[BugFix] Limit number of threads in workers for .map()

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 124. Improved: $\large\color{#35bf28}37$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	48.2410μs	15.9998μs	62.5006 KOps/s	54.7986 KOps/s	$\textbf{\color{#35bf28}+14.06\%}$
test_plain_set_stack_nested	0.1782ms	0.1419ms	7.0451 KOps/s	6.4607 KOps/s	$\textbf{\color{#35bf28}+9.05\%}$
test_plain_set_nested_inplace	62.5370μs	18.0239μs	55.4820 KOps/s	50.6606 KOps/s	$\textbf{\color{#35bf28}+9.52\%}$
test_plain_set_stack_nested_inplace	0.2961ms	0.1778ms	5.6242 KOps/s	5.1307 KOps/s	$\textbf{\color{#35bf28}+9.62\%}$
test_items	12.9250μs	2.3710μs	421.7685 KOps/s	376.7549 KOps/s	$\textbf{\color{#35bf28}+11.95\%}$
test_items_nested	0.4645ms	0.2716ms	3.6817 KOps/s	3.6405 KOps/s	$\color{#35bf28}+1.13\%$
test_items_nested_locked	0.4906ms	0.2699ms	3.7047 KOps/s	3.6342 KOps/s	$\color{#35bf28}+1.94\%$
test_items_nested_leaf	0.5221ms	0.1666ms	6.0022 KOps/s	5.9451 KOps/s	$\color{#35bf28}+0.96\%$
test_items_stack_nested	2.0924ms	1.3196ms	757.8235 Ops/s	728.7630 Ops/s	$\color{#35bf28}+3.99\%$
test_items_stack_nested_leaf	2.1966ms	1.2056ms	829.4341 Ops/s	812.4826 Ops/s	$\color{#35bf28}+2.09\%$
test_items_stack_nested_locked	1.4740ms	0.8639ms	1.1576 KOps/s	1.1169 KOps/s	$\color{#35bf28}+3.64\%$
test_keys	19.4970μs	3.9091μs	255.8163 KOps/s	258.3890 KOps/s	$\color{#d91a1a}-1.00\%$
test_keys_nested	51.3228ms	0.1583ms	6.3175 KOps/s	6.7278 KOps/s	$\textbf{\color{#d91a1a}-6.10\%}$
test_keys_nested_locked	0.2923ms	0.1525ms	6.5592 KOps/s	6.5653 KOps/s	$\color{#d91a1a}-0.09\%$
test_keys_nested_leaf	0.2384ms	0.1316ms	7.5981 KOps/s	7.5870 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_stack_nested	1.9761ms	1.2599ms	793.7287 Ops/s	754.9971 Ops/s	$\textbf{\color{#35bf28}+5.13\%}$
test_keys_stack_nested_leaf	2.1421ms	1.2516ms	798.9588 Ops/s	754.6222 Ops/s	$\textbf{\color{#35bf28}+5.88\%}$
test_keys_stack_nested_locked	1.3308ms	0.7871ms	1.2704 KOps/s	1.1881 KOps/s	$\textbf{\color{#35bf28}+6.93\%}$
test_values	9.7233μs	1.1605μs	861.7289 KOps/s	860.0241 KOps/s	$\color{#35bf28}+0.20\%$
test_values_nested	92.5600μs	52.4736μs	19.0572 KOps/s	19.6201 KOps/s	$\color{#d91a1a}-2.87\%$
test_values_nested_locked	0.1076ms	52.1911μs	19.1604 KOps/s	19.7165 KOps/s	$\color{#d91a1a}-2.82\%$
test_values_nested_leaf	0.1137ms	46.6121μs	21.4537 KOps/s	21.9615 KOps/s	$\color{#d91a1a}-2.31\%$
test_values_stack_nested	1.2691ms	1.0438ms	958.0660 Ops/s	937.8966 Ops/s	$\color{#35bf28}+2.15\%$
test_values_stack_nested_leaf	1.1792ms	1.0286ms	972.1581 Ops/s	943.1555 Ops/s	$\color{#35bf28}+3.08\%$
test_values_stack_nested_locked	0.7784ms	0.5968ms	1.6755 KOps/s	1.6475 KOps/s	$\color{#35bf28}+1.70\%$
test_membership	11.1900μs	1.3365μs	748.2395 KOps/s	744.6730 KOps/s	$\color{#35bf28}+0.48\%$
test_membership_nested	26.6800μs	3.4621μs	288.8385 KOps/s	282.2811 KOps/s	$\color{#35bf28}+2.32\%$
test_membership_nested_leaf	20.1080μs	3.4619μs	288.8612 KOps/s	274.8820 KOps/s	$\textbf{\color{#35bf28}+5.09\%}$
test_membership_stacked_nested	55.9940μs	11.7174μs	85.3432 KOps/s	78.1193 KOps/s	$\textbf{\color{#35bf28}+9.25\%}$
test_membership_stacked_nested_leaf	34.0240μs	11.8067μs	84.6980 KOps/s	73.6338 KOps/s	$\textbf{\color{#35bf28}+15.03\%}$
