[Feature] More copy() refactors for compile friendliness

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	39.4110μs	14.2917μs	69.9706 KOps/s	67.8649 KOps/s	$\color{#35bf28}+3.10\%$
test_plain_set_stack_nested	48.3610μs	14.6457μs	68.2792 KOps/s	66.7027 KOps/s	$\color{#35bf28}+2.36\%$
test_plain_set_nested_inplace	45.2610μs	16.0877μs	62.1594 KOps/s	60.5927 KOps/s	$\color{#35bf28}+2.59\%$
test_plain_set_stack_nested_inplace	61.1720μs	16.2388μs	61.5808 KOps/s	61.5126 KOps/s	$\color{#35bf28}+0.11\%$
test_items	29.1500μs	5.5321μs	180.7628 KOps/s	176.3758 KOps/s	$\color{#35bf28}+2.49\%$
test_items_nested	0.5658ms	0.5124ms	1.9518 KOps/s	1.9371 KOps/s	$\color{#35bf28}+0.75\%$
test_items_nested_locked	0.5809ms	0.5150ms	1.9416 KOps/s	1.9482 KOps/s	$\color{#d91a1a}-0.34\%$
test_items_nested_leaf	0.1270ms	90.9810μs	10.9913 KOps/s	11.1311 KOps/s	$\color{#d91a1a}-1.26\%$
test_items_stack_nested	0.5759ms	0.5111ms	1.9567 KOps/s	1.9577 KOps/s	$\color{#d91a1a}-0.06\%$
test_items_stack_nested_leaf	0.1370ms	90.5914μs	11.0386 KOps/s	10.8483 KOps/s	$\color{#35bf28}+1.75\%$
test_items_stack_nested_locked	0.6200ms	0.5153ms	1.9405 KOps/s	1.9619 KOps/s	$\color{#d91a1a}-1.09\%$
test_keys	29.6300μs	4.4454μs	224.9533 KOps/s	243.4742 KOps/s	$\textbf{\color{#d91a1a}-7.61\%}$
test_keys_nested	0.1896ms	0.1162ms	8.6060 KOps/s	8.3833 KOps/s	$\color{#35bf28}+2.66\%$
test_keys_nested_locked	0.7596ms	0.1248ms	8.0133 KOps/s	7.7883 KOps/s	$\color{#35bf28}+2.89\%$
test_keys_nested_leaf	0.5159ms	0.1065ms	9.3896 KOps/s	9.1036 KOps/s	$\color{#35bf28}+3.14\%$
test_keys_stack_nested	0.1681ms	0.1165ms	8.5811 KOps/s	8.4228 KOps/s	$\color{#35bf28}+1.88\%$
test_keys_stack_nested_leaf	0.1530ms	0.1069ms	9.3514 KOps/s	9.1281 KOps/s	$\color{#35bf28}+2.45\%$
test_keys_stack_nested_locked	0.1695ms	0.1250ms	7.9999 KOps/s	7.8489 KOps/s	$\color{#35bf28}+1.92\%$
test_values	9.9642μs	0.9949μs	1.0052 MOps/s	998.2798 KOps/s	$\color{#35bf28}+0.69\%$
test_values_nested	74.8810μs	46.1557μs	21.6658 KOps/s	21.3263 KOps/s	$\color{#35bf28}+1.59\%$
test_values_nested_locked	86.5410μs	48.9560μs	20.4265 KOps/s	20.1557 KOps/s	$\color{#35bf28}+1.34\%$
test_values_nested_leaf	82.3820μs	52.6010μs	19.0110 KOps/s	18.7264 KOps/s	$\color{#35bf28}+1.52\%$
test_values_stack_nested	77.4710μs	46.7130μs	21.4073 KOps/s	21.3093 KOps/s	$\color{#35bf28}+0.46\%$
test_values_stack_nested_leaf	87.5510μs	52.5056μs	19.0456 KOps/s	18.4974 KOps/s	$\color{#35bf28}+2.96\%$
test_values_stack_nested_locked	81.5920μs	49.0837μs	20.3734 KOps/s	20.1358 KOps/s	$\color{#35bf28}+1.18\%$
test_membership	5.1768μs	0.8021μs	1.2468 MOps/s	1.2394 MOps/s	$\color{#35bf28}+0.60\%$
test_membership_nested	35.3400μs	2.9597μs	337.8666 KOps/s	332.9217 KOps/s	$\color{#35bf28}+1.49\%$
test_membership_nested_leaf	31.4110μs	3.0059μs	332.6784 KOps/s	331.8123 KOps/s	$\color{#35bf28}+0.26\%$
test_membership_stacked_nested	36.8700μs	3.0080μs	332.4422 KOps/s	332.6410 KOps/s	$\color{#d91a1a}-0.06\%$
test_membership_stacked_nested_leaf	24.5010μs	2.9802μs	335.5514 KOps/s	334.5166 KOps/s	$\color{#35bf28}+0.31\%$
test_membership_nested_last	38.2410μs	4.3628μs	229.2130 KOps/s	227.2465 KOps/s	$\color{#35bf28}+0.87\%$
test_membership_nested_leaf_last	34.9810μs	4.3499μs	229.8906 KOps/s	226.1016 KOps/s	$\color{#35bf28}+1.68\%$
test_membership_stacked_nested_last	38.6110μs	4.3729μs	228.6810 KOps/s	226.7032 KOps/s	$\color{#35bf28}+0.87\%$
test_membership_stacked_nested_leaf_last	25.7910μs	4.3633μs	229.1837 KOps/s	230.2510 KOps/s	$\color{#d91a1a}-0.46\%$
test_nested_getleaf	51.8310μs	20.8263μs	48.0161 KOps/s	48.5970 KOps/s	$\color{#d91a1a}-1.20\%$
test_nested_get	56.9710μs	19.7159μs	50.7206 KOps/s	51.1790 KOps/s	$\color{#d91a1a}-0.90\%$
test_stacked_getleaf	44.2310μs	20.5867μs	48.5751 KOps/s	48.5586 KOps/s	$\color{#35bf28}+0.03\%$
test_stacked_get	45.1410μs	19.4989μs	51.2848 KOps/s	50.9654 KOps/s	$\color{#35bf28}+0.63\%$
test_nested_getitemleaf	58.7610μs	21.3793μs	46.7743 KOps/s	47.1397 KOps/s	$\color{#d91a1a}-0.78\%$
test_nested_getitem	49.0010μs	20.0216μs	49.9460 KOps/s	49.6388 KOps/s	$\color{#35bf28}+0.62\%$
test_stacked_getitemleaf	47.4710μs	21.0812μs	47.4356 KOps/s	46.9325 KOps/s	$\color{#35bf28}+1.07\%$
test_stacked_getitem	56.8410μs	19.6923μs	50.7813 KOps/s	49.4489 KOps/s	$\color{#35bf28}+2.69\%$
test_lock_nested	7.9938ms	0.4566ms	2.1901 KOps/s	2.2271 KOps/s	$\color{#d91a1a}-1.66\%$
test_lock_stack_nested	0.5013ms	0.4536ms	2.2047 KOps/s	2.1888 KOps/s	$\color{#35bf28}+0.73\%$
test_unlock_nested	0.6410ms	0.3625ms	2.7590 KOps/s	2.7326 KOps/s	$\color{#35bf28}+0.96\%$
test_unlock_stack_nested	0.4167ms	0.3618ms	2.7636 KOps/s	2.7195 KOps/s	$\color{#35bf28}+1.62\%$
test_flatten_speed	0.1483ms	0.1160ms	8.6198 KOps/s	8.5904 KOps/s	$\color{#35bf28}+0.34\%$
test_unflatten_speed	0.7009ms	0.5665ms	1.7651 KOps/s	1.7523 KOps/s	$\color{#35bf28}+0.73\%$
test_common_ops	0.8464ms	0.7255ms	1.3783 KOps/s	1.3671 KOps/s	$\color{#35bf28}+0.82\%$
test_creation	0.1017ms	2.5760μs	388.2019 KOps/s	387.7784 KOps/s	$\color{#35bf28}+0.11\%$
test_creation_empty	33.4200μs	8.6166μs	116.0546 KOps/s	115.7373 KOps/s	$\color{#35bf28}+0.27\%$
test_creation_nested_1	34.9910μs	11.3324μs	88.2428 KOps/s	87.7010 KOps/s	$\color{#35bf28}+0.62\%$
test_creation_nested_2	46.0500μs	15.2289μs	65.6646 KOps/s	65.4914 KOps/s	$\color{#35bf28}+0.26\%$
test_clone	51.8610μs	12.8003μs	78.1234 KOps/s	76.6450 KOps/s	$\color{#35bf28}+1.93\%$
test_getitem[int]	1.1603ms	13.4946μs	74.1037 KOps/s	73.2298 KOps/s	$\color{#35bf28}+1.19\%$
test_getitem[slice_int]	0.1357ms	23.4165μs	42.7049 KOps/s	42.6213 KOps/s	$\color{#35bf28}+0.20\%$
test_getitem[range]	0.1659ms	56.9163μs	17.5697 KOps/s	17.0981 KOps/s	$\color{#35bf28}+2.76\%$
test_getitem[tuple]	0.1571ms	23.4405μs	42.6612 KOps/s	42.9003 KOps/s	$\color{#d91a1a}-0.56\%$
test_getitem[list]	0.1730ms	52.3349μs	19.1077 KOps/s	19.1545 KOps/s	$\color{#d91a1a}-0.24\%$
test_setitem_dim[int]	48.7010μs	23.9593μs	41.7374 KOps/s	41.5266 KOps/s	$\color{#35bf28}+0.51\%$
test_setitem_dim[slice_int]	80.9020μs	43.8717μs	22.7937 KOps/s	23.0591 KOps/s	$\color{#d91a1a}-1.15\%$
