{\color{#35bf28}+11.30\%}$
test_instantiation_functorch	2.7970ms	1.5615ms	640.3988 Ops/s	626.0928 Ops/s	$\color{#35bf28}+2.28\%$
test_instantiation_td	1.8306ms	1.1638ms	859.2528 Ops/s	862.9671 Ops/s	$\color{#d91a1a}-0.43\%$
test_exec_functorch	0.3679ms	0.1876ms	5.3292 KOps/s	5.3453 KOps/s	$\color{#d91a1a}-0.30\%$
test_exec_functional_call	0.4049ms	0.1733ms	5.7705 KOps/s	5.5922 KOps/s	$\color{#35bf28}+3.19\%$
test_exec_td	0.2612ms	0.1661ms	6.0187 KOps/s	5.6806 KOps/s	$\textbf{\color{#35bf28}+5.95\%}$
test_exec_td_decorator	1.1586ms	0.2264ms	4.4162 KOps/s	4.3032 KOps/s	$\color{#35bf28}+2.63\%$
test_vmap_mlp_speed[True-True]	0.8772ms	0.6418ms	1.5580 KOps/s	1.5108 KOps/s	$\color{#35bf28}+3.12\%$
test_vmap_mlp_speed[True-False]	0.7519ms	0.6309ms	1.5850 KOps/s	1.5371 KOps/s	$\color{#35bf28}+3.11\%$
test_vmap_mlp_speed[False-True]	0.7575ms	0.4944ms	2.0226 KOps/s	1.9880 KOps/s	$\color{#35bf28}+1.74\%$
test_vmap_mlp_speed[False-False]	0.6885ms	0.4908ms	2.0373 KOps/s	1.9881 KOps/s	$\color{#35bf28}+2.48\%$
test_vmap_mlp_speed_decorator[True-True]	1.3675ms	0.6136ms	1.6296 KOps/s	1.5905 KOps/s	$\color{#35bf28}+2.46\%$
test_vmap_mlp_speed_decorator[True-False]	1.0647ms	0.6210ms	1.6103 KOps/s	1.5895 KOps/s	$\color{#35bf28}+1.31\%$
test_vmap_mlp_speed_decorator[False-True]	0.7666ms	0.5075ms	1.9705 KOps/s	1.9467 KOps/s	$\color{#35bf28}+1.22\%$
test_vmap_mlp_speed_decorator[False-False]	0.7737ms	0.5070ms	1.9723 KOps/s	1.9430 KOps/s	$\color{#35bf28}+1.51\%$
test_to_module_speed[True]	1.4427ms	1.2921ms	773.9208 Ops/s	758.9256 Ops/s	$\color{#35bf28}+1.98\%$
test_to_module_speed[False]	2.0324ms	1.2777ms	782.6337 Ops/s	784.6731 Ops/s	$\color{#d91a1a}-0.26\%$
test_tc_init	75.1420μs	42.4004μs	23.5847 KOps/s	22.2737 KOps/s	$\textbf{\color{#35bf28}+5.89\%}$
test_tc_init_nested	0.1528ms	83.0681μs	12.0383 KOps/s	10.9700 KOps/s	$\textbf{\color{#35bf28}+9.74\%}$
test_tc_first_layer_tensor	47.7500μs	1.5779μs	633.7672 KOps/s	662.6703 KOps/s	$\color{#d91a1a}-4.36\%$
test_tc_first_layer_nontensor	22.1920μs	4.7386μs	211.0312 KOps/s	213.6784 KOps/s	$\color{#d91a1a}-1.24\%$
test_tc_second_layer_tensor	39.4250μs	2.8887μs	346.1820 KOps/s	359.8134 KOps/s	$\color{#d91a1a}-3.79\%$
test_tc_second_layer_nontensor	30.6680μs	6.1111μs	163.6356 KOps/s	164.9867 KOps/s	$\color{#d91a1a}-0.82\%$
test_unbind	0.4895s	14.2859ms	69.9990 Ops/s	64.9331 Ops/s	$\textbf{\color{#35bf28}+7.80\%}$
test_full_like	14.4053ms	8.8629ms	112.8301 Ops/s	139.3967 Ops/s	$\textbf{\color{#d91a1a}-19.06\%}$
test_zeros_like	4.6895ms	3.3727ms	296.4979 Ops/s	361.2277 Ops/s	$\textbf{\color{#d91a1a}-17.92\%}$
test_ones_like	8.2147ms	3.4775ms	287.5591 Ops/s	310.3643 Ops/s	$\textbf{\color{#d91a1a}-7.35\%}$
test_clone	6.5593ms	5.4929ms	182.0528 Ops/s	183.9498 Ops/s	$\color{#d91a1a}-1.03\%$
test_squeeze	0.1040ms	13.0432μs	76.6686 KOps/s	76.2556 KOps/s	$\color{#35bf28}+0.54\%$
test_unsqueeze	0.2003ms	93.6518μs	10.6778 KOps/s	10.8544 KOps/s	$\color{#d91a1a}-1.63\%$
test_split	0.3619ms	0.1988ms	5.0293 KOps/s	5.2740 KOps/s	$\color{#d91a1a}-4.64\%$
test_permute	0.3595ms	0.2206ms	4.5324 KOps/s	4.6006 KOps/s	$\color{#d91a1a}-1.48\%$
test_stack	45.7423ms	30.4320ms	32.8602 Ops/s	39.5462 Ops/s	$\textbf{\color{#d91a1a}-16.91\%}$
test_cat	30.8122ms	26.0593ms	38.3740 Ops/s	39.8274 Ops/s	$\color{#d91a1a}-3.65\%$

[github-actions]

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