test_membership_nested_last	47.0380μs	6.6400μs	150.6033 KOps/s	149.9788 KOps/s	$\color{#35bf28}+0.42\%$
test_membership_nested_leaf_last	49.9730μs	6.7281μs	148.6308 KOps/s	147.9757 KOps/s	$\color{#35bf28}+0.44\%$
test_membership_stacked_nested_last	0.3139ms	0.1747ms	5.7254 KOps/s	5.5477 KOps/s	$\color{#35bf28}+3.20\%$
test_membership_stacked_nested_leaf_last	0.2810ms	14.5718μs	68.6259 KOps/s	66.0890 KOps/s	$\color{#35bf28}+3.84\%$
test_nested_getleaf	45.7050μs	10.4908μs	95.3213 KOps/s	93.1436 KOps/s	$\color{#35bf28}+2.34\%$
test_nested_get	30.7270μs	10.0350μs	99.6515 KOps/s	98.1711 KOps/s	$\color{#35bf28}+1.51\%$
test_stacked_getleaf	0.6153ms	0.3985ms	2.5092 KOps/s	2.4435 KOps/s	$\color{#35bf28}+2.69\%$
test_stacked_get	0.5686ms	0.3607ms	2.7722 KOps/s	2.6842 KOps/s	$\color{#35bf28}+3.28\%$
test_nested_getitemleaf	42.4600μs	11.8930μs	84.0834 KOps/s	83.4490 KOps/s	$\color{#35bf28}+0.76\%$
test_nested_getitem	52.1470μs	11.4404μs	87.4097 KOps/s	86.8074 KOps/s	$\color{#35bf28}+0.69\%$
test_stacked_getitemleaf	0.9166ms	0.4001ms	2.4995 KOps/s	2.4248 KOps/s	$\color{#35bf28}+3.08\%$
test_stacked_getitem	0.5631ms	0.3703ms	2.7008 KOps/s	2.6843 KOps/s	$\color{#35bf28}+0.62\%$
test_lock_nested	0.6990ms	0.3328ms	3.0045 KOps/s	2.9383 KOps/s	$\color{#35bf28}+2.25\%$
test_lock_stack_nested	70.7185ms	5.2704ms	189.7403 Ops/s	186.1908 Ops/s	$\color{#35bf28}+1.91\%$
test_unlock_nested	0.7455ms	0.3334ms	2.9991 KOps/s	2.4932 KOps/s	$\textbf{\color{#35bf28}+20.29\%}$
test_unlock_stack_nested	70.8509ms	5.3876ms	185.6123 Ops/s	179.8277 Ops/s	$\color{#35bf28}+3.22\%$
test_flatten_speed	0.7078ms	0.3667ms	2.7269 KOps/s	2.7104 KOps/s	$\color{#35bf28}+0.61\%$
test_unflatten_speed	0.5337ms	0.4624ms	2.1628 KOps/s	2.1183 KOps/s	$\color{#35bf28}+2.10\%$
test_common_ops	1.1489ms	0.6287ms	1.5906 KOps/s	1.4745 KOps/s	$\textbf{\color{#35bf28}+7.87\%}$
test_creation	17.9440μs	1.9088μs	523.8922 KOps/s	531.2583 KOps/s	$\color{#d91a1a}-1.39\%$
test_creation_empty	23.8750μs	7.7038μs	129.8056 KOps/s	111.0733 KOps/s	$\textbf{\color{#35bf28}+16.86\%}$
test_creation_nested_1	31.8100μs	10.3712μs	96.4210 KOps/s	85.6024 KOps/s	$\textbf{\color{#35bf28}+12.64\%}$
test_creation_nested_2	46.7370μs	13.5295μs	73.9128 KOps/s	66.8783 KOps/s	$\textbf{\color{#35bf28}+10.52\%}$
test_clone	88.9960μs	12.8792μs	77.6445 KOps/s	76.4470 KOps/s	$\color{#35bf28}+1.57\%$
test_getitem[int]	49.8630μs	11.0124μs	90.8067 KOps/s	90.4465 KOps/s	$\color{#35bf28}+0.40\%$
test_getitem[slice_int]	59.6120μs	21.9455μs	45.5675 KOps/s	43.4093 KOps/s	$\color{#35bf28}+4.97\%$
test_getitem[range]	90.2490μs	39.6951μs	25.1920 KOps/s	23.2619 KOps/s	$\textbf{\color{#35bf28}+8.30\%}$
test_getitem[tuple]	51.5570μs	17.8465μs	56.0333 KOps/s	54.3051 KOps/s	$\color{#35bf28}+3.18\%$
test_getitem[list]	0.3881ms	34.8161μs	28.7223 KOps/s	26.7472 KOps/s	$\textbf{\color{#35bf28}+7.38\%}$
test_setitem_dim[int]	72.5050μs	26.9957μs	37.0429 KOps/s	33.3409 KOps/s	$\textbf{\color{#35bf28}+11.10\%}$
test_setitem_dim[slice_int]	87.8150μs	51.5211μs	19.4095 KOps/s	17.3795 KOps/s	$\textbf{\color{#35bf28}+11.68\%}$
test_setitem_dim[range]	0.1361ms	68.9732μs	14.4984 KOps/s	13.1685 KOps/s	$\textbf{\color{#35bf28}+10.10\%}$