test_setitem_dim[range]	0.1227ms	83.8834μs	11.9213 KOps/s	11.8525 KOps/s	$\color{#35bf28}+0.58\%$
test_setitem_dim[tuple]	77.4810μs	40.3408μs	24.7888 KOps/s	24.6977 KOps/s	$\color{#35bf28}+0.37\%$
test_setitem	55.8010μs	17.4751μs	57.2242 KOps/s	56.2016 KOps/s	$\color{#35bf28}+1.82\%$
test_set	50.6810μs	16.7478μs	59.7095 KOps/s	59.8043 KOps/s	$\color{#d91a1a}-0.16\%$
test_set_shared	0.4878ms	0.2042ms	4.8969 KOps/s	4.9306 KOps/s	$\color{#d91a1a}-0.68\%$
test_update	0.3653ms	21.7170μs	46.0468 KOps/s	46.1854 KOps/s	$\color{#d91a1a}-0.30\%$
test_update_nested	73.9910μs	33.4785μs	29.8699 KOps/s	29.9962 KOps/s	$\color{#d91a1a}-0.42\%$
test_update__nested	0.4755ms	32.5405μs	30.7310 KOps/s	28.5983 KOps/s	$\textbf{\color{#35bf28}+7.46\%}$
test_set_nested	57.8410μs	18.6393μs	53.6502 KOps/s	53.9088 KOps/s	$\color{#d91a1a}-0.48\%$
test_set_nested_new	65.8710μs	23.7942μs	42.0270 KOps/s	41.6626 KOps/s	$\color{#35bf28}+0.87\%$
test_select	83.6620μs	41.0057μs	24.3868 KOps/s	23.6990 KOps/s	$\color{#35bf28}+2.90\%$
test_select_nested	0.1177ms	70.2030μs	14.2444 KOps/s	13.9679 KOps/s	$\color{#35bf28}+1.98\%$
test_exclude_nested	0.1351ms	92.1692μs	10.8496 KOps/s	10.8672 KOps/s	$\color{#d91a1a}-0.16\%$
test_empty[True]	0.4709ms	0.4161ms	2.4032 KOps/s	2.3980 KOps/s	$\color{#35bf28}+0.22\%$
test_empty[False]	8.0450μs	1.2517μs	798.9207 KOps/s	793.6258 KOps/s	$\color{#35bf28}+0.67\%$
test_to	0.1001ms	69.6001μs	14.3678 KOps/s	14.1144 KOps/s	$\color{#35bf28}+1.80\%$
test_to_nonblocking	0.1559ms	63.7245μs	15.6925 KOps/s	15.9133 KOps/s	$\color{#d91a1a}-1.39\%$
test_unbind_speed	0.3559ms	0.3102ms	3.2235 KOps/s	3.2177 KOps/s	$\color{#35bf28}+0.18\%$
test_unbind_speed_stack0	0.3716ms	0.3080ms	3.2466 KOps/s	3.2282 KOps/s	$\color{#35bf28}+0.57\%$
test_unbind_speed_stack1	98.7742ms	0.9027ms	1.1078 KOps/s	1.1910 KOps/s	$\textbf{\color{#d91a1a}-6.98\%}$
test_split	1.1545ms	1.1098ms	901.0621 Ops/s	677.3472 Ops/s	$\textbf{\color{#35bf28}+33.03\%}$
test_chunk	98.3580ms	1.1652ms	858.2038 Ops/s	949.5347 Ops/s	$\textbf{\color{#d91a1a}-9.62\%}$
test_consolidate[False-None]	3.9466ms	3.7423ms	267.2189 Ops/s	269.1021 Ops/s	$\color{#d91a1a}-0.70\%$
test_consolidate[default-None]	2.0974ms	1.9920ms	502.0201 Ops/s	484.0340 Ops/s	$\color{#35bf28}+3.72\%$
test_consolidate[reduce-overhead-None]	1.9891ms	1.9017ms	525.8361 Ops/s	498.4121 Ops/s	$\textbf{\color{#35bf28}+5.50\%}$
test_consolidate_njt[False-None]	8.8058ms	8.6156ms	116.0686 Ops/s	117.5496 Ops/s	$\color{#d91a1a}-1.26\%$
test_to[False-False-None]	2.1072ms	2.0236ms	494.1697 Ops/s	493.7708 Ops/s	$\color{#35bf28}+0.08\%$
test_to[True-False-None]	2.0824ms	1.7918ms	558.1081 Ops/s	556.5970 Ops/s	$\color{#35bf28}+0.27\%$
test_to[within-False-None]	5.7096ms	5.6133ms	178.1474 Ops/s	181.2369 Ops/s	$\color{#d91a1a}-1.70\%$
test_to[True-default-None]	11.9039ms	11.7398ms	85.1802 Ops/s	86.5040 Ops/s	$\color{#d91a1a}-1.53\%$
test_to_njt[False-False-None]	8.4819ms	8.3897ms	119.1934 Ops/s	119.5545 Ops/s	$\color{#d91a1a}-0.30\%$
test_to_njt[True-False-None]	7.3714ms	7.2244ms	138.4206 Ops/s	139.2561 Ops/s	$\color{#d91a1a}-0.60\%$
test_to_njt[within-False-None]	16.1139ms	16.0300ms	62.3831 Ops/s	62.9222 Ops/s	$\color{#d91a1a}-0.86\%$
test_creation[device0]	0.3496ms	0.1063ms	9.4040 KOps/s	9.2146 KOps/s	$\color{#35bf28}+2.06\%$
test_creation_from_tensor	0.3540ms	0.1086ms	9.2117 KOps/s	9.0758 KOps/s	$\color{#35bf28}+1.50\%$
test_add_one[memmap_tensor0]	0.1661ms	6.5742μs	152.1092 KOps/s	152.7741 KOps/s	$\color{#d91a1a}-0.44\%$
test_contiguous[memmap_tensor0]	25.7200μs	0.6346μs	1.5758 MOps/s	2.1192 MOps/s	$\textbf{\color{#d91a1a}-25.64\%}$
test_stack[memmap_tensor0]	31.7110μs	4.5066μs	221.8979 KOps/s	218.6604 KOps/s	$\color{#35bf28}+1.48\%$
test_memmaptd_index	1.0695ms	0.2700ms	3.7035 KOps/s	3.6898 KOps/s	$\color{#35bf28}+0.37\%$
test_memmaptd_index_astensor	0.5209ms	0.3551ms	2.8158 KOps/s	2.8187 KOps/s	$\color{#d91a1a}-0.10\%$
test_memmaptd_index_op	0.8046ms	0.6008ms	1.6644 KOps/s	1.6541 KOps/s	$\color{#35bf28}+0.62\%$
test_serialize_model	0.1349s	0.1337s	7.4771 Ops/s	7.4681 Ops/s	$\color{#35bf28}+0.12\%$
test_serialize_model_pickle	1.3749s	1.2160s	0.8224 Ops/s	0.8431 Ops/s	$\color{#d91a1a}-2.46\%$
test_serialize_weights	0.1341s	0.1337s	7.4810 Ops/s	5.1143 Ops/s	$\textbf{\color{#35bf28}+46.28\%}$
test_serialize_weights_returnearly	67.7410ms	51.8055ms	19.3030 Ops/s	19.6459 Ops/s	$\color{#d91a1a}-1.75\%$
test_serialize_weights_pickle	1.3648s	1.2212s	0.8188 Ops/s	0.8394 Ops/s	$\color{#d91a1a}-2.45\%$
test_reshape_pytree	0.3613ms	32.4561μs	30.8109 KOps/s	31.3724 KOps/s	$\color{#d91a1a}-1.79\%$
test_reshape_td	79.5620μs	38.3370μs	26.0845 KOps/s	26.9806 KOps/s	$\color{#d91a1a}-3.32\%$
test_view_pytree	0.2180ms	31.4169μs	31.8300 KOps/s	31.6772 KOps/s	$\color{#35bf28}+0.48\%$
test_view_td	77.2620μs	44.0102μs	22.7220 KOps/s	22.4781 KOps/s	$\color{#35bf28}+1.09\%$
test_unbind_pytree	0.2342ms	37.0498μs	26.9907 KOps/s	27.4004 KOps/s	$\color{#d91a1a}-1.50\%$
test_unbind_td	0.1613ms	45.6581μs	21.9019 KOps/s	21.0984 KOps/s	$\color{#35bf28}+3.81\%$
test_split_pytree	0.1955ms	42.3427μs	23.6168 KOps/s	23.6828 KOps/s	$\color{#d91a1a}-0.28\%$
test_split_td	0.1954ms	61.8069μs	16.1794 KOps/s	15.8221 KOps/s	$\color{#35bf28}+2.26\%$
test_add_pytree	0.2478ms	45.0905μs	22.1776 KOps/s	23.2474 KOps/s	$\color{#d91a1a}-4.60\%$
test_add_td	96.6720μs	55.8715μs	17.8982 KOps/s	19.0106 KOps/s	$\textbf{\color{#d91a1a}-5.85\%}$
test_compile_add_one_nested[tensordict-compile]	0.2828ms	0.1767ms	5.6602 KOps/s	5.5834 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_add_one_nested[tensordict-eager]	0.2764ms	0.1886ms	5.3026 KOps/s	5.3799 KOps/s	$\color{#d91a1a}-1.44\%$
test_compile_add_one_nested[pytree-compile]	0.1933ms	0.1482ms	6.7470 KOps/s	6.6412 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_add_one_nested[pytree-eager]	0.4451ms	0.1850ms	5.4067 KOps/s	5.4760 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_copy_nested[tensordict-compile]	59.7910μs	27.4877μs	36.3799 KOps/s	35.1448 KOps/s	$\color{#35bf28}+3.51\%$