Total Benchmarks: 228. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1176ms	13.7790μs	72.5744 KOps/s	72.2996 KOps/s	$\color{#35bf28}+0.38\%$
test_plain_set_stack_nested	43.0800μs	13.8870μs	72.0098 KOps/s	71.7719 KOps/s	$\color{#35bf28}+0.33\%$
test_plain_set_nested_inplace	49.8510μs	14.7769μs	67.6732 KOps/s	66.5799 KOps/s	$\color{#35bf28}+1.64\%$
test_plain_set_stack_nested_inplace	48.7200μs	14.7000μs	68.0273 KOps/s	67.4084 KOps/s	$\color{#35bf28}+0.92\%$
test_items	31.2100μs	2.8648μs	349.0703 KOps/s	345.3667 KOps/s	$\color{#35bf28}+1.07\%$
test_items_nested	0.3572ms	0.3221ms	3.1044 KOps/s	3.0548 KOps/s	$\color{#35bf28}+1.62\%$
test_items_nested_locked	0.3785ms	0.3242ms	3.0845 KOps/s	3.0603 KOps/s	$\color{#35bf28}+0.79\%$
test_items_nested_leaf	76.8710μs	55.5622μs	17.9978 KOps/s	17.9342 KOps/s	$\color{#35bf28}+0.35\%$
test_items_stack_nested	0.3582ms	0.3259ms	3.0681 KOps/s	3.0776 KOps/s	$\color{#d91a1a}-0.31\%$
test_items_stack_nested_leaf	81.9310μs	55.7591μs	17.9343 KOps/s	17.5254 KOps/s	$\color{#35bf28}+2.33\%$
test_items_stack_nested_locked	0.3755ms	0.3239ms	3.0874 KOps/s	3.0347 KOps/s	$\color{#35bf28}+1.74\%$
test_keys	27.0900μs	3.4275μs	291.7587 KOps/s	291.3376 KOps/s	$\color{#35bf28}+0.14\%$
test_keys_nested	85.9910μs	55.3590μs	18.0639 KOps/s	18.3757 KOps/s	$\color{#d91a1a}-1.70\%$
test_keys_nested_locked	0.8157ms	61.7944μs	16.1827 KOps/s	16.1765 KOps/s	$\color{#35bf28}+0.04\%$
test_keys_nested_leaf	77.0310μs	47.1619μs	21.2036 KOps/s	21.6342 KOps/s	$\color{#d91a1a}-1.99\%$
test_keys_stack_nested	99.4610μs	54.7361μs	18.2695 KOps/s	18.2189 KOps/s	$\color{#35bf28}+0.28\%$
test_keys_stack_nested_leaf	82.0410μs	47.1480μs	21.2098 KOps/s	20.8447 KOps/s	$\color{#35bf28}+1.75\%$
test_keys_stack_nested_locked	93.1910μs	61.1073μs	16.3647 KOps/s	16.2116 KOps/s	$\color{#35bf28}+0.94\%$
test_values	5.1585μs	0.8482μs	1.1790 MOps/s	1.1958 MOps/s	$\color{#d91a1a}-1.40\%$
test_values_nested	63.8810μs	40.8805μs	24.4615 KOps/s	24.5684 KOps/s	$\color{#d91a1a}-0.43\%$
test_values_nested_locked	90.9210μs	42.7700μs	23.3809 KOps/s	23.5395 KOps/s	$\color{#d91a1a}-0.67\%$
test_values_nested_leaf	66.4610μs	35.4402μs	28.2165 KOps/s	28.4427 KOps/s	$\color{#d91a1a}-0.80\%$
test_values_stack_nested	75.2010μs	40.8580μs	24.4750 KOps/s	23.9962 KOps/s	$\color{#35bf28}+2.00\%$
test_values_stack_nested_leaf	66.9710μs	35.7244μs	27.9920 KOps/s	27.5192 KOps/s	$\color{#35bf28}+1.72\%$
test_values_stack_nested_locked	74.8210μs	42.7722μs	23.3797 KOps/s	22.8676 KOps/s	$\color{#35bf28}+2.24\%$
test_membership	1.7265μs	0.5071μs	1.9720 MOps/s	1.9971 MOps/s	$\color{#d91a1a}-1.25\%$
test_membership_nested	17.8800μs	1.7987μs	555.9582 KOps/s	545.2477 KOps/s	$\color{#35bf28}+1.96\%$
test_membership_nested_leaf	16.5400μs	1.8322μs	545.7984 KOps/s	553.3762 KOps/s	$\color{#d91a1a}-1.37\%$
test_membership_stacked_nested	24.3310μs	1.8356μs	544.7760 KOps/s	529.8085 KOps/s	$\color{#35bf28}+2.83\%$
test_membership_stacked_nested_leaf	31.6310μs	1.8704μs	534.6340 KOps/s	533.3935 KOps/s	$\color{#35bf28}+0.23\%$
test_membership_nested_last	25.8410μs	2.7584μs	362.5239 KOps/s	366.0320 KOps/s	$\color{#d91a1a}-0.96\%$
test_membership_nested_leaf_last	32.3500μs	2.7947μs	357.8153 KOps/s	362.3194 KOps/s	$\color{#d91a1a}-1.24\%$
test_membership_stacked_nested_last	23.5500μs	2.7689μs	361.1538 KOps/s	367.5362 KOps/s	$\color{#d91a1a}-1.74\%$