test_setitem_dim[tuple]	76.4430μs	41.4102μs	24.1486 KOps/s	21.6863 KOps/s	$\textbf{\color{#35bf28}+11.35\%}$
test_setitem	0.1286ms	17.4908μs	57.1730 KOps/s	52.4268 KOps/s	$\textbf{\color{#35bf28}+9.05\%}$
test_set	86.3620μs	16.8985μs	59.1769 KOps/s	54.0410 KOps/s	$\textbf{\color{#35bf28}+9.50\%}$
test_set_shared	3.0789ms	0.1447ms	6.9128 KOps/s	6.9692 KOps/s	$\color{#d91a1a}-0.81\%$
test_update	93.0540μs	18.7601μs	53.3047 KOps/s	47.4821 KOps/s	$\textbf{\color{#35bf28}+12.26\%}$
test_update_nested	0.1004ms	25.8360μs	38.7057 KOps/s	34.6340 KOps/s	$\textbf{\color{#35bf28}+11.76\%}$
test_set_nested	78.9680μs	18.7825μs	53.2410 KOps/s	49.0411 KOps/s	$\textbf{\color{#35bf28}+8.56\%}$
test_set_nested_new	98.9550μs	22.9877μs	43.5015 KOps/s	40.9623 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_select	0.1090ms	36.3046μs	27.5447 KOps/s	25.9725 KOps/s	$\textbf{\color{#35bf28}+6.05\%}$
test_select_nested	0.1208ms	57.7477μs	17.3167 KOps/s	17.3298 KOps/s	$\color{#d91a1a}-0.08\%$
test_exclude_nested	0.2564ms	0.1173ms	8.5234 KOps/s	8.3383 KOps/s	$\color{#35bf28}+2.22\%$
test_empty[True]	0.4862ms	0.4096ms	2.4412 KOps/s	2.3703 KOps/s	$\color{#35bf28}+2.99\%$
test_empty[False]	9.0610μs	1.0935μs	914.5257 KOps/s	883.2895 KOps/s	$\color{#35bf28}+3.54\%$
test_unbind_speed	0.3558ms	0.2424ms	4.1246 KOps/s	3.9851 KOps/s	$\color{#35bf28}+3.50\%$
test_unbind_speed_stack0	70.3489ms	3.2225ms	310.3138 Ops/s	330.2691 Ops/s	$\textbf{\color{#d91a1a}-6.04\%}$
test_unbind_speed_stack1	35.6160μs	1.9100μs	523.5516 KOps/s	508.7441 KOps/s	$\color{#35bf28}+2.91\%$
test_split	66.6641ms	1.6245ms	615.5661 Ops/s	629.2814 Ops/s	$\color{#d91a1a}-2.18\%$
test_chunk	68.6627ms	1.5643ms	639.2590 Ops/s	639.1402 Ops/s	$\color{#35bf28}+0.02\%$
test_creation[device0]	0.2117ms	0.1004ms	9.9579 KOps/s	10.0033 KOps/s	$\color{#d91a1a}-0.45\%$
test_creation_from_tensor	3.5889ms	81.0833μs	12.3330 KOps/s	12.2684 KOps/s	$\color{#35bf28}+0.53\%$
test_add_one[memmap_tensor0]	0.3647ms	5.3486μs	186.9634 KOps/s	180.4440 KOps/s	$\color{#35bf28}+3.61\%$
test_contiguous[memmap_tensor0]	24.1160μs	0.6322μs	1.5817 MOps/s	1.5651 MOps/s	$\color{#35bf28}+1.06\%$
test_stack[memmap_tensor0]	68.9790μs	3.4318μs	291.3906 KOps/s	277.5042 KOps/s	$\textbf{\color{#35bf28}+5.00\%}$
test_memmaptd_index	0.8507ms	0.2198ms	4.5498 KOps/s	4.4305 KOps/s	$\color{#35bf28}+2.69\%$
test_memmaptd_index_astensor	0.5201ms	0.2779ms	3.5984 KOps/s	3.4616 KOps/s	$\color{#35bf28}+3.95\%$
test_memmaptd_index_op	1.1999ms	0.5317ms	1.8807 KOps/s	1.8054 KOps/s	$\color{#35bf28}+4.17\%$
test_serialize_model	0.1692s	0.1043s	9.5841 Ops/s	9.0928 Ops/s	$\textbf{\color{#35bf28}+5.40\%}$
test_serialize_model_pickle	0.4555s	0.3790s	2.6388 Ops/s	2.6165 Ops/s	$\color{#35bf28}+0.85\%$
test_serialize_weights	0.1589s	0.1038s	9.6310 Ops/s	9.3916 Ops/s	$\color{#35bf28}+2.55\%$
test_serialize_weights_returnearly	0.3174s	0.1472s	6.7915 Ops/s	7.5604 Ops/s	$\textbf{\color{#d91a1a}-10.17\%}$
test_serialize_weights_pickle	0.8916s	0.5208s	1.9202 Ops/s	2.4169 Ops/s	$\textbf{\color{#d91a1a}-20.55\%}$
test_serialize_weights_filesystem	0.1484s	95.4235ms	10.4796 Ops/s	10.9124 Ops/s	$\color{#d91a1a}-3.97\%$
test_serialize_model_filesystem	99.9152ms	90.9125ms	10.9996 Ops/s	9.6846 Ops/s	$\textbf{\color{#35bf28}+13.58\%}$