test_compile_copy_nested[tensordict-eager]	79.0310μs	50.4877μs	19.8068 KOps/s	19.8471 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_copy_nested[pytree-compile]	58.0510μs	14.0807μs	71.0190 KOps/s	68.4436 KOps/s	$\color{#35bf28}+3.76\%$
test_compile_copy_nested[pytree-eager]	0.3954ms	71.6789μs	13.9511 KOps/s	13.8288 KOps/s	$\color{#35bf28}+0.88\%$
test_compile_add_one_flat[tensordict-compile]	0.2396ms	0.2038ms	4.9060 KOps/s	4.7993 KOps/s	$\color{#35bf28}+2.22\%$
test_compile_add_one_flat[tensordict-eager]	0.3024ms	0.2602ms	3.8436 KOps/s	3.9037 KOps/s	$\color{#d91a1a}-1.54\%$
test_compile_add_one_flat[tensorclass-compile]	0.1974ms	0.1521ms	6.5731 KOps/s	6.4480 KOps/s	$\color{#35bf28}+1.94\%$
test_compile_add_one_flat[tensorclass-eager]	0.1118ms	70.6916μs	14.1459 KOps/s	14.2113 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_add_one_flat[pytree-compile]	0.2596ms	0.1995ms	5.0138 KOps/s	4.9192 KOps/s	$\color{#35bf28}+1.92\%$
test_compile_add_one_flat[pytree-eager]	0.9006ms	0.5438ms	1.8390 KOps/s	1.8534 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_add_self_flat[tensordict-eager]	0.3489ms	0.3093ms	3.2330 KOps/s	3.2364 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_add_self_flat[tensordict-compile]	0.2439ms	0.2073ms	4.8244 KOps/s	4.6522 KOps/s	$\color{#35bf28}+3.70\%$
test_compile_add_self_flat[tensorclass-eager]	0.1277ms	89.1271μs	11.2199 KOps/s	11.8617 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_compile_add_self_flat[tensorclass-compile]	0.2067ms	0.1563ms	6.3976 KOps/s	6.2562 KOps/s	$\color{#35bf28}+2.26\%$
test_compile_add_self_flat[pytree-eager]	0.6791ms	0.4514ms	2.2151 KOps/s	2.2356 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_add_self_flat[pytree-compile]	0.2559ms	0.2011ms	4.9717 KOps/s	4.7953 KOps/s	$\color{#35bf28}+3.68\%$
test_compile_copy_flat[tensordict-compile]	57.8310μs	23.4353μs	42.6706 KOps/s	38.3995 KOps/s	$\textbf{\color{#35bf28}+11.12\%}$
test_compile_copy_flat[tensordict-eager]	70.9520μs	40.4105μs	24.7461 KOps/s	25.1815 KOps/s	$\color{#d91a1a}-1.73\%$
test_compile_copy_flat[pytree-compile]	56.6410μs	21.9194μs	45.6217 KOps/s	46.8941 KOps/s	$\color{#d91a1a}-2.71\%$
test_compile_copy_flat[pytree-eager]	0.3552ms	66.6207μs	15.0104 KOps/s	14.9426 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_assign_and_add[tensordict-compile]	2.0216ms	0.2090ms	4.7857 KOps/s	4.7070 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_assign_and_add[tensordict-eager]	3.5527ms	3.3125ms	301.8873 Ops/s	310.8999 Ops/s	$\color{#d91a1a}-2.90\%$
test_compile_assign_and_add[pytree-compile]	1.9975ms	0.2036ms	4.9116 KOps/s	4.8565 KOps/s	$\color{#35bf28}+1.14\%$
test_compile_assign_and_add[pytree-eager]	3.2111ms	2.9686ms	336.8626 Ops/s	348.6521 Ops/s	$\color{#d91a1a}-3.38\%$
test_compile_indexing[tensor-tensordict-compile]	0.2126ms	0.1462ms	6.8417 KOps/s	6.7336 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_indexing[tensor-tensordict-eager]	0.3025ms	70.2515μs	14.2346 KOps/s	14.4883 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1892ms	0.1413ms	7.0766 KOps/s	7.3716 KOps/s	$\color{#d91a1a}-4.00\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2579ms	46.6914μs	21.4172 KOps/s	21.2026 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_indexing[tensor-pytree-compile]	0.1737ms	0.1351ms	7.4022 KOps/s	7.4243 KOps/s	$\color{#d91a1a}-0.30\%$
test_compile_indexing[tensor-pytree-eager]	0.2843ms	47.7238μs	20.9539 KOps/s	20.9964 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_indexing[slice-tensordict-compile]	0.1258ms	85.9901μs	11.6293 KOps/s	11.2498 KOps/s	$\color{#35bf28}+3.37\%$
test_compile_indexing[slice-tensordict-eager]	0.2129ms	27.5569μs	36.2886 KOps/s	38.0379 KOps/s	$\color{#d91a1a}-4.60\%$
test_compile_indexing[slice-tensorclass-compile]	0.1326ms	80.3813μs	12.4407 KOps/s	12.3665 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_indexing[slice-tensorclass-eager]	0.2255ms	22.8066μs	43.8470 KOps/s	43.3293 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_indexing[slice-pytree-compile]	0.1198ms	82.8216μs	12.0741 KOps/s	11.6586 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_indexing[slice-pytree-eager]	0.2539ms	22.8436μs	43.7759 KOps/s	43.9353 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_indexing[int-tensordict-compile]	0.1345ms	90.2239μs	11.0835 KOps/s	11.0686 KOps/s	$\color{#35bf28}+0.13\%$
test_compile_indexing[int-tensordict-eager]	0.2546ms	25.2749μs	39.5649 KOps/s	38.6627 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_indexing[int-tensorclass-compile]	0.1205ms	80.4968μs	12.4229 KOps/s	11.7668 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_compile_indexing[int-tensorclass-eager]	0.2313ms	22.4667μs	44.5103 KOps/s	43.8204 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_indexing[int-pytree-compile]	0.1687ms	81.2074μs	12.3142 KOps/s	11.7397 KOps/s	$\color{#35bf28}+4.89\%$
test_compile_indexing[int-pytree-eager]	0.2564ms	22.4236μs	44.5959 KOps/s	43.2019 KOps/s	$\color{#35bf28}+3.23\%$
test_mod_add[eager]	92.6820μs	52.4787μs	19.0554 KOps/s	19.8376 KOps/s	$\color{#d91a1a}-3.94\%$
test_mod_add[compile]	0.2041ms	0.1478ms	6.7644 KOps/s	6.3207 KOps/s	$\textbf{\color{#35bf28}+7.02\%}$
test_mod_add[compile-overhead]	0.2824ms	0.1948ms	5.1341 KOps/s	4.9239 KOps/s	$\color{#35bf28}+4.27\%$
test_mod_wrap[eager]	0.4071ms	0.3067ms	3.2605 KOps/s	3.1168 KOps/s	$\color{#35bf28}+4.61\%$
test_mod_wrap[compile]	0.5221ms	0.3962ms	2.5239 KOps/s	2.5421 KOps/s	$\color{#d91a1a}-0.71\%$
test_mod_wrap[compile-overhead]	7.4018ms	3.9177ms	255.2511 Ops/s	261.4842 Ops/s	$\color{#d91a1a}-2.38\%$
test_mod_wrap_and_backward[eager]	1.6646ms	1.5680ms	637.7359 Ops/s	637.3953 Ops/s	$\color{#35bf28}+0.05\%$
test_mod_wrap_and_backward[compile]	1.6502ms	1.5840ms	631.3066 Ops/s	629.0205 Ops/s	$\color{#35bf28}+0.36\%$
test_mod_wrap_and_backward[compile-overhead]	1.3013ms	0.9647ms	1.0366 KOps/s	1.0309 KOps/s	$\color{#35bf28}+0.56\%$
test_seq_add[eager]	0.2104ms	0.1580ms	6.3299 KOps/s	6.5648 KOps/s	$\color{#d91a1a}-3.58\%$
test_seq_add[compile]	0.2127ms	0.1589ms	6.2924 KOps/s	6.3306 KOps/s	$\color{#d91a1a}-0.60\%$
test_seq_add[compile-overhead]	0.2415ms	0.1989ms	5.0284 KOps/s	5.0080 KOps/s	$\color{#35bf28}+0.41\%$