test_membership_stacked_nested_leaf_last	19.4900μs	2.7190μs	367.7805 KOps/s	366.8846 KOps/s	$\color{#35bf28}+0.24\%$
test_nested_getleaf	34.1600μs	6.0528μs	165.2126 KOps/s	164.3478 KOps/s	$\color{#35bf28}+0.53\%$
test_nested_get	42.8410μs	5.7056μs	175.2660 KOps/s	176.2142 KOps/s	$\color{#d91a1a}-0.54\%$
test_stacked_getleaf	31.3310μs	6.1039μs	163.8296 KOps/s	164.3479 KOps/s	$\color{#d91a1a}-0.32\%$
test_stacked_get	31.7000μs	5.7007μs	175.4157 KOps/s	179.3316 KOps/s	$\color{#d91a1a}-2.18\%$
test_nested_getitemleaf	29.7710μs	6.0892μs	164.2255 KOps/s	166.0118 KOps/s	$\color{#d91a1a}-1.08\%$
test_nested_getitem	41.7400μs	5.7600μs	173.6122 KOps/s	176.4821 KOps/s	$\color{#d91a1a}-1.63\%$
test_stacked_getitemleaf	28.7400μs	6.0505μs	165.2768 KOps/s	166.2272 KOps/s	$\color{#d91a1a}-0.57\%$
test_stacked_getitem	33.4310μs	5.7168μs	174.9227 KOps/s	176.4234 KOps/s	$\color{#d91a1a}-0.85\%$
test_lock_nested	7.7919ms	0.4201ms	2.3803 KOps/s	2.3949 KOps/s	$\color{#d91a1a}-0.61\%$
test_lock_stack_nested	0.4727ms	0.3782ms	2.6441 KOps/s	2.6406 KOps/s	$\color{#35bf28}+0.13\%$
test_unlock_nested	0.7557ms	0.3519ms	2.8418 KOps/s	2.8476 KOps/s	$\color{#d91a1a}-0.21\%$
test_unlock_stack_nested	0.3927ms	0.3180ms	3.1448 KOps/s	3.1501 KOps/s	$\color{#d91a1a}-0.17\%$
test_flatten_speed	0.1452ms	68.1000μs	14.6843 KOps/s	14.5190 KOps/s	$\color{#35bf28}+1.14\%$
test_unflatten_speed	0.3343ms	0.2795ms	3.5773 KOps/s	3.5736 KOps/s	$\color{#35bf28}+0.11\%$
test_common_ops	1.5411ms	1.1970ms	835.4067 Ops/s	821.4684 Ops/s	$\color{#35bf28}+1.70\%$
test_creation	24.9300μs	1.4361μs	696.3242 KOps/s	692.6138 KOps/s	$\color{#35bf28}+0.54\%$
test_creation_empty	49.9200μs	14.8824μs	67.1935 KOps/s	66.0165 KOps/s	$\color{#35bf28}+1.78\%$
test_creation_nested_1	41.4610μs	16.8141μs	59.4740 KOps/s	59.3690 KOps/s	$\color{#35bf28}+0.18\%$
test_creation_nested_2	47.7300μs	19.3268μs	51.7417 KOps/s	50.6401 KOps/s	$\color{#35bf28}+2.18\%$
test_clone	59.7200μs	27.4162μs	36.4748 KOps/s	35.6328 KOps/s	$\color{#35bf28}+2.36\%$
test_getitem[int]	1.4211ms	15.4877μs	64.5672 KOps/s	63.1719 KOps/s	$\color{#35bf28}+2.21\%$
test_getitem[slice_int]	0.1176ms	26.4725μs	37.7750 KOps/s	37.2154 KOps/s	$\color{#35bf28}+1.50\%$
test_getitem[range]	0.2205ms	0.1061ms	9.4285 KOps/s	9.3029 KOps/s	$\color{#35bf28}+1.35\%$
test_getitem[tuple]	0.1174ms	22.7077μs	44.0379 KOps/s	43.7303 KOps/s	$\color{#35bf28}+0.70\%$
test_getitem[list]	0.1902ms	95.3700μs	10.4855 KOps/s	10.5459 KOps/s	$\color{#d91a1a}-0.57\%$
test_setitem_dim[int]	64.6200μs	42.3337μs	23.6218 KOps/s	23.2940 KOps/s	$\color{#35bf28}+1.41\%$
test_setitem_dim[slice_int]	0.1098ms	63.3431μs	15.7870 KOps/s	15.6827 KOps/s	$\color{#35bf28}+0.67\%$
test_setitem_dim[range]	0.1494ms	0.1218ms	8.2094 KOps/s	8.1837 KOps/s	$\color{#35bf28}+0.31\%$
test_setitem_dim[tuple]	83.9210μs	57.5588μs	17.3735 KOps/s	16.2956 KOps/s	$\textbf{\color{#35bf28}+6.61\%}$
test_setitem	70.3110μs	39.9180μs	25.0513 KOps/s	23.1617 KOps/s	$\textbf{\color{#35bf28}+8.16\%}$
test_set	81.4910μs	38.9541μs	25.6712 KOps/s	25.0018 KOps/s	$\color{#35bf28}+2.68\%$
test_set_shared	0.3608ms	48.8086μs	20.4882 KOps/s	20.2925 KOps/s	$\color{#35bf28}+0.96\%$
test_update	0.2522ms	47.1652μs	21.2021 KOps/s	20.6528 KOps/s	$\color{#35bf28}+2.66\%$
test_update_nested	0.1165ms	53.7761μs	18.5956 KOps/s	17.2912 KOps/s	$\textbf{\color{#35bf28}+7.54\%}$
test_update__nested	98.2510μs	56.3069μs	17.7598 KOps/s	16.2919 KOps/s	$\textbf{\color{#35bf28}+9.01\%}$