test_reshape_pytree	68.0200μs	22.9886μs	43.4999 KOps/s	43.2926 KOps/s	$\color{#35bf28}+0.48\%$
test_reshape_td	71.2130μs	29.7901μs	33.5682 KOps/s	33.5645 KOps/s	$\color{#35bf28}+0.01\%$
test_view_pytree	60.0230μs	22.9333μs	43.6048 KOps/s	42.6318 KOps/s	$\color{#35bf28}+2.28\%$
test_view_td	38.6820μs	4.8373μs	206.7276 KOps/s	199.6755 KOps/s	$\color{#35bf28}+3.53\%$
test_unbind_pytree	86.7630μs	26.1760μs	38.2030 KOps/s	37.5254 KOps/s	$\color{#35bf28}+1.81\%$
test_unbind_td	0.4888ms	35.6288μs	28.0672 KOps/s	27.5371 KOps/s	$\color{#35bf28}+1.92\%$
test_split_pytree	81.1710μs	26.2772μs	38.0558 KOps/s	37.9291 KOps/s	$\color{#35bf28}+0.33\%$
test_split_td	0.1178ms	39.8682μs	25.0826 KOps/s	24.5838 KOps/s	$\color{#35bf28}+2.03\%$
test_add_pytree	81.9740μs	32.2038μs	31.0522 KOps/s	30.2658 KOps/s	$\color{#35bf28}+2.60\%$
test_add_td	0.1174ms	45.7201μs	21.8722 KOps/s	20.1138 KOps/s	$\textbf{\color{#35bf28}+8.74\%}$
test_distributed	0.1808ms	97.8180μs	10.2231 KOps/s	9.9139 KOps/s	$\color{#35bf28}+3.12\%$
test_tdmodule	0.1040ms	20.5742μs	48.6045 KOps/s	44.2514 KOps/s	$\textbf{\color{#35bf28}+9.84\%}$
test_tdmodule_dispatch	0.2583ms	40.3205μs	24.8013 KOps/s	23.5297 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_tdseq	46.4160μs	24.0803μs	41.5276 KOps/s	39.9060 KOps/s	$\color{#35bf28}+4.06\%$
test_tdseq_dispatch	0.1386ms	44.2809μs	22.5831 KOps/s	21.0886 KOps/s	$\textbf{\color{#35bf28}+7.09\%}$
test_instantiation_functorch	1.8414ms	1.3054ms	766.0606 Ops/s	758.9714 Ops/s	$\color{#35bf28}+0.93\%$
test_instantiation_td	1.5301ms	1.0126ms	987.5982 Ops/s	988.0826 Ops/s	$\color{#d91a1a}-0.05\%$
test_exec_functorch	0.2956ms	0.1565ms	6.3912 KOps/s	6.2830 KOps/s	$\color{#35bf28}+1.72\%$
test_exec_functional_call	0.2893ms	0.1468ms	6.8101 KOps/s	6.6942 KOps/s	$\color{#35bf28}+1.73\%$
test_exec_td	0.2829ms	0.1437ms	6.9583 KOps/s	6.7897 KOps/s	$\color{#35bf28}+2.48\%$
test_exec_td_decorator	0.8709ms	0.1804ms	5.5436 KOps/s	5.5794 KOps/s	$\color{#d91a1a}-0.64\%$
test_vmap_mlp_speed[True-True]	1.1845ms	0.8640ms	1.1573 KOps/s	1.1031 KOps/s	$\color{#35bf28}+4.91\%$
test_vmap_mlp_speed[True-False]	0.6725ms	0.4597ms	2.1755 KOps/s	2.1138 KOps/s	$\color{#35bf28}+2.92\%$
test_vmap_mlp_speed[False-True]	1.1628ms	0.7648ms	1.3075 KOps/s	1.2499 KOps/s	$\color{#35bf28}+4.61\%$
test_vmap_mlp_speed[False-False]	0.5364ms	0.3797ms	2.6335 KOps/s	2.5584 KOps/s	$\color{#35bf28}+2.93\%$
test_vmap_mlp_speed_decorator[True-True]	69.0421ms	2.3836ms	419.5325 Ops/s	420.6313 Ops/s	$\color{#d91a1a}-0.26\%$
test_vmap_mlp_speed_decorator[True-False]	0.8169ms	0.5156ms	1.9393 KOps/s	1.9015 KOps/s	$\color{#35bf28}+1.99\%$
test_vmap_mlp_speed_decorator[False-True]	2.3915ms	1.8185ms	549.8926 Ops/s	518.4389 Ops/s	$\textbf{\color{#35bf28}+6.07\%}$
test_vmap_mlp_speed_decorator[False-False]	0.7704ms	0.3982ms	2.5114 KOps/s	2.4879 KOps/s	$\color{#35bf28}+0.95\%$

[github-actions]

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

Total Benchmarks: 132. Improved: $\large\color{#35bf28}3$. Worsened: $\large\color{#d91a1a}18$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.6514ms	14.8121μs	67.5122 KOps/s	74.0097 KOps/s	$\textbf{\color{#d91a1a}-8.78\%}$
test_plain_set_stack_nested	0.1645ms	0.1214ms	8.2361 KOps/s	8.3208 KOps/s	$\color{#d91a1a}-1.02\%$