test_seq_wrap[eager]	0.7061ms	0.5594ms	1.7878 KOps/s	1.8904 KOps/s	$\textbf{\color{#d91a1a}-5.43\%}$
test_seq_wrap[compile]	0.5087ms	0.4235ms	2.3613 KOps/s	2.4511 KOps/s	$\color{#d91a1a}-3.66\%$
test_seq_wrap[compile-overhead]	0.3639ms	0.3063ms	3.2645 KOps/s	3.2636 KOps/s	$\color{#35bf28}+0.03\%$
test_func_call_runtime[False-eager]	0.9672ms	0.8787ms	1.1380 KOps/s	1.1514 KOps/s	$\color{#d91a1a}-1.16\%$
test_func_call_runtime[False-compile]	0.9791ms	0.9249ms	1.0812 KOps/s	1.0881 KOps/s	$\color{#d91a1a}-0.64\%$
test_func_call_runtime[False-compile-overhead]	0.5216ms	0.4786ms	2.0895 KOps/s	2.0741 KOps/s	$\color{#35bf28}+0.74\%$
test_func_call_runtime[True-eager]	1.2306ms	1.1155ms	896.4767 Ops/s	899.8950 Ops/s	$\color{#d91a1a}-0.38\%$
test_func_call_runtime[True-compile]	1.1603ms	0.9868ms	1.0134 KOps/s	1.0538 KOps/s	$\color{#d91a1a}-3.84\%$
test_func_call_runtime[True-compile-overhead]	0.5662ms	0.5018ms	1.9926 KOps/s	1.9868 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[False-eager]	1.0951ms	0.9239ms	1.0824 KOps/s	1.1483 KOps/s	$\textbf{\color{#d91a1a}-5.74\%}$
test_func_call_cm_runtime[False-compile]	1.0289ms	0.9609ms	1.0406 KOps/s	1.0842 KOps/s	$\color{#d91a1a}-4.02\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5910ms	0.4837ms	2.0675 KOps/s	2.0670 KOps/s	$\color{#35bf28}+0.02\%$
test_func_call_cm_runtime[True-eager]	1.4062ms	1.2532ms	797.9391 Ops/s	802.0192 Ops/s	$\color{#d91a1a}-0.51\%$
test_func_call_cm_runtime[True-compile]	1.0743ms	0.9803ms	1.0200 KOps/s	1.0243 KOps/s	$\color{#d91a1a}-0.42\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6229ms	0.5338ms	1.8734 KOps/s	1.8593 KOps/s	$\color{#35bf28}+0.76\%$
test_vmap_func_call_cm_runtime[eager]	2.8475ms	2.3560ms	424.4502 Ops/s	423.7846 Ops/s	$\color{#35bf28}+0.16\%$
test_vmap_func_call_cm_runtime[compile]	1.2603ms	1.0017ms	998.2969 Ops/s	983.7563 Ops/s	$\color{#35bf28}+1.48\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5806ms	0.5300ms	1.8868 KOps/s	1.8803 KOps/s	$\color{#35bf28}+0.35\%$
test_distributed	0.5479ms	0.1525ms	6.5595 KOps/s	6.4848 KOps/s	$\color{#35bf28}+1.15\%$
test_tdmodule	0.2856ms	27.2107μs	36.7502 KOps/s	36.9967 KOps/s	$\color{#d91a1a}-0.67\%$
test_tdmodule_dispatch	81.5910μs	46.3524μs	21.5739 KOps/s	21.5767 KOps/s	$\color{#d91a1a}-0.01\%$
test_tdseq	47.1110μs	26.2718μs	38.0636 KOps/s	38.4374 KOps/s	$\color{#d91a1a}-0.97\%$
test_tdseq_dispatch	84.9920μs	48.5391μs	20.6020 KOps/s	20.6134 KOps/s	$\color{#d91a1a}-0.06\%$
test_instantiation_functorch	2.0849ms	1.9973ms	500.6755 Ops/s	504.5698 Ops/s	$\color{#d91a1a}-0.77\%$
test_exec_functorch	0.2259ms	0.1793ms	5.5763 KOps/s	5.5039 KOps/s	$\color{#35bf28}+1.32\%$
test_exec_functional_call	0.2293ms	0.1588ms	6.2961 KOps/s	6.1801 KOps/s	$\color{#35bf28}+1.88\%$
test_exec_td_decorator	0.4721ms	0.2347ms	4.2615 KOps/s	4.2791 KOps/s	$\color{#d91a1a}-0.41\%$
test_vmap_mlp_speed_decorator[True-True]	0.9794ms	0.7985ms	1.2524 KOps/s	1.2473 KOps/s	$\color{#35bf28}+0.41\%$
test_vmap_mlp_speed_decorator[True-False]	0.9998ms	0.8054ms	1.2416 KOps/s	1.2385 KOps/s	$\color{#35bf28}+0.26\%$
test_vmap_mlp_speed_decorator[False-True]	0.9037ms	0.6870ms	1.4556 KOps/s	1.4592 KOps/s	$\color{#d91a1a}-0.25\%$
test_vmap_mlp_speed_decorator[False-False]	0.8792ms	0.6942ms	1.4405 KOps/s	1.4269 KOps/s	$\color{#35bf28}+0.95\%$
test_vmap_transformer_speed_decorator[True-True]	20.7898ms	20.7011ms	48.3067 Ops/s	48.5750 Ops/s	$\color{#d91a1a}-0.55\%$
test_vmap_transformer_speed_decorator[True-False]	21.7234ms	20.7364ms	48.2244 Ops/s	48.5881 Ops/s	$\color{#d91a1a}-0.75\%$
test_vmap_transformer_speed_decorator[False-True]	21.1608ms	20.5243ms	48.7226 Ops/s	48.9938 Ops/s	$\color{#d91a1a}-0.55\%$
test_vmap_transformer_speed_decorator[False-False]	20.8117ms	20.5137ms	48.7478 Ops/s	48.9937 Ops/s	$\color{#d91a1a}-0.50\%$
test_to_module_speed[True]	1.5035ms	1.3969ms	715.8791 Ops/s	717.6078 Ops/s	$\color{#d91a1a}-0.24\%$
test_to_module_speed[False]	1.4679ms	1.3662ms	731.9571 Ops/s	727.6130 Ops/s	$\color{#35bf28}+0.60\%$
test_tc_init	84.4410μs	50.4089μs	19.8378 KOps/s	19.7888 KOps/s	$\color{#35bf28}+0.25\%$
test_tc_init_tensor_only	47.3710μs	14.1487μs	70.6777 KOps/s	70.2866 KOps/s	$\color{#35bf28}+0.56\%$
test_tc_init_nested	0.1507ms	99.0695μs	10.0939 KOps/s	10.0019 KOps/s	$\color{#35bf28}+0.92\%$
test_tc_first_layer_tensor	33.3810μs	1.6775μs	596.1402 KOps/s	589.4324 KOps/s	$\color{#35bf28}+1.14\%$
test_tc_first_layer_tensor_only	7.5089μs	0.6523μs	1.5331 MOps/s	1.4854 MOps/s	$\color{#35bf28}+3.21\%$
test_tc_first_layer_tensor_set	35.4910μs	3.9539μs	252.9177 KOps/s	248.1826 KOps/s	$\color{#35bf28}+1.91\%$
test_tc_first_layer_tensor_only_set	29.4200μs	2.9743μs	336.2183 KOps/s	332.8704 KOps/s	$\color{#35bf28}+1.01\%$
test_tc_first_layer_nontensor	28.3000μs	5.6245μs	177.7932 KOps/s	177.0970 KOps/s	$\color{#35bf28}+0.39\%$
test_tc_second_layer_tensor	23.9800μs	4.0028μs	249.8221 KOps/s	246.0809 KOps/s	$\color{#35bf28}+1.52\%$
test_tc_second_layer_nontensor	38.1710μs	7.8463μs	127.4487 KOps/s	128.6794 KOps/s	$\color{#d91a1a}-0.96\%$
test_unbind	0.2565s	13.0842ms	76.4282 Ops/s	66.5074 Ops/s	$\textbf{\color{#35bf28}+14.92\%}$
test_full_like	4.9771ms	4.4000ms	227.2729 Ops/s	225.9471 Ops/s	$\color{#35bf28}+0.59\%$
test_zeros_like	5.0191ms	4.3773ms	228.4525 Ops/s	228.3409 Ops/s	$\color{#35bf28}+0.05\%$
test_ones_like	4.5904ms	4.3815ms	228.2336 Ops/s	227.8812 Ops/s	$\color{#35bf28}+0.15\%$
test_clone	6.8983ms	6.5757ms	152.0759 Ops/s	152.2187 Ops/s	$\color{#d91a1a}-0.09\%$
test_squeeze	0.2099ms	13.5285μs	73.9181 KOps/s	73.2633 KOps/s	$\color{#35bf28}+0.89\%$
test_unsqueeze	0.2798ms	0.1076ms	9.2962 KOps/s	9.3099 KOps/s	$\color{#d91a1a}-0.15\%$
test_split	0.2376ms	0.1772ms	5.6447 KOps/s	5.4820 KOps/s	$\color{#35bf28}+2.97\%$
test_permute	0.2542ms	0.2017ms	4.9572 KOps/s	5.1247 KOps/s	$\color{#d91a1a}-3.27\%$
test_stack	52.1193ms	51.6900ms	19.3461 Ops/s	19.3245 Ops/s	$\color{#35bf28}+0.11\%$
test_cat	52.2121ms	51.8118ms	19.3006 Ops/s	23.0480 Ops/s	$\textbf{\color{#d91a1a}-16.26\%}$

[github-actions]