test_set_nested	78.4410μs	40.7538μs	24.5376 KOps/s	23.5386 KOps/s	$\color{#35bf28}+4.24\%$
test_set_nested_new	94.7010μs	44.8013μs	22.3208 KOps/s	21.6472 KOps/s	$\color{#35bf28}+3.11\%$
test_select	0.1184ms	57.4844μs	17.3960 KOps/s	15.9761 KOps/s	$\textbf{\color{#35bf28}+8.89\%}$
test_select_nested	66.8010μs	41.3196μs	24.2016 KOps/s	24.0380 KOps/s	$\color{#35bf28}+0.68\%$
test_exclude_nested	82.4610μs	57.7623μs	17.3123 KOps/s	17.5050 KOps/s	$\color{#d91a1a}-1.10\%$
test_empty[True]	0.2977ms	0.2380ms	4.2019 KOps/s	4.1268 KOps/s	$\color{#35bf28}+1.82\%$
test_empty[False]	2.9680μs	0.7392μs	1.3529 MOps/s	1.3651 MOps/s	$\color{#d91a1a}-0.89\%$
test_to	47.6200μs	24.2129μs	41.3003 KOps/s	40.2506 KOps/s	$\color{#35bf28}+2.61\%$
test_to_nonblocking	58.8010μs	23.8795μs	41.8769 KOps/s	41.2012 KOps/s	$\color{#35bf28}+1.64\%$
test_unbind_speed	0.3265ms	0.2732ms	3.6608 KOps/s	3.6524 KOps/s	$\color{#35bf28}+0.23\%$
test_unbind_speed_stack0	0.4426ms	0.2708ms	3.6929 KOps/s	3.7284 KOps/s	$\color{#d91a1a}-0.95\%$
test_unbind_speed_stack1	92.6469ms	0.7065ms	1.4154 KOps/s	1.4380 KOps/s	$\color{#d91a1a}-1.57\%$
test_split	94.9279ms	2.1279ms	469.9515 Ops/s	467.8443 Ops/s	$\color{#35bf28}+0.45\%$
test_chunk	96.8783ms	2.1358ms	468.2064 Ops/s	468.5588 Ops/s	$\color{#d91a1a}-0.08\%$
test_creation[device0]	0.3531ms	0.1240ms	8.0624 KOps/s	8.0406 KOps/s	$\color{#35bf28}+0.27\%$
test_creation_from_tensor	0.4203ms	0.1265ms	7.9076 KOps/s	7.8665 KOps/s	$\color{#35bf28}+0.52\%$
test_add_one[memmap_tensor0]	0.2879ms	8.3054μs	120.4033 KOps/s	120.4991 KOps/s	$\color{#d91a1a}-0.08\%$
test_contiguous[memmap_tensor0]	32.9800μs	2.1045μs	475.1701 KOps/s	482.0211 KOps/s	$\color{#d91a1a}-1.42\%$
test_stack[memmap_tensor0]	33.1700μs	6.5790μs	151.9991 KOps/s	152.6706 KOps/s	$\color{#d91a1a}-0.44\%$
test_memmaptd_index	1.1003ms	0.4137ms	2.4172 KOps/s	2.4585 KOps/s	$\color{#d91a1a}-1.68\%$
test_memmaptd_index_astensor	0.7328ms	0.4710ms	2.1231 KOps/s	2.1556 KOps/s	$\color{#d91a1a}-1.51\%$
test_memmaptd_index_op	1.3586ms	0.9866ms	1.0136 KOps/s	1.0107 KOps/s	$\color{#35bf28}+0.29\%$
test_serialize_model	0.1309s	0.1301s	7.6862 Ops/s	7.6817 Ops/s	$\color{#35bf28}+0.06\%$
test_serialize_model_pickle	1.3509s	1.2130s	0.8244 Ops/s	0.8245 Ops/s	$\color{#d91a1a}-0.01\%$
test_serialize_weights	0.2242s	0.1428s	7.0031 Ops/s	7.6650 Ops/s	$\textbf{\color{#d91a1a}-8.63\%}$
test_serialize_weights_returnearly	0.2241s	55.6103ms	17.9823 Ops/s	18.3614 Ops/s	$\color{#d91a1a}-2.06\%$
test_serialize_weights_pickle	1.3467s	1.2161s	0.8223 Ops/s	0.8184 Ops/s	$\color{#35bf28}+0.48\%$
test_reshape_pytree	82.7710μs	34.5255μs	28.9641 KOps/s	28.9173 KOps/s	$\color{#35bf28}+0.16\%$
test_reshape_td	74.6510μs	41.2811μs	24.2241 KOps/s	24.8497 KOps/s	$\color{#d91a1a}-2.52\%$
test_view_pytree	61.8910μs	34.8783μs	28.6711 KOps/s	28.8789 KOps/s	$\color{#d91a1a}-0.72\%$
test_view_td	86.0610μs	47.5371μs	21.0362 KOps/s	21.7438 KOps/s	$\color{#d91a1a}-3.25\%$
test_unbind_pytree	74.5500μs	33.6865μs	29.6855 KOps/s	29.8926 KOps/s	$\color{#d91a1a}-0.69\%$
test_unbind_td	0.4912ms	42.1444μs	23.7279 KOps/s	23.5985 KOps/s	$\color{#35bf28}+0.55\%$
test_split_pytree	81.1800μs	44.6298μs	22.4066 KOps/s	21.7930 KOps/s	$\color{#35bf28}+2.82\%$
test_split_td	0.6557ms	55.2846μs	18.0882 KOps/s	15.7018 KOps/s	$\textbf{\color{#35bf28}+15.20\%}$
test_add_pytree	0.1048ms	54.4377μs	18.3696 KOps/s	18.7224 KOps/s	$\color{#d91a1a}-1.88\%$