test_plain_set_nested_inplace	34.6320μs	16.1540μs	61.9041 KOps/s	67.2996 KOps/s	$\textbf{\color{#d91a1a}-8.02\%}$
test_plain_set_stack_nested_inplace	0.2166ms	0.1506ms	6.6404 KOps/s	6.6764 KOps/s	$\color{#d91a1a}-0.54\%$
test_items	20.9010μs	4.7669μs	209.7799 KOps/s	211.5981 KOps/s	$\color{#d91a1a}-0.86\%$
test_items_nested	0.4128ms	0.3452ms	2.8966 KOps/s	2.8759 KOps/s	$\color{#35bf28}+0.72\%$
test_items_nested_locked	0.4247ms	0.3465ms	2.8862 KOps/s	2.8583 KOps/s	$\color{#35bf28}+0.98\%$
test_items_nested_leaf	0.2818ms	0.2037ms	4.9102 KOps/s	4.8894 KOps/s	$\color{#35bf28}+0.43\%$
test_items_stack_nested	1.4481ms	1.3510ms	740.2167 Ops/s	739.3137 Ops/s	$\color{#35bf28}+0.12\%$
test_items_stack_nested_leaf	1.2683ms	1.1761ms	850.2512 Ops/s	845.0361 Ops/s	$\color{#35bf28}+0.62\%$
test_items_stack_nested_locked	1.9767ms	0.9283ms	1.0773 KOps/s	1.0825 KOps/s	$\color{#d91a1a}-0.48\%$
test_keys	25.7300μs	4.7685μs	209.7111 KOps/s	217.7950 KOps/s	$\color{#d91a1a}-3.71\%$
test_keys_nested	0.4810ms	95.0969μs	10.5156 KOps/s	10.5191 KOps/s	$\color{#d91a1a}-0.03\%$
test_keys_nested_locked	0.1233ms	98.8030μs	10.1211 KOps/s	10.1250 KOps/s	$\color{#d91a1a}-0.04\%$
test_keys_nested_leaf	0.1808ms	78.7316μs	12.7014 KOps/s	12.5779 KOps/s	$\color{#35bf28}+0.98\%$
test_keys_stack_nested	1.2743ms	1.1784ms	848.6267 Ops/s	852.0999 Ops/s	$\color{#d91a1a}-0.41\%$
test_keys_stack_nested_leaf	1.2687ms	1.1717ms	853.4668 Ops/s	858.0167 Ops/s	$\color{#d91a1a}-0.53\%$
test_keys_stack_nested_locked	0.8479ms	0.7507ms	1.3320 KOps/s	1.3296 KOps/s	$\color{#35bf28}+0.18\%$
test_values	8.5567μs	1.9076μs	524.2316 KOps/s	528.5651 KOps/s	$\color{#d91a1a}-0.82\%$
test_values_nested	67.4410μs	45.8945μs	21.7891 KOps/s	22.0990 KOps/s	$\color{#d91a1a}-1.40\%$
test_values_nested_locked	69.4010μs	47.6624μs	20.9809 KOps/s	21.0460 KOps/s	$\color{#d91a1a}-0.31\%$
test_values_nested_leaf	62.5710μs	39.7297μs	25.1701 KOps/s	25.1507 KOps/s	$\color{#35bf28}+0.08\%$
test_values_stack_nested	1.3992ms	0.9911ms	1.0090 KOps/s	996.8528 Ops/s	$\color{#35bf28}+1.22\%$
test_values_stack_nested_leaf	1.1085ms	0.9947ms	1.0054 KOps/s	1.0112 KOps/s	$\color{#d91a1a}-0.58\%$
test_values_stack_nested_locked	0.6824ms	0.6020ms	1.6610 KOps/s	1.6722 KOps/s	$\color{#d91a1a}-0.67\%$
test_membership	4.3462μs	0.9366μs	1.0677 MOps/s	1.0668 MOps/s	$\color{#35bf28}+0.08\%$
test_membership_nested	29.5500μs	2.8686μs	348.6017 KOps/s	350.8112 KOps/s	$\color{#d91a1a}-0.63\%$
test_membership_nested_leaf	26.2700μs	2.8736μs	347.9977 KOps/s	346.8671 KOps/s	$\color{#35bf28}+0.33\%$
test_membership_stacked_nested	54.5510μs	11.6325μs	85.9657 KOps/s	85.3086 KOps/s	$\color{#35bf28}+0.77\%$
test_membership_stacked_nested_leaf	30.5210μs	11.6283μs	85.9969 KOps/s	86.0874 KOps/s	$\color{#d91a1a}-0.11\%$
test_membership_nested_last	21.1110μs	5.2593μs	190.1397 KOps/s	191.7591 KOps/s	$\color{#d91a1a}-0.84\%$
test_membership_nested_leaf_last	20.7310μs	5.2640μs	189.9711 KOps/s	191.2049 KOps/s	$\color{#d91a1a}-0.65\%$
test_membership_stacked_nested_last	0.2423ms	0.1568ms	6.3775 KOps/s	6.4116 KOps/s	$\color{#d91a1a}-0.53\%$
test_membership_stacked_nested_leaf_last	39.1100μs	13.4596μs	74.2963 KOps/s	74.7339 KOps/s	$\color{#d91a1a}-0.59\%$