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	39.6810μs	15.2491μs	65.5776 KOps/s	66.5274 KOps/s	$\color{#d91a1a}-1.43\%$
test_plain_set_stack_nested	59.8720μs	15.4145μs	64.8738 KOps/s	65.5916 KOps/s	$\color{#d91a1a}-1.09\%$
test_plain_set_nested_inplace	47.6310μs	16.8351μs	59.3997 KOps/s	59.6539 KOps/s	$\color{#d91a1a}-0.43\%$
test_plain_set_stack_nested_inplace	52.7610μs	16.8144μs	59.4730 KOps/s	59.4228 KOps/s	$\color{#35bf28}+0.08\%$
test_items	40.7210μs	5.7736μs	173.2008 KOps/s	172.3257 KOps/s	$\color{#35bf28}+0.51\%$
test_items_nested	0.5950ms	0.5398ms	1.8525 KOps/s	1.8723 KOps/s	$\color{#d91a1a}-1.06\%$
test_items_nested_locked	0.6230ms	0.5438ms	1.8389 KOps/s	1.8595 KOps/s	$\color{#d91a1a}-1.11\%$
test_items_nested_leaf	0.1425ms	95.3100μs	10.4921 KOps/s	10.3491 KOps/s	$\color{#35bf28}+1.38\%$
test_items_stack_nested	0.6134ms	0.5343ms	1.8717 KOps/s	1.8544 KOps/s	$\color{#35bf28}+0.94\%$
test_items_stack_nested_leaf	0.1241ms	97.0849μs	10.3003 KOps/s	10.2276 KOps/s	$\color{#35bf28}+0.71\%$
test_items_stack_nested_locked	0.7592ms	0.5346ms	1.8706 KOps/s	1.8525 KOps/s	$\color{#35bf28}+0.98\%$
test_keys	61.5620μs	4.1983μs	238.1928 KOps/s	238.3800 KOps/s	$\color{#d91a1a}-0.08\%$
test_keys_nested	0.1536ms	0.1196ms	8.3637 KOps/s	8.2927 KOps/s	$\color{#35bf28}+0.86\%$
test_keys_nested_locked	0.7452ms	0.1295ms	7.7198 KOps/s	7.7845 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys_nested_leaf	0.1394ms	0.1110ms	9.0082 KOps/s	9.0914 KOps/s	$\color{#d91a1a}-0.92\%$
test_keys_stack_nested	0.1597ms	0.1213ms	8.2465 KOps/s	8.2688 KOps/s	$\color{#d91a1a}-0.27\%$
test_keys_stack_nested_leaf	0.1490ms	0.1110ms	9.0118 KOps/s	8.9709 KOps/s	$\color{#35bf28}+0.46\%$
test_keys_stack_nested_locked	0.1768ms	0.1288ms	7.7649 KOps/s	7.7613 KOps/s	$\color{#35bf28}+0.05\%$
test_values	7.4440μs	1.0273μs	973.4640 KOps/s	973.9872 KOps/s	$\color{#d91a1a}-0.05\%$
test_values_nested	72.4920μs	48.2981μs	20.7048 KOps/s	20.5968 KOps/s	$\color{#35bf28}+0.52\%$
test_values_nested_locked	82.0220μs	50.9152μs	19.6405 KOps/s	19.2631 KOps/s	$\color{#35bf28}+1.96\%$
test_values_nested_leaf	78.9010μs	54.7401μs	18.2681 KOps/s	18.1949 KOps/s	$\color{#35bf28}+0.40\%$
test_values_stack_nested	73.5410μs	48.2555μs	20.7230 KOps/s	20.6595 KOps/s	$\color{#35bf28}+0.31\%$
test_values_stack_nested_leaf	1.0766ms	54.5058μs	18.3467 KOps/s	18.1678 KOps/s	$\color{#35bf28}+0.98\%$
test_values_stack_nested_locked	88.3520μs	50.8287μs	19.6739 KOps/s	19.4764 KOps/s	$\color{#35bf28}+1.01\%$
test_membership	5.4785μs	0.8483μs	1.1788 MOps/s	1.1863 MOps/s	$\color{#d91a1a}-0.63\%$
test_membership_nested	29.3910μs	3.1779μs	314.6748 KOps/s	313.3996 KOps/s	$\color{#35bf28}+0.41\%$
test_membership_nested_leaf	28.8200μs	3.1948μs	313.0100 KOps/s	313.1632 KOps/s	$\color{#d91a1a}-0.05\%$
test_membership_stacked_nested	37.0910μs	3.1955μs	312.9432 KOps/s	311.3553 KOps/s	$\color{#35bf28}+0.51\%$
test_membership_stacked_nested_leaf	33.3710μs	3.1869μs	313.7876 KOps/s	313.0446 KOps/s	$\color{#35bf28}+0.24\%$
test_membership_nested_last	29.3610μs	4.6914μs	213.1581 KOps/s	216.9031 KOps/s	$\color{#d91a1a}-1.73\%$
test_membership_nested_leaf_last	39.0610μs	4.6398μs	215.5276 KOps/s	215.9277 KOps/s	$\color{#d91a1a}-0.19\%$
test_membership_stacked_nested_last	37.9510μs	4.7140μs	212.1329 KOps/s	214.9179 KOps/s	$\color{#d91a1a}-1.30\%$
test_membership_stacked_nested_leaf_last	37.5110μs	4.6504μs	215.0347 KOps/s	214.2371 KOps/s	$\color{#35bf28}+0.37\%$
test_nested_getleaf	50.8810μs	21.5394μs	46.4266 KOps/s	46.8255 KOps/s	$\color{#d91a1a}-0.85\%$
test_nested_get	48.1710μs	20.4094μs	48.9970 KOps/s	48.6764 KOps/s	$\color{#35bf28}+0.66\%$
test_stacked_getleaf	45.7010μs	21.7926μs	45.8872 KOps/s	46.5525 KOps/s	$\color{#d91a1a}-1.43\%$
test_stacked_get	49.2410μs	20.5657μs	48.6246 KOps/s	49.4609 KOps/s	$\color{#d91a1a}-1.69\%$
test_nested_getitemleaf	66.7410μs	21.8199μs	45.8297 KOps/s	45.6686 KOps/s	$\color{#35bf28}+0.35\%$
test_nested_getitem	65.5210μs	20.8628μs	47.9322 KOps/s	48.8069 KOps/s	$\color{#d91a1a}-1.79\%$
test_stacked_getitemleaf	47.2500μs	22.2245μs	44.9953 KOps/s	45.3942 KOps/s	$\color{#d91a1a}-0.88\%$
test_stacked_getitem	44.3110μs	21.0119μs	47.5921 KOps/s	48.0660 KOps/s	$\color{#d91a1a}-0.99\%$
test_lock_nested	7.9643ms	0.4825ms	2.0725 KOps/s	2.0907 KOps/s	$\color{#d91a1a}-0.87\%$
test_lock_stack_nested	0.6237ms	0.4791ms	2.0874 KOps/s	2.0912 KOps/s	$\color{#d91a1a}-0.18\%$
test_unlock_nested	0.6410ms	0.3832ms	2.6094 KOps/s	2.5969 KOps/s	$\color{#35bf28}+0.48\%$
test_unlock_stack_nested	0.4580ms	0.3821ms	2.6170 KOps/s	2.6038 KOps/s	$\color{#35bf28}+0.51\%$
test_flatten_speed	0.1817ms	0.1212ms	8.2495 KOps/s	8.2691 KOps/s	$\color{#d91a1a}-0.24\%$
test_unflatten_speed	0.6620ms	0.5988ms	1.6701 KOps/s	1.6826 KOps/s	$\color{#d91a1a}-0.74\%$
test_common_ops	0.8441ms	0.7407ms	1.3500 KOps/s	1.3343 KOps/s	$\color{#35bf28}+1.18\%$
test_creation	95.5220μs	2.7504μs	363.5792 KOps/s	367.0978 KOps/s	$\color{#d91a1a}-0.96\%$
test_creation_empty	51.5610μs	9.1204μs	109.6448 KOps/s	110.4981 KOps/s	$\color{#d91a1a}-0.77\%$
test_creation_nested_1	41.2610μs	11.9364μs	83.7774 KOps/s	83.7090 KOps/s	$\color{#35bf28}+0.08\%$
test_creation_nested_2	62.3310μs	16.0818μs	62.1822 KOps/s	62.6666 KOps/s	$\color{#d91a1a}-0.77\%$
test_clone	54.1110μs	13.4680μs	74.2503 KOps/s	75.0112 KOps/s	$\color{#d91a1a}-1.01\%$
test_getitem[int]	1.1588ms	14.0459μs	71.1951 KOps/s	70.2571 KOps/s	$\color{#35bf28}+1.34\%$
test_getitem[slice_int]	0.1444ms	24.7029μs	40.4811 KOps/s	40.0782 KOps/s	$\color{#35bf28}+1.01\%$
test_getitem[range]	0.1729ms	60.1378μs	16.6285 KOps/s	16.7303 KOps/s	$\color{#d91a1a}-0.61\%$
test_getitem[tuple]	0.1440ms	24.5876μs	40.6709 KOps/s	40.5711 KOps/s	$\color{#35bf28}+0.25\%$
test_getitem[list]	0.1770ms	54.7612μs	18.2611 KOps/s	18.6590 KOps/s	$\color{#d91a1a}-2.13\%$
test_setitem_dim[int]	46.6910μs	24.5039μs	40.8099 KOps/s	41.1236 KOps/s	$\color{#d91a1a}-0.76\%$
test_setitem_dim[slice_int]	68.9610μs	45.3908μs	22.0309 KOps/s	21.8688 KOps/s	$\color{#35bf28}+0.74\%$
test_setitem_dim[range]	0.1077ms	85.9221μs	11.6384 KOps/s	11.6358 KOps/s	$\color{#35bf28}+0.02\%$