test_add_td	0.1327ms	87.7166μs	11.4004 KOps/s	11.4209 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_add_one_nested[tensordict-compile]	0.4047ms	0.2051ms	4.8748 KOps/s	4.9030 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_add_one_nested[tensordict-eager]	0.2184ms	0.1468ms	6.8125 KOps/s	6.6814 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_add_one_nested[pytree-compile]	0.1969ms	0.1412ms	7.0842 KOps/s	7.1989 KOps/s	$\color{#d91a1a}-1.59\%$
test_compile_add_one_nested[pytree-eager]	0.2327ms	0.1771ms	5.6472 KOps/s	5.6750 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_copy_nested[tensordict-compile]	57.9310μs	21.1983μs	47.1735 KOps/s	48.0951 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_copy_nested[tensordict-eager]	0.1744ms	42.8284μs	23.3490 KOps/s	22.8150 KOps/s	$\color{#35bf28}+2.34\%$
test_compile_copy_nested[pytree-compile]	0.2173ms	64.1601μs	15.5860 KOps/s	15.4849 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_copy_nested[pytree-eager]	87.9810μs	48.9435μs	20.4317 KOps/s	20.3726 KOps/s	$\color{#35bf28}+0.29\%$
test_compile_add_one_flat[tensordict-compile]	0.4059ms	0.3089ms	3.2370 KOps/s	3.2476 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_add_one_flat[tensordict-eager]	0.2594ms	0.2062ms	4.8500 KOps/s	4.8785 KOps/s	$\color{#d91a1a}-0.59\%$
test_compile_add_one_flat[tensorclass-compile]	0.1714ms	0.1269ms	7.8812 KOps/s	8.0870 KOps/s	$\color{#d91a1a}-2.54\%$
test_compile_add_one_flat[tensorclass-eager]	0.1044ms	61.3613μs	16.2969 KOps/s	17.0373 KOps/s	$\color{#d91a1a}-4.35\%$
test_compile_add_one_flat[pytree-compile]	0.3902ms	0.3064ms	3.2638 KOps/s	3.2870 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_add_one_flat[pytree-eager]	0.7105ms	0.6212ms	1.6099 KOps/s	1.6935 KOps/s	$\color{#d91a1a}-4.94\%$
test_compile_add_self_flat[tensordict-eager]	0.3027ms	0.2455ms	4.0739 KOps/s	4.0522 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_add_self_flat[tensordict-compile]	0.3490ms	0.3099ms	3.2271 KOps/s	3.2541 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_self_flat[tensorclass-eager]	0.1169ms	71.8140μs	13.9249 KOps/s	14.4556 KOps/s	$\color{#d91a1a}-3.67\%$
test_compile_add_self_flat[tensorclass-compile]	0.1852ms	0.1320ms	7.5730 KOps/s	8.0080 KOps/s	$\textbf{\color{#d91a1a}-5.43\%}$
test_compile_add_self_flat[pytree-eager]	0.6793ms	0.5211ms	1.9192 KOps/s	1.9306 KOps/s	$\color{#d91a1a}-0.59\%$
test_compile_add_self_flat[pytree-compile]	0.3564ms	0.3084ms	3.2426 KOps/s	3.2599 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_copy_flat[tensordict-compile]	0.1046ms	18.7146μs	53.4341 KOps/s	55.6178 KOps/s	$\color{#d91a1a}-3.93\%$
test_compile_copy_flat[tensordict-eager]	0.1865ms	26.1437μs	38.2501 KOps/s	35.9410 KOps/s	$\textbf{\color{#35bf28}+6.42\%}$
test_compile_copy_flat[pytree-compile]	0.1133ms	70.1830μs	14.2485 KOps/s	14.1584 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_copy_flat[pytree-eager]	82.9410μs	51.1855μs	19.5368 KOps/s	19.2930 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_assign_and_add[tensordict-compile]	2.2993ms	0.8044ms	1.2431 KOps/s	1.1633 KOps/s	$\textbf{\color{#35bf28}+6.86\%}$
test_compile_assign_and_add[tensordict-eager]	3.4213ms	3.2001ms	312.4867 Ops/s	312.6084 Ops/s	$\color{#d91a1a}-0.04\%$
test_compile_assign_and_add[pytree-compile]	2.2182ms	0.7893ms	1.2669 KOps/s	1.1679 KOps/s	$\textbf{\color{#35bf28}+8.48\%}$