test_nested_getleaf	29.1500μs	8.4457μs	118.4029 KOps/s	119.0714 KOps/s	$\color{#d91a1a}-0.56\%$
test_nested_get	28.4500μs	7.9616μs	125.6028 KOps/s	126.6091 KOps/s	$\color{#d91a1a}-0.79\%$
test_stacked_getleaf	0.4052ms	0.3361ms	2.9756 KOps/s	2.9845 KOps/s	$\color{#d91a1a}-0.30\%$
test_stacked_get	0.3564ms	0.2994ms	3.3398 KOps/s	3.3452 KOps/s	$\color{#d91a1a}-0.16\%$
test_nested_getitemleaf	25.0000μs	9.8807μs	101.2076 KOps/s	101.3146 KOps/s	$\color{#d91a1a}-0.11\%$
test_nested_getitem	31.2300μs	9.3963μs	106.4253 KOps/s	106.4302 KOps/s	$-0.00\%$
test_stacked_getitemleaf	0.3932ms	0.3404ms	2.9381 KOps/s	2.9421 KOps/s	$\color{#d91a1a}-0.14\%$
test_stacked_getitem	0.3555ms	0.3013ms	3.3188 KOps/s	3.3089 KOps/s	$\color{#35bf28}+0.30\%$
test_lock_nested	0.7959ms	0.3601ms	2.7774 KOps/s	2.7667 KOps/s	$\color{#35bf28}+0.38\%$
test_lock_stack_nested	84.0660ms	6.3094ms	158.4943 Ops/s	159.0229 Ops/s	$\color{#d91a1a}-0.33\%$
test_unlock_nested	0.7883ms	0.3591ms	2.7848 KOps/s	2.3023 KOps/s	$\textbf{\color{#35bf28}+20.96\%}$
test_unlock_stack_nested	82.8834ms	6.4013ms	156.2183 Ops/s	157.3844 Ops/s	$\color{#d91a1a}-0.74\%$
test_flatten_speed	0.5255ms	0.2679ms	3.7326 KOps/s	3.7459 KOps/s	$\color{#d91a1a}-0.36\%$
test_unflatten_speed	0.4263ms	0.3637ms	2.7497 KOps/s	2.7468 KOps/s	$\color{#35bf28}+0.10\%$
test_common_ops	1.2675ms	0.6727ms	1.4865 KOps/s	1.6043 KOps/s	$\textbf{\color{#d91a1a}-7.34\%}$
test_creation	42.5610μs	1.5518μs	644.4102 KOps/s	630.0611 KOps/s	$\color{#35bf28}+2.28\%$
test_creation_empty	38.6800μs	10.3482μs	96.6352 KOps/s	126.6962 KOps/s	$\textbf{\color{#d91a1a}-23.73\%}$
test_creation_nested_1	31.3900μs	12.1126μs	82.5584 KOps/s	103.9835 KOps/s	$\textbf{\color{#d91a1a}-20.60\%}$
test_creation_nested_2	32.5500μs	14.5462μs	68.7464 KOps/s	82.8264 KOps/s	$\textbf{\color{#d91a1a}-17.00\%}$
test_clone	0.1590ms	14.9495μs	66.8920 KOps/s	68.7666 KOps/s	$\color{#d91a1a}-2.73\%$
test_getitem[int]	36.9310μs	11.5541μs	86.5492 KOps/s	86.5429 KOps/s	$+0.01\%$
test_getitem[slice_int]	45.2310μs	21.9761μs	45.5039 KOps/s	44.2766 KOps/s	$\color{#35bf28}+2.77\%$
test_getitem[range]	69.2410μs	38.4036μs	26.0392 KOps/s	25.8335 KOps/s	$\color{#35bf28}+0.80\%$
test_getitem[tuple]	45.9900μs	19.5334μs	51.1943 KOps/s	50.4676 KOps/s	$\color{#35bf28}+1.44\%$
test_getitem[list]	71.1010μs	34.2663μs	29.1832 KOps/s	26.9678 KOps/s	$\textbf{\color{#35bf28}+8.22\%}$
test_setitem_dim[int]	55.8610μs	30.7432μs	32.5275 KOps/s	34.6015 KOps/s	$\textbf{\color{#d91a1a}-5.99\%}$
test_setitem_dim[slice_int]	75.8510μs	52.4024μs	19.0831 KOps/s	18.9924 KOps/s	$\color{#35bf28}+0.48\%$
test_setitem_dim[range]	0.1031ms	66.5795μs	15.0196 KOps/s	14.6027 KOps/s	$\color{#35bf28}+2.86\%$
test_setitem_dim[tuple]	76.6810μs	45.3207μs	22.0650 KOps/s	22.4092 KOps/s	$\color{#d91a1a}-1.54\%$
test_setitem	0.1594ms	20.7362μs	48.2247 KOps/s	47.6637 KOps/s	$\color{#35bf28}+1.18\%$
test_set	0.1389ms	20.1837μs	49.5450 KOps/s	50.7756 KOps/s	$\color{#d91a1a}-2.42\%$
test_set_shared	2.8093ms	0.1077ms	9.2840 KOps/s	9.2764 KOps/s	$\color{#35bf28}+0.08\%$
test_update	0.1166ms	23.5378μs	42.4848 KOps/s	46.7591 KOps/s	$\textbf{\color{#d91a1a}-9.14\%}$
test_update_nested	0.1745ms	30.4847μs	32.8034 KOps/s	34.5282 KOps/s	$\color{#d91a1a}-5.00\%$