test_setitem_dim[tuple]	61.7010μs	41.1467μs	24.3033 KOps/s	23.7351 KOps/s	$\color{#35bf28}+2.39\%$
test_setitem	61.9810μs	18.3417μs	54.5207 KOps/s	54.7117 KOps/s	$\color{#d91a1a}-0.35\%$
test_set	61.8720μs	17.3932μs	57.4937 KOps/s	54.4256 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_set_shared	0.4878ms	0.2067ms	4.8381 KOps/s	4.7924 KOps/s	$\color{#35bf28}+0.95\%$
test_update	0.2219ms	22.1930μs	45.0593 KOps/s	45.0765 KOps/s	$\color{#d91a1a}-0.04\%$
test_update_nested	77.1210μs	34.4232μs	29.0501 KOps/s	28.7666 KOps/s	$\color{#35bf28}+0.99\%$
test_update__nested	0.4692ms	34.4117μs	29.0599 KOps/s	27.9399 KOps/s	$\color{#35bf28}+4.01\%$
test_set_nested	65.0810μs	19.5238μs	51.2197 KOps/s	51.8631 KOps/s	$\color{#d91a1a}-1.24\%$
test_set_nested_new	68.7320μs	24.3964μs	40.9896 KOps/s	41.3419 KOps/s	$\color{#d91a1a}-0.85\%$
test_select	87.3410μs	43.0414μs	23.2334 KOps/s	23.3637 KOps/s	$\color{#d91a1a}-0.56\%$
test_select_nested	0.1198ms	75.3120μs	13.2781 KOps/s	13.4287 KOps/s	$\color{#d91a1a}-1.12\%$
test_exclude_nested	0.1290ms	97.7450μs	10.2307 KOps/s	10.2211 KOps/s	$\color{#35bf28}+0.09\%$
test_empty[True]	0.5040ms	0.4354ms	2.2969 KOps/s	2.2795 KOps/s	$\color{#35bf28}+0.76\%$
test_empty[False]	7.2102μs	1.3321μs	750.6684 KOps/s	752.1901 KOps/s	$\color{#d91a1a}-0.20\%$
test_to	0.1047ms	73.2250μs	13.6565 KOps/s	13.4820 KOps/s	$\color{#35bf28}+1.29\%$
test_to_nonblocking	0.1146ms	65.7830μs	15.2015 KOps/s	15.1850 KOps/s	$\color{#35bf28}+0.11\%$
test_unbind_speed	0.3730ms	0.3276ms	3.0521 KOps/s	3.0963 KOps/s	$\color{#d91a1a}-1.43\%$
test_unbind_speed_stack0	0.4793ms	0.3234ms	3.0922 KOps/s	3.1068 KOps/s	$\color{#d91a1a}-0.47\%$
test_unbind_speed_stack1	99.7706ms	0.9452ms	1.0580 KOps/s	1.1508 KOps/s	$\textbf{\color{#d91a1a}-8.07\%}$
test_split	1.2505ms	1.1565ms	864.6605 Ops/s	866.8946 Ops/s	$\color{#d91a1a}-0.26\%$
test_chunk	99.3968ms	1.2175ms	821.3437 Ops/s	754.4221 Ops/s	$\textbf{\color{#35bf28}+8.87\%}$
test_consolidate[False-None]	4.0166ms	3.8816ms	257.6259 Ops/s	256.6516 Ops/s	$\color{#35bf28}+0.38\%$
test_consolidate[default-None]	2.1657ms	2.0552ms	486.5690 Ops/s	460.4803 Ops/s	$\textbf{\color{#35bf28}+5.67\%}$
test_consolidate[reduce-overhead-None]	2.1229ms	1.9772ms	505.7621 Ops/s	473.2462 Ops/s	$\textbf{\color{#35bf28}+6.87\%}$
test_consolidate_njt[False-None]	9.1576ms	8.8269ms	113.2901 Ops/s	111.2589 Ops/s	$\color{#35bf28}+1.83\%$
test_to[False-False-None]	2.1856ms	2.0700ms	483.0924 Ops/s	477.4677 Ops/s	$\color{#35bf28}+1.18\%$
test_to[True-False-None]	0.1754s	2.2159ms	451.2796 Ops/s	537.7088 Ops/s	$\textbf{\color{#d91a1a}-16.07\%}$
test_to[within-False-None]	6.1657ms	5.8684ms	170.4044 Ops/s	170.5205 Ops/s	$\color{#d91a1a}-0.07\%$
test_to[True-default-None]	13.0327ms	12.2995ms	81.3044 Ops/s	83.2909 Ops/s	$\color{#d91a1a}-2.39\%$
test_to_njt[False-False-None]	9.0310ms	8.6061ms	116.1961 Ops/s	114.6585 Ops/s	$\color{#35bf28}+1.34\%$
test_to_njt[True-False-None]	7.8852ms	7.6333ms	131.0055 Ops/s	133.7059 Ops/s	$\color{#d91a1a}-2.02\%$
test_to_njt[within-False-None]	17.3925ms	16.5348ms	60.4786 Ops/s	59.1906 Ops/s	$\color{#35bf28}+2.18\%$
test_creation[device0]	0.3398ms	0.1138ms	8.7852 KOps/s	9.0124 KOps/s	$\color{#d91a1a}-2.52\%$
test_creation_from_tensor	0.3609ms	0.1141ms	8.7627 KOps/s	8.9393 KOps/s	$\color{#d91a1a}-1.98\%$
test_add_one[memmap_tensor0]	0.1640ms	6.7522μs	148.0991 KOps/s	147.8935 KOps/s	$\color{#35bf28}+0.14\%$
test_contiguous[memmap_tensor0]	26.6600μs	0.6933μs	1.4424 MOps/s	1.9975 MOps/s	$\textbf{\color{#d91a1a}-27.79\%}$
test_stack[memmap_tensor0]	32.2800μs	4.8623μs	205.6644 KOps/s	209.2184 KOps/s	$\color{#d91a1a}-1.70\%$
test_memmaptd_index	0.9880ms	0.2849ms	3.5103 KOps/s	3.6142 KOps/s	$\color{#d91a1a}-2.87\%$
test_memmaptd_index_astensor	1.1176ms	0.3792ms	2.6369 KOps/s	2.7042 KOps/s	$\color{#d91a1a}-2.49\%$
test_memmaptd_index_op	0.8071ms	0.6296ms	1.5883 KOps/s	1.6104 KOps/s	$\color{#d91a1a}-1.37\%$
test_serialize_model	0.1355s	0.1342s	7.4536 Ops/s	7.5002 Ops/s	$\color{#d91a1a}-0.62\%$
test_serialize_model_pickle	1.3479s	1.2111s	0.8257 Ops/s	0.8227 Ops/s	$\color{#35bf28}+0.36\%$
test_serialize_weights	0.1346s	0.1329s	7.5254 Ops/s	7.5363 Ops/s	$\color{#d91a1a}-0.14\%$
test_serialize_weights_returnearly	0.4108s	70.4879ms	14.1868 Ops/s	19.4168 Ops/s	$\textbf{\color{#d91a1a}-26.94\%}$
test_serialize_weights_pickle	1.3745s	1.1977s	0.8349 Ops/s	0.8353 Ops/s	$\color{#d91a1a}-0.05\%$
test_reshape_pytree	9.8355ms	34.0884μs	29.3355 KOps/s	30.1389 KOps/s	$\color{#d91a1a}-2.67\%$
test_reshape_td	72.1710μs	39.0750μs	25.5918 KOps/s	26.6442 KOps/s	$\color{#d91a1a}-3.95\%$
test_view_pytree	0.2272ms	33.2988μs	30.0311 KOps/s	30.1433 KOps/s	$\color{#d91a1a}-0.37\%$
test_view_td	80.5310μs	46.2754μs	21.6097 KOps/s	21.7122 KOps/s	$\color{#d91a1a}-0.47\%$
test_unbind_pytree	0.2436ms	38.3926μs	26.0467 KOps/s	26.0559 KOps/s	$\color{#d91a1a}-0.04\%$
test_unbind_td	0.2127ms	49.0330μs	20.3944 KOps/s	20.4696 KOps/s	$\color{#d91a1a}-0.37\%$
test_split_pytree	0.2553ms	45.5126μs	21.9719 KOps/s	22.4310 KOps/s	$\color{#d91a1a}-2.05\%$
test_split_td	0.2038ms	65.0622μs	15.3699 KOps/s	15.4813 KOps/s	$\color{#d91a1a}-0.72\%$
test_add_pytree	0.2321ms	46.7285μs	21.4002 KOps/s	22.5332 KOps/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_add_td	96.7020μs	57.5609μs	17.3729 KOps/s	19.1388 KOps/s	$\textbf{\color{#d91a1a}-9.23\%}$
test_compile_add_one_nested[tensordict-compile]	0.3119ms	0.1847ms	5.4135 KOps/s	5.4786 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_add_one_nested[tensordict-eager]	0.2513ms	0.1933ms	5.1741 KOps/s	5.1280 KOps/s	$\color{#35bf28}+0.90\%$
test_compile_add_one_nested[pytree-compile]	0.2283ms	0.1559ms	6.4136 KOps/s	6.1010 KOps/s	$\textbf{\color{#35bf28}+5.12\%}$
test_compile_add_one_nested[pytree-eager]	0.4346ms	0.1896ms	5.2747 KOps/s	5.2860 KOps/s	$\color{#d91a1a}-0.21\%$