test_compile_assign_and_add[pytree-eager]	3.1431ms	3.0944ms	323.1673 Ops/s	323.2633 Ops/s	$\color{#d91a1a}-0.03\%$
test_compile_indexing[tensor-tensordict-compile]	0.1774ms	0.1061ms	9.4211 KOps/s	9.5113 KOps/s	$\color{#d91a1a}-0.95\%$
test_compile_indexing[tensor-tensordict-eager]	0.1878ms	56.9767μs	17.5510 KOps/s	16.7283 KOps/s	$\color{#35bf28}+4.92\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1731ms	0.1043ms	9.5920 KOps/s	10.0123 KOps/s	$\color{#d91a1a}-4.20\%$
test_compile_indexing[tensor-tensorclass-eager]	85.3410μs	43.1984μs	23.1490 KOps/s	24.6009 KOps/s	$\textbf{\color{#d91a1a}-5.90\%}$
test_compile_indexing[tensor-pytree-compile]	0.1623ms	0.1056ms	9.4678 KOps/s	9.9411 KOps/s	$\color{#d91a1a}-4.76\%$
test_compile_indexing[tensor-pytree-eager]	89.4410μs	42.9741μs	23.2698 KOps/s	24.4453 KOps/s	$\color{#d91a1a}-4.81\%$
test_compile_indexing[slice-tensordict-compile]	0.1749ms	0.1344ms	7.4396 KOps/s	7.5289 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_indexing[slice-tensordict-eager]	0.1630ms	24.6604μs	40.5508 KOps/s	40.6209 KOps/s	$\color{#d91a1a}-0.17\%$
test_compile_indexing[slice-tensorclass-compile]	0.1663ms	0.1270ms	7.8729 KOps/s	7.9254 KOps/s	$\color{#d91a1a}-0.66\%$
test_compile_indexing[slice-tensorclass-eager]	54.9100μs	20.5070μs	48.7639 KOps/s	49.6705 KOps/s	$\color{#d91a1a}-1.83\%$
test_compile_indexing[slice-pytree-compile]	0.1761ms	0.1337ms	7.4799 KOps/s	7.9184 KOps/s	$\textbf{\color{#d91a1a}-5.54\%}$
test_compile_indexing[slice-pytree-eager]	56.7610μs	20.1896μs	49.5304 KOps/s	49.4215 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_indexing[int-tensordict-compile]	0.2149ms	0.1409ms	7.0977 KOps/s	7.5186 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_compile_indexing[int-tensordict-eager]	0.5268ms	24.7900μs	40.3388 KOps/s	40.9161 KOps/s	$\color{#d91a1a}-1.41\%$
test_compile_indexing[int-tensorclass-compile]	0.1715ms	0.1285ms	7.7834 KOps/s	7.6482 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_indexing[int-tensorclass-eager]	54.3810μs	19.8653μs	50.3390 KOps/s	48.9264 KOps/s	$\color{#35bf28}+2.89\%$
test_compile_indexing[int-pytree-compile]	0.1802ms	0.1282ms	7.7977 KOps/s	7.7242 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_indexing[int-pytree-eager]	62.3200μs	20.4527μs	48.8933 KOps/s	48.7962 KOps/s	$\color{#35bf28}+0.20\%$
test_mod_add[eager]	75.9710μs	30.8925μs	32.3704 KOps/s	32.3568 KOps/s	$\color{#35bf28}+0.04\%$
test_mod_add[compile]	0.1788ms	70.4368μs	14.1971 KOps/s	14.2779 KOps/s	$\color{#d91a1a}-0.57\%$
test_mod_add[compile-overhead]	0.2696ms	0.1344ms	7.4385 KOps/s	7.1644 KOps/s	$\color{#35bf28}+3.83\%$
test_mod_wrap[eager]	0.3634ms	0.2399ms	4.1691 KOps/s	4.2885 KOps/s	$\color{#d91a1a}-2.78\%$
test_mod_wrap[compile]	1.4441ms	0.2900ms	3.4478 KOps/s	3.4338 KOps/s	$\color{#35bf28}+0.41\%$
test_mod_wrap[compile-overhead]	7.6253ms	4.1141ms	243.0688 Ops/s	246.5231 Ops/s	$\color{#d91a1a}-1.40\%$
test_mod_wrap_and_backward[eager]	1.7021ms	1.2967ms	771.1804 Ops/s	721.8503 Ops/s	$\textbf{\color{#35bf28}+6.83\%}$
test_mod_wrap_and_backward[compile]	1.5036ms	1.2793ms	781.6865 Ops/s	712.8300 Ops/s	$\textbf{\color{#35bf28}+9.66\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3274ms	0.8859ms	1.1288 KOps/s	1.0144 KOps/s	$\textbf{\color{#35bf28}+11.28\%}$
test_seq_add[eager]	0.1540ms	97.1850μs	10.2897 KOps/s	10.2427 KOps/s	$\color{#35bf28}+0.46\%$