test_set_nested	66.3210μs	21.5678μs	46.3654 KOps/s	49.0645 KOps/s	$\textbf{\color{#d91a1a}-5.50\%}$
test_set_nested_new	0.1340ms	24.5943μs	40.6598 KOps/s	42.0934 KOps/s	$\color{#d91a1a}-3.41\%$
test_select	62.8110μs	37.3846μs	26.7490 KOps/s	26.5357 KOps/s	$\color{#35bf28}+0.80\%$
test_select_nested	71.5920μs	53.0315μs	18.8567 KOps/s	18.8053 KOps/s	$\color{#35bf28}+0.27\%$
test_exclude_nested	0.1472ms	0.1127ms	8.8751 KOps/s	8.8579 KOps/s	$\color{#35bf28}+0.19\%$
test_empty[True]	0.4540ms	0.3895ms	2.5672 KOps/s	2.5907 KOps/s	$\color{#d91a1a}-0.91\%$
test_empty[False]	2.8470μs	0.8675μs	1.1528 MOps/s	1.1571 MOps/s	$\color{#d91a1a}-0.38\%$
test_to	73.8910μs	55.2708μs	18.0927 KOps/s	18.0021 KOps/s	$\color{#35bf28}+0.50\%$
test_to_nonblocking	84.1110μs	35.6125μs	28.0800 KOps/s	28.1722 KOps/s	$\color{#d91a1a}-0.33\%$
test_unbind_speed	0.3458ms	0.2775ms	3.6031 KOps/s	3.6748 KOps/s	$\color{#d91a1a}-1.95\%$
test_unbind_speed_stack0	81.1868ms	3.7485ms	266.7717 Ops/s	263.7376 Ops/s	$\color{#35bf28}+1.15\%$
test_unbind_speed_stack1	46.1120μs	1.8795μs	532.0616 KOps/s	541.8418 KOps/s	$\color{#d91a1a}-1.80\%$
test_split	2.2329ms	1.6311ms	613.0752 Ops/s	557.4659 Ops/s	$\textbf{\color{#35bf28}+9.98\%}$
test_chunk	76.6039ms	1.7542ms	570.0626 Ops/s	576.6036 Ops/s	$\color{#d91a1a}-1.13\%$
test_creation[device0]	0.1399ms	74.2811μs	13.4624 KOps/s	13.5914 KOps/s	$\color{#d91a1a}-0.95\%$
test_creation_from_tensor	0.1265ms	55.1898μs	18.1193 KOps/s	18.4123 KOps/s	$\color{#d91a1a}-1.59\%$
test_add_one[memmap_tensor0]	0.1350ms	7.9930μs	125.1100 KOps/s	127.4107 KOps/s	$\color{#d91a1a}-1.81\%$
test_contiguous[memmap_tensor0]	11.8000μs	0.6440μs	1.5528 MOps/s	1.5420 MOps/s	$\color{#35bf28}+0.70\%$
test_stack[memmap_tensor0]	30.7900μs	4.8325μs	206.9306 KOps/s	208.4729 KOps/s	$\color{#d91a1a}-0.74\%$
test_memmaptd_index	1.0458ms	0.2747ms	3.6402 KOps/s	3.7177 KOps/s	$\color{#d91a1a}-2.08\%$
test_memmaptd_index_astensor	0.6353ms	0.3273ms	3.0555 KOps/s	3.0816 KOps/s	$\color{#d91a1a}-0.85\%$
test_memmaptd_index_op	1.0995ms	0.6825ms	1.4653 KOps/s	1.5731 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_serialize_model	0.1713s	98.6042ms	10.1416 Ops/s	10.6871 Ops/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_serialize_model_pickle	1.3537s	1.2373s	0.8082 Ops/s	0.8055 Ops/s	$\color{#35bf28}+0.33\%$
test_serialize_weights	0.1666s	96.3373ms	10.3802 Ops/s	10.0594 Ops/s	$\color{#35bf28}+3.19\%$
test_serialize_weights_returnearly	0.2599s	81.7552ms	12.2316 Ops/s	14.2611 Ops/s	$\textbf{\color{#d91a1a}-14.23\%}$
test_serialize_weights_pickle	1.3484s	1.2377s	0.8079 Ops/s	0.8083 Ops/s	$\color{#d91a1a}-0.04\%$
test_reshape_pytree	58.6810μs	25.6503μs	38.9860 KOps/s	39.5798 KOps/s	$\color{#d91a1a}-1.50\%$
test_reshape_td	0.2896ms	30.8126μs	32.4543 KOps/s	33.7042 KOps/s	$\color{#d91a1a}-3.71\%$
test_view_pytree	54.6500μs	25.1815μs	39.7117 KOps/s	40.2753 KOps/s	$\color{#d91a1a}-1.40\%$
test_view_td	27.1310μs	4.3147μs	231.7679 KOps/s	231.5113 KOps/s	$\color{#35bf28}+0.11\%$
test_unbind_pytree	0.2238ms	31.8164μs	31.4303 KOps/s	32.2123 KOps/s	$\color{#d91a1a}-2.43\%$
test_unbind_td	0.5065ms	42.2490μs	23.6692 KOps/s	24.3883 KOps/s	$\color{#d91a1a}-2.95\%$