test_compile_copy_nested[tensordict-compile]	0.1313ms	28.3698μs	35.2488 KOps/s	35.8946 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_copy_nested[tensordict-eager]	86.5020μs	52.9995μs	18.8681 KOps/s	18.9189 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_copy_nested[pytree-compile]	0.1441ms	14.2704μs	70.0750 KOps/s	69.1206 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_copy_nested[pytree-eager]	0.3828ms	76.4335μs	13.0833 KOps/s	13.0012 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_add_one_flat[tensordict-compile]	0.3290ms	0.2074ms	4.8215 KOps/s	4.7359 KOps/s	$\color{#35bf28}+1.81\%$
test_compile_add_one_flat[tensordict-eager]	0.3350ms	0.2611ms	3.8300 KOps/s	3.8155 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_add_one_flat[tensorclass-compile]	0.2624ms	0.1549ms	6.4566 KOps/s	6.3605 KOps/s	$\color{#35bf28}+1.51\%$
test_compile_add_one_flat[tensorclass-eager]	0.1204ms	71.2063μs	14.0437 KOps/s	14.2562 KOps/s	$\color{#d91a1a}-1.49\%$
test_compile_add_one_flat[pytree-compile]	0.2478ms	0.2036ms	4.9123 KOps/s	4.6381 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_compile_add_one_flat[pytree-eager]	0.8213ms	0.5422ms	1.8445 KOps/s	1.8471 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_add_self_flat[tensordict-eager]	0.4517ms	0.3166ms	3.1584 KOps/s	3.1319 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_add_self_flat[tensordict-compile]	0.3308ms	0.2116ms	4.7265 KOps/s	4.5346 KOps/s	$\color{#35bf28}+4.23\%$
test_compile_add_self_flat[tensorclass-eager]	0.1275ms	86.2577μs	11.5932 KOps/s	11.5412 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_add_self_flat[tensorclass-compile]	0.2045ms	0.1570ms	6.3708 KOps/s	6.2213 KOps/s	$\color{#35bf28}+2.40\%$
test_compile_add_self_flat[pytree-eager]	0.6780ms	0.4512ms	2.2164 KOps/s	2.2343 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_add_self_flat[pytree-compile]	0.2509ms	0.2037ms	4.9086 KOps/s	4.7983 KOps/s	$\color{#35bf28}+2.30\%$
test_compile_copy_flat[tensordict-compile]	0.1411ms	23.9691μs	41.7204 KOps/s	40.2482 KOps/s	$\color{#35bf28}+3.66\%$
test_compile_copy_flat[tensordict-eager]	0.1417ms	41.1042μs	24.3284 KOps/s	24.0033 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_copy_flat[pytree-compile]	92.2810μs	20.8024μs	48.0715 KOps/s	47.5301 KOps/s	$\color{#35bf28}+1.14\%$
test_compile_copy_flat[pytree-eager]	0.3655ms	70.4728μs	14.1899 KOps/s	14.3753 KOps/s	$\color{#d91a1a}-1.29\%$
test_compile_assign_and_add[tensordict-compile]	2.0976ms	0.2127ms	4.7019 KOps/s	4.6276 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_assign_and_add[tensordict-eager]	3.6568ms	3.3959ms	294.4744 Ops/s	297.3034 Ops/s	$\color{#d91a1a}-0.95\%$
test_compile_assign_and_add[pytree-compile]	2.0585ms	0.2105ms	4.7503 KOps/s	4.7211 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_assign_and_add[pytree-eager]	3.0658ms	2.9414ms	339.9774 Ops/s	356.8828 Ops/s	$\color{#d91a1a}-4.74\%$
test_compile_indexing[tensor-tensordict-compile]	0.2117ms	0.1430ms	6.9921 KOps/s	6.9194 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_indexing[tensor-tensordict-eager]	0.2982ms	67.1802μs	14.8853 KOps/s	15.0236 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2820ms	0.1368ms	7.3126 KOps/s	7.2186 KOps/s	$\color{#35bf28}+1.30\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2578ms	46.3027μs	21.5970 KOps/s	21.3313 KOps/s	$\color{#35bf28}+1.25\%$
test_compile_indexing[tensor-pytree-compile]	0.1997ms	0.1379ms	7.2538 KOps/s	7.1696 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_indexing[tensor-pytree-eager]	0.2776ms	49.7517μs	20.0998 KOps/s	21.3797 KOps/s	$\textbf{\color{#d91a1a}-5.99\%}$
test_compile_indexing[slice-tensordict-compile]	0.2388ms	86.7788μs	11.5236 KOps/s	11.1641 KOps/s	$\color{#35bf28}+3.22\%$
test_compile_indexing[slice-tensordict-eager]	0.2155ms	27.4071μs	36.4869 KOps/s	36.8401 KOps/s	$\color{#d91a1a}-0.96\%$
test_compile_indexing[slice-tensorclass-compile]	0.1912ms	81.3403μs	12.2940 KOps/s	12.1938 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_indexing[slice-tensorclass-eager]	0.2373ms	24.1537μs	41.4015 KOps/s	42.0176 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_indexing[slice-pytree-compile]	0.1349ms	81.7467μs	12.2329 KOps/s	12.1491 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_indexing[slice-pytree-eager]	0.2304ms	23.9748μs	41.7104 KOps/s	42.0352 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_indexing[int-tensordict-compile]	0.1605ms	87.6873μs	11.4042 KOps/s	11.3795 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_indexing[int-tensordict-eager]	0.2641ms	28.4033μs	35.2072 KOps/s	37.0830 KOps/s	$\textbf{\color{#d91a1a}-5.06\%}$
test_compile_indexing[int-tensorclass-compile]	0.1206ms	81.9988μs	12.1953 KOps/s	12.1056 KOps/s	$\color{#35bf28}+0.74\%$
test_compile_indexing[int-tensorclass-eager]	0.2321ms	23.7510μs	42.1035 KOps/s	41.8062 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_indexing[int-pytree-compile]	0.1237ms	81.4219μs	12.2817 KOps/s	12.1520 KOps/s	$\color{#35bf28}+1.07\%$
test_compile_indexing[int-pytree-eager]	0.2441ms	23.8345μs	41.9560 KOps/s	41.9823 KOps/s	$\color{#d91a1a}-0.06\%$
test_mod_add[eager]	99.1920μs	52.3394μs	19.1061 KOps/s	19.2268 KOps/s	$\color{#d91a1a}-0.63\%$
test_mod_add[compile]	0.2122ms	0.1532ms	6.5268 KOps/s	6.5024 KOps/s	$\color{#35bf28}+0.37\%$
test_mod_add[compile-overhead]	0.6106ms	0.2008ms	4.9811 KOps/s	4.9360 KOps/s	$\color{#35bf28}+0.91\%$
test_mod_wrap[eager]	0.4089ms	0.3134ms	3.1903 KOps/s	3.2223 KOps/s	$\color{#d91a1a}-0.99\%$
test_mod_wrap[compile]	0.5179ms	0.4056ms	2.4654 KOps/s	2.4325 KOps/s	$\color{#35bf28}+1.35\%$
test_mod_wrap[compile-overhead]	7.5845ms	4.0001ms	249.9931 Ops/s	253.5575 Ops/s	$\color{#d91a1a}-1.41\%$
test_mod_wrap_and_backward[eager]	1.7003ms	1.5690ms	637.3298 Ops/s	603.1250 Ops/s	$\textbf{\color{#35bf28}+5.67\%}$
test_mod_wrap_and_backward[compile]	1.8017ms	1.6266ms	614.7829 Ops/s	568.1210 Ops/s	$\textbf{\color{#35bf28}+8.21\%}$
test_mod_wrap_and_backward[compile-overhead]	1.5117ms	0.9986ms	1.0014 KOps/s	894.1144 Ops/s	$\textbf{\color{#35bf28}+12.00\%}$
test_seq_add[eager]	0.2154ms	0.1590ms	6.2876 KOps/s	6.3356 KOps/s	$\color{#d91a1a}-0.76\%$
test_seq_add[compile]	0.2162ms	0.1623ms	6.1618 KOps/s	6.1159 KOps/s	$\color{#35bf28}+0.75\%$
test_seq_add[compile-overhead]	0.2584ms	0.2108ms	4.7440 KOps/s	4.8906 KOps/s	$\color{#d91a1a}-3.00\%$