test_seq_add[compile]	0.1506ms	81.8419μs	12.2187 KOps/s	12.1808 KOps/s	$\color{#35bf28}+0.31\%$
test_seq_add[compile-overhead]	0.1879ms	0.1161ms	8.6167 KOps/s	8.9553 KOps/s	$\color{#d91a1a}-3.78\%$
test_seq_wrap[eager]	0.4607ms	0.3826ms	2.6136 KOps/s	2.6713 KOps/s	$\color{#d91a1a}-2.16\%$
test_seq_wrap[compile]	0.3695ms	0.3119ms	3.2064 KOps/s	3.1023 KOps/s	$\color{#35bf28}+3.35\%$
test_seq_wrap[compile-overhead]	0.2670ms	0.2153ms	4.6452 KOps/s	4.6130 KOps/s	$\color{#35bf28}+0.70\%$
test_func_call_runtime[False-eager]	0.8561ms	0.7395ms	1.3522 KOps/s	1.4000 KOps/s	$\color{#d91a1a}-3.41\%$
test_func_call_runtime[False-compile]	1.0381ms	0.7553ms	1.3240 KOps/s	1.2827 KOps/s	$\color{#35bf28}+3.22\%$
test_func_call_runtime[False-compile-overhead]	0.4159ms	0.3499ms	2.8579 KOps/s	2.8591 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_runtime[True-eager]	1.0051ms	0.8683ms	1.1517 KOps/s	1.1420 KOps/s	$\color{#35bf28}+0.86\%$
test_func_call_runtime[True-compile]	0.8974ms	0.7825ms	1.2779 KOps/s	1.2638 KOps/s	$\color{#35bf28}+1.12\%$
test_func_call_runtime[True-compile-overhead]	0.4390ms	0.3728ms	2.6824 KOps/s	2.7001 KOps/s	$\color{#d91a1a}-0.65\%$
test_func_call_cm_runtime[False-eager]	0.7994ms	0.7247ms	1.3799 KOps/s	1.4163 KOps/s	$\color{#d91a1a}-2.57\%$
test_func_call_cm_runtime[False-compile]	0.8580ms	0.7660ms	1.3055 KOps/s	1.2961 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4900ms	0.3521ms	2.8404 KOps/s	2.8583 KOps/s	$\color{#d91a1a}-0.62\%$
test_func_call_cm_runtime[True-eager]	1.0536ms	0.9582ms	1.0436 KOps/s	1.0262 KOps/s	$\color{#35bf28}+1.69\%$
test_func_call_cm_runtime[True-compile]	0.8610ms	0.8100ms	1.2346 KOps/s	1.2165 KOps/s	$\color{#35bf28}+1.49\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4840ms	0.3942ms	2.5368 KOps/s	2.5365 KOps/s	$\color{#35bf28}+0.01\%$
test_vmap_func_call_cm_runtime[eager]	2.4322ms	1.9708ms	507.4013 Ops/s	505.1941 Ops/s	$\color{#35bf28}+0.44\%$
test_vmap_func_call_cm_runtime[compile]	0.9930ms	0.8197ms	1.2199 KOps/s	1.1744 KOps/s	$\color{#35bf28}+3.87\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4469ms	0.3989ms	2.5066 KOps/s	2.5181 KOps/s	$\color{#d91a1a}-0.46\%$
test_distributed	3.0114ms	0.1804ms	5.5434 KOps/s	8.8804 KOps/s	$\textbf{\color{#d91a1a}-37.58\%}$
test_tdmodule	0.4129ms	15.0639μs	66.3840 KOps/s	66.6909 KOps/s	$\color{#d91a1a}-0.46\%$
test_tdmodule_dispatch	56.0510μs	28.2576μs	35.3887 KOps/s	34.3009 KOps/s	$\color{#35bf28}+3.17\%$
test_tdseq	37.6310μs	15.5780μs	64.1932 KOps/s	64.4216 KOps/s	$\color{#d91a1a}-0.35\%$
test_tdseq_dispatch	52.3700μs	30.9920μs	32.2664 KOps/s	31.9886 KOps/s	$\color{#35bf28}+0.87\%$
test_instantiation_functorch	1.9668ms	1.8083ms	552.9976 Ops/s	550.2546 Ops/s	$\color{#35bf28}+0.50\%$
test_instantiation_td	1.7809ms	1.1649ms	858.4295 Ops/s	854.0178 Ops/s	$\color{#35bf28}+0.52\%$
test_exec_functorch	0.3371ms	0.2049ms	4.8803 KOps/s	4.8751 KOps/s	$\color{#35bf28}+0.11\%$
test_exec_functional_call	0.2698ms	0.2023ms	4.9434 KOps/s	4.8975 KOps/s	$\color{#35bf28}+0.94\%$
test_exec_td	0.3091ms	0.2060ms	4.8551 KOps/s	4.8361 KOps/s	$\color{#35bf28}+0.39\%$
test_exec_td_decorator	1.1083ms	0.2463ms	4.0608 KOps/s	3.9821 KOps/s	$\color{#35bf28}+1.98\%$
test_vmap_mlp_speed[True-True]	0.7563ms	0.6578ms	1.5201 KOps/s	1.4961 KOps/s	$\color{#35bf28}+1.61\%$
test_vmap_mlp_speed[True-False]	0.7183ms	0.6555ms	1.5257 KOps/s	1.5208 KOps/s	$\color{#35bf28}+0.32\%$