test_split_pytree	53.7210μs	30.1316μs	33.1878 KOps/s	33.6857 KOps/s	$\color{#d91a1a}-1.48\%$
test_split_td	0.5422ms	40.8122μs	24.5025 KOps/s	24.5412 KOps/s	$\color{#d91a1a}-0.16\%$
test_add_pytree	66.6710μs	38.9536μs	25.6715 KOps/s	26.4233 KOps/s	$\color{#d91a1a}-2.84\%$
test_add_td	0.2785ms	56.3453μs	17.7477 KOps/s	20.0792 KOps/s	$\textbf{\color{#d91a1a}-11.61\%}$
test_distributed	1.9386ms	72.7804μs	13.7400 KOps/s	13.3631 KOps/s	$\color{#35bf28}+2.82\%$
test_tdmodule	40.5310μs	19.2959μs	51.8245 KOps/s	56.4623 KOps/s	$\textbf{\color{#d91a1a}-8.21\%}$
test_tdmodule_dispatch	0.2391ms	40.6789μs	24.5828 KOps/s	27.7649 KOps/s	$\textbf{\color{#d91a1a}-11.46\%}$
test_tdseq	38.0600μs	22.4111μs	44.6208 KOps/s	48.3888 KOps/s	$\textbf{\color{#d91a1a}-7.79\%}$
test_tdseq_dispatch	71.9610μs	42.5842μs	23.4829 KOps/s	25.5123 KOps/s	$\textbf{\color{#d91a1a}-7.95\%}$
test_instantiation_functorch	1.8622ms	1.7251ms	579.6725 Ops/s	579.2599 Ops/s	$\color{#35bf28}+0.07\%$
test_instantiation_td	1.8350ms	1.1959ms	836.1831 Ops/s	842.3804 Ops/s	$\color{#d91a1a}-0.74\%$
test_exec_functorch	0.2068ms	0.1676ms	5.9661 KOps/s	5.9259 KOps/s	$\color{#35bf28}+0.68\%$
test_exec_functional_call	0.2328ms	0.1707ms	5.8584 KOps/s	5.9465 KOps/s	$\color{#d91a1a}-1.48\%$
test_exec_td	0.2156ms	0.1652ms	6.0528 KOps/s	6.2574 KOps/s	$\color{#d91a1a}-3.27\%$
test_exec_td_decorator	0.8846ms	0.1981ms	5.0475 KOps/s	5.0605 KOps/s	$\color{#d91a1a}-0.26\%$
test_vmap_mlp_speed[True-True]	1.1771ms	1.0918ms	915.9018 Ops/s	933.9979 Ops/s	$\color{#d91a1a}-1.94\%$
test_vmap_mlp_speed[True-False]	0.7217ms	0.6320ms	1.5823 KOps/s	1.6116 KOps/s	$\color{#d91a1a}-1.82\%$
test_vmap_mlp_speed[False-True]	1.0831ms	0.9989ms	1.0011 KOps/s	1.0128 KOps/s	$\color{#d91a1a}-1.15\%$
test_vmap_mlp_speed[False-False]	0.6352ms	0.5616ms	1.7807 KOps/s	1.8149 KOps/s	$\color{#d91a1a}-1.88\%$
test_vmap_mlp_speed_decorator[True-True]	2.9707ms	2.4401ms	409.8271 Ops/s	422.3000 Ops/s	$\color{#d91a1a}-2.95\%$
test_vmap_mlp_speed_decorator[True-False]	1.0849ms	0.6749ms	1.4818 KOps/s	1.4864 KOps/s	$\color{#d91a1a}-0.31\%$
test_vmap_mlp_speed_decorator[False-True]	2.4839ms	2.0341ms	491.6083 Ops/s	497.7863 Ops/s	$\color{#d91a1a}-1.24\%$
test_vmap_mlp_speed_decorator[False-False]	1.0333ms	0.5746ms	1.7403 KOps/s	1.7553 KOps/s	$\color{#d91a1a}-0.86\%$
test_vmap_transformer_speed[True-True]	13.3241ms	12.8107ms	78.0599 Ops/s	78.6799 Ops/s	$\color{#d91a1a}-0.79\%$
test_vmap_transformer_speed[True-False]	8.6482ms	8.3282ms	120.0746 Ops/s	120.3795 Ops/s	$\color{#d91a1a}-0.25\%$
test_vmap_transformer_speed[False-True]	13.1566ms	12.7216ms	78.6062 Ops/s	79.7476 Ops/s	$\color{#d91a1a}-1.43\%$
test_vmap_transformer_speed[False-False]	8.5945ms	8.2315ms	121.4846 Ops/s	122.0227 Ops/s	$\color{#d91a1a}-0.44\%$
test_vmap_transformer_speed_decorator[True-True]	78.1087ms	75.4882ms	13.2471 Ops/s	13.4497 Ops/s	$\color{#d91a1a}-1.51\%$
test_vmap_transformer_speed_decorator[True-False]	22.3113ms	20.2458ms	49.3930 Ops/s	49.9900 Ops/s	$\color{#d91a1a}-1.19\%$
test_vmap_transformer_speed_decorator[False-True]	69.6092ms	68.7264ms	14.5504 Ops/s	14.8138 Ops/s	$\color{#d91a1a}-1.78\%$
test_vmap_transformer_speed_decorator[False-False]	0.1144s	21.6380ms	46.2150 Ops/s	51.0099 Ops/s	$\textbf{\color{#d91a1a}-9.40\%}$