test_seq_wrap[eager]	0.6397ms	0.5505ms	1.8164 KOps/s	1.8111 KOps/s	$\color{#35bf28}+0.29\%$
test_seq_wrap[compile]	0.5354ms	0.4203ms	2.3790 KOps/s	2.3719 KOps/s	$\color{#35bf28}+0.30\%$
test_seq_wrap[compile-overhead]	0.4137ms	0.3165ms	3.1591 KOps/s	3.1631 KOps/s	$\color{#d91a1a}-0.13\%$
test_func_call_runtime[False-eager]	1.0052ms	0.8956ms	1.1165 KOps/s	1.1205 KOps/s	$\color{#d91a1a}-0.35\%$
test_func_call_runtime[False-compile]	1.0202ms	0.9532ms	1.0491 KOps/s	1.0487 KOps/s	$\color{#35bf28}+0.04\%$
test_func_call_runtime[False-compile-overhead]	0.5617ms	0.5015ms	1.9941 KOps/s	1.9761 KOps/s	$\color{#35bf28}+0.91\%$
test_func_call_runtime[True-eager]	1.2328ms	1.1331ms	882.4972 Ops/s	885.2426 Ops/s	$\color{#d91a1a}-0.31\%$
test_func_call_runtime[True-compile]	1.0828ms	0.9793ms	1.0212 KOps/s	1.0201 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_runtime[True-compile-overhead]	0.5731ms	0.5208ms	1.9200 KOps/s	1.8929 KOps/s	$\color{#35bf28}+1.43\%$
test_func_call_cm_runtime[False-eager]	0.9628ms	0.8957ms	1.1164 KOps/s	1.0804 KOps/s	$\color{#35bf28}+3.33\%$
test_func_call_cm_runtime[False-compile]	1.1682ms	0.9637ms	1.0377 KOps/s	1.0447 KOps/s	$\color{#d91a1a}-0.68\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5495ms	0.4995ms	2.0022 KOps/s	1.9773 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_cm_runtime[True-eager]	1.4043ms	1.2799ms	781.3394 Ops/s	775.8520 Ops/s	$\color{#35bf28}+0.71\%$
test_func_call_cm_runtime[True-compile]	1.0617ms	1.0116ms	988.5148 Ops/s	992.0090 Ops/s	$\color{#d91a1a}-0.35\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6256ms	0.5544ms	1.8037 KOps/s	1.7719 KOps/s	$\color{#35bf28}+1.79\%$
test_vmap_func_call_cm_runtime[eager]	3.0085ms	2.4027ms	416.1938 Ops/s	419.8471 Ops/s	$\color{#d91a1a}-0.87\%$
test_vmap_func_call_cm_runtime[compile]	1.0969ms	1.0273ms	973.4159 Ops/s	980.0453 Ops/s	$\color{#d91a1a}-0.68\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6369ms	0.5471ms	1.8277 KOps/s	1.7961 KOps/s	$\color{#35bf28}+1.76\%$
test_distributed	3.1234ms	0.1622ms	6.1667 KOps/s	6.4100 KOps/s	$\color{#d91a1a}-3.80\%$
test_tdmodule	0.1990ms	27.9144μs	35.8238 KOps/s	35.0459 KOps/s	$\color{#35bf28}+2.22\%$
test_tdmodule_dispatch	77.6520μs	48.0846μs	20.7967 KOps/s	20.7148 KOps/s	$\color{#35bf28}+0.40\%$
test_tdseq	46.3010μs	27.4248μs	36.4634 KOps/s	36.6021 KOps/s	$\color{#d91a1a}-0.38\%$
test_tdseq_dispatch	70.0220μs	50.4938μs	19.8044 KOps/s	19.6038 KOps/s	$\color{#35bf28}+1.02\%$
test_instantiation_functorch	2.2485ms	2.0960ms	477.1013 Ops/s	476.0608 Ops/s	$\color{#35bf28}+0.22\%$
test_exec_functorch	0.2282ms	0.1854ms	5.3937 KOps/s	5.4067 KOps/s	$\color{#d91a1a}-0.24\%$
test_exec_functional_call	0.2603ms	0.1679ms	5.9563 KOps/s	6.1075 KOps/s	$\color{#d91a1a}-2.48\%$
test_exec_td_decorator	0.4462ms	0.2419ms	4.1337 KOps/s	4.1607 KOps/s	$\color{#d91a1a}-0.65\%$
test_vmap_mlp_speed_decorator[True-True]	0.9982ms	0.8188ms	1.2213 KOps/s	1.2194 KOps/s	$\color{#35bf28}+0.15\%$
test_vmap_mlp_speed_decorator[True-False]	0.9727ms	0.8150ms	1.2270 KOps/s	1.2198 KOps/s	$\color{#35bf28}+0.59\%$
test_vmap_mlp_speed_decorator[False-True]	0.8552ms	0.7036ms	1.4212 KOps/s	1.4216 KOps/s	$\color{#d91a1a}-0.03\%$
test_vmap_mlp_speed_decorator[False-False]	0.8595ms	0.7040ms	1.4204 KOps/s	1.4184 KOps/s	$\color{#35bf28}+0.14\%$
test_vmap_transformer_speed_decorator[True-True]	20.9935ms	20.9076ms	47.8296 Ops/s	47.8546 Ops/s	$\color{#d91a1a}-0.05\%$
test_vmap_transformer_speed_decorator[True-False]	21.1929ms	20.8887ms	47.8727 Ops/s	48.0362 Ops/s	$\color{#d91a1a}-0.34\%$
test_vmap_transformer_speed_decorator[False-True]	20.9547ms	20.7035ms	48.3010 Ops/s	48.5755 Ops/s	$\color{#d91a1a}-0.57\%$
test_vmap_transformer_speed_decorator[False-False]	20.8796ms	20.7259ms	48.2487 Ops/s	48.2625 Ops/s	$\color{#d91a1a}-0.03\%$
test_to_module_speed[True]	1.6514ms	1.4837ms	674.0051 Ops/s	670.0258 Ops/s	$\color{#35bf28}+0.59\%$
test_to_module_speed[False]	1.5627ms	1.4712ms	679.7203 Ops/s	677.9734 Ops/s	$\color{#35bf28}+0.26\%$
test_tc_init	80.5510μs	52.4490μs	19.0661 KOps/s	19.3271 KOps/s	$\color{#d91a1a}-1.35\%$
test_tc_init_tensor_only	48.4010μs	14.9573μs	66.8568 KOps/s	66.7052 KOps/s	$\color{#35bf28}+0.23\%$
test_tc_init_nested	0.1361ms	0.1054ms	9.4910 KOps/s	9.6992 KOps/s	$\color{#d91a1a}-2.15\%$
test_tc_first_layer_tensor	16.2810μs	1.8043μs	554.2213 KOps/s	555.5795 KOps/s	$\color{#d91a1a}-0.24\%$
test_tc_first_layer_tensor_only	4.2571μs	0.7053μs	1.4178 MOps/s	1.4114 MOps/s	$\color{#35bf28}+0.45\%$
test_tc_first_layer_tensor_set	32.1410μs	4.2138μs	237.3159 KOps/s	234.5683 KOps/s	$\color{#35bf28}+1.17\%$
test_tc_first_layer_tensor_only_set	24.5005μs	3.0635μs	326.4195 KOps/s	325.4673 KOps/s	$\color{#35bf28}+0.29\%$
test_tc_first_layer_nontensor	30.2500μs	6.0395μs	165.5775 KOps/s	162.9135 KOps/s	$\color{#35bf28}+1.64\%$
test_tc_second_layer_tensor	27.5000μs	4.3378μs	230.5323 KOps/s	231.0323 KOps/s	$\color{#d91a1a}-0.22\%$
test_tc_second_layer_nontensor	41.8910μs	8.5649μs	116.7554 KOps/s	116.2420 KOps/s	$\color{#35bf28}+0.44\%$
test_unbind	0.2564s	16.1047ms	62.0936 Ops/s	63.1422 Ops/s	$\color{#d91a1a}-1.66\%$
test_full_like	4.4932ms	4.3768ms	228.4785 Ops/s	222.6022 Ops/s	$\color{#35bf28}+2.64\%$
test_zeros_like	4.4628ms	4.3611ms	229.2986 Ops/s	223.1669 Ops/s	$\color{#35bf28}+2.75\%$
test_ones_like	5.0356ms	4.4025ms	227.1414 Ops/s	226.7161 Ops/s	$\color{#35bf28}+0.19\%$
test_clone	6.5554ms	6.4023ms	156.1929 Ops/s	103.3493 Ops/s	$\textbf{\color{#35bf28}+51.13\%}$
test_squeeze	0.2001ms	14.3693μs	69.5929 KOps/s	69.5834 KOps/s	$\color{#35bf28}+0.01\%$
test_unsqueeze	0.1516ms	0.1144ms	8.7434 KOps/s	9.0802 KOps/s	$\color{#d91a1a}-3.71\%$
test_split	0.3603ms	0.1870ms	5.3474 KOps/s	5.3797 KOps/s	$\color{#d91a1a}-0.60\%$
test_permute	0.2853ms	0.2110ms	4.7400 KOps/s	4.9360 KOps/s	$\color{#d91a1a}-3.97\%$
test_stack	43.2234ms	42.9533ms	23.2811 Ops/s	19.0580 Ops/s	$\textbf{\color{#35bf28}+22.16\%}$
test_cat	43.0785ms	42.8632ms	23.3300 Ops/s	19.0696 Ops/s	$\textbf{\color{#35bf28}+22.34\%}$

[vmoens]

unrelated failing tests