test_vmap_mlp_speed[False-True]	0.7190ms	0.5499ms	1.8184 KOps/s	1.8187 KOps/s	$\color{#d91a1a}-0.02\%$
test_vmap_mlp_speed[False-False]	0.6105ms	0.5473ms	1.8272 KOps/s	1.8085 KOps/s	$\color{#35bf28}+1.04\%$
test_vmap_mlp_speed_decorator[True-True]	1.3675ms	0.6462ms	1.5474 KOps/s	1.5577 KOps/s	$\color{#d91a1a}-0.66\%$
test_vmap_mlp_speed_decorator[True-False]	0.7662ms	0.6438ms	1.5533 KOps/s	1.5489 KOps/s	$\color{#35bf28}+0.29\%$
test_vmap_mlp_speed_decorator[False-True]	0.6847ms	0.5629ms	1.7764 KOps/s	1.7755 KOps/s	$\color{#35bf28}+0.05\%$
test_vmap_mlp_speed_decorator[False-False]	0.6846ms	0.5664ms	1.7657 KOps/s	1.7206 KOps/s	$\color{#35bf28}+2.62\%$
test_vmap_transformer_speed[True-True]	8.7600ms	7.9734ms	125.4171 Ops/s	124.4053 Ops/s	$\color{#35bf28}+0.81\%$
test_vmap_transformer_speed[True-False]	8.1053ms	7.9390ms	125.9603 Ops/s	124.7698 Ops/s	$\color{#35bf28}+0.95\%$
test_vmap_transformer_speed[False-True]	7.9270ms	7.7614ms	128.8425 Ops/s	126.7856 Ops/s	$\color{#35bf28}+1.62\%$
test_vmap_transformer_speed[False-False]	8.1094ms	7.7860ms	128.4351 Ops/s	127.5118 Ops/s	$\color{#35bf28}+0.72\%$
test_vmap_transformer_speed_decorator[True-True]	19.5412ms	18.6590ms	53.5935 Ops/s	53.8603 Ops/s	$\color{#d91a1a}-0.50\%$
test_vmap_transformer_speed_decorator[True-False]	19.1354ms	18.7188ms	53.4224 Ops/s	53.5788 Ops/s	$\color{#d91a1a}-0.29\%$
test_vmap_transformer_speed_decorator[False-True]	18.6280ms	18.5589ms	53.8824 Ops/s	54.1023 Ops/s	$\color{#d91a1a}-0.41\%$
test_vmap_transformer_speed_decorator[False-False]	19.8194ms	18.5835ms	53.8113 Ops/s	54.0899 Ops/s	$\color{#d91a1a}-0.52\%$
test_to_module_speed[True]	1.9749ms	0.9313ms	1.0738 KOps/s	1.0691 KOps/s	$\color{#35bf28}+0.44\%$
test_to_module_speed[False]	0.9841ms	0.8997ms	1.1115 KOps/s	1.0860 KOps/s	$\color{#35bf28}+2.35\%$
test_tc_init	61.0710μs	34.5744μs	28.9231 KOps/s	29.3490 KOps/s	$\color{#d91a1a}-1.45\%$
test_tc_init_nested	0.1102ms	71.6088μs	13.9648 KOps/s	14.8388 KOps/s	$\textbf{\color{#d91a1a}-5.89\%}$
test_tc_first_layer_tensor	16.0401μs	0.6635μs	1.5072 MOps/s	1.5031 MOps/s	$\color{#35bf28}+0.27\%$
test_tc_first_layer_nontensor	30.1900μs	2.2020μs	454.1286 KOps/s	453.3373 KOps/s	$\color{#35bf28}+0.17\%$
test_tc_second_layer_tensor	8.7575μs	1.3379μs	747.4442 KOps/s	740.5515 KOps/s	$\color{#35bf28}+0.93\%$
test_tc_second_layer_nontensor	33.1900μs	2.8723μs	348.1505 KOps/s	345.4798 KOps/s	$\color{#35bf28}+0.77\%$
test_unbind	0.1961s	10.8551ms	92.1223 Ops/s	94.1153 Ops/s	$\color{#d91a1a}-2.12\%$
test_full_like	0.6404ms	0.5762ms	1.7356 KOps/s	1.7382 KOps/s	$\color{#d91a1a}-0.15\%$
test_zeros_like	0.2620ms	0.1980ms	5.0495 KOps/s	5.0518 KOps/s	$\color{#d91a1a}-0.04\%$
test_ones_like	0.2334ms	0.1979ms	5.0535 KOps/s	5.0559 KOps/s	$\color{#d91a1a}-0.05\%$
test_clone	0.4411ms	0.4145ms	2.4126 KOps/s	2.4189 KOps/s	$\color{#d91a1a}-0.26\%$
test_squeeze	41.3600μs	9.7764μs	102.2871 KOps/s	100.8873 KOps/s	$\color{#35bf28}+1.39\%$
test_unsqueeze	0.2320ms	74.4689μs	13.4284 KOps/s	13.5519 KOps/s	$\color{#d91a1a}-0.91\%$
test_split	0.4442ms	0.1571ms	6.3662 KOps/s	6.2998 KOps/s	$\color{#35bf28}+1.05\%$
test_permute	0.2936ms	0.1767ms	5.6578 KOps/s	5.7232 KOps/s	$\color{#d91a1a}-1.14\%$
test_stack	1.2537ms	0.8564ms	1.1676 KOps/s	1.1573 KOps/s	$\color{#35bf28}+0.89\%$
test_cat	1.2548ms	1.2315ms	812.0121 Ops/s	811.6978 Ops/s	$\color{#35bf28}+0.04\%$