[Feature] Weakref for unlocking tds

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 113. Improved: $\large\color{#35bf28}31$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	44.6330μs	15.6949μs	63.7148 KOps/s	62.7375 KOps/s	$\color{#35bf28}+1.56\%$
test_plain_set_stack_nested	0.2640ms	0.1421ms	7.0374 KOps/s	6.9127 KOps/s	$\color{#35bf28}+1.80\%$
test_plain_set_nested_inplace	52.4380μs	17.9004μs	55.8648 KOps/s	55.1978 KOps/s	$\color{#35bf28}+1.21\%$
test_plain_set_stack_nested_inplace	0.3489ms	0.1778ms	5.6244 KOps/s	5.6387 KOps/s	$\color{#d91a1a}-0.25\%$
test_items	33.5130μs	2.5984μs	384.8462 KOps/s	414.5071 KOps/s	$\textbf{\color{#d91a1a}-7.16\%}$
test_items_nested	0.4807ms	0.2767ms	3.6137 KOps/s	3.7050 KOps/s	$\color{#d91a1a}-2.46\%$
test_items_nested_locked	0.3487ms	0.2790ms	3.5841 KOps/s	3.7388 KOps/s	$\color{#d91a1a}-4.14\%$
test_items_nested_leaf	0.5597ms	0.1691ms	5.9123 KOps/s	6.1599 KOps/s	$\color{#d91a1a}-4.02\%$
test_items_stack_nested	1.9235ms	1.4859ms	672.9881 Ops/s	669.1528 Ops/s	$\color{#35bf28}+0.57\%$
test_items_stack_nested_leaf	2.0437ms	1.3446ms	743.7139 Ops/s	734.3787 Ops/s	$\color{#35bf28}+1.27\%$
test_items_stack_nested_locked	1.8011ms	0.7736ms	1.2926 KOps/s	1.3074 KOps/s	$\color{#d91a1a}-1.13\%$
test_keys	14.4370μs	3.8559μs	259.3413 KOps/s	256.8981 KOps/s	$\color{#35bf28}+0.95\%$
test_keys_nested	0.5172ms	0.1409ms	7.0987 KOps/s	6.6860 KOps/s	$\textbf{\color{#35bf28}+6.17\%}$
test_keys_nested_locked	0.1927ms	0.1397ms	7.1573 KOps/s	7.1563 KOps/s	$\color{#35bf28}+0.01\%$
test_keys_nested_leaf	0.2778ms	0.1402ms	7.1332 KOps/s	7.0850 KOps/s	$\color{#35bf28}+0.68\%$
test_keys_stack_nested	2.1601ms	1.4214ms	703.5276 Ops/s	695.9553 Ops/s	$\color{#35bf28}+1.09\%$
test_keys_stack_nested_leaf	2.1588ms	1.4188ms	704.8384 Ops/s	702.7179 Ops/s	$\color{#35bf28}+0.30\%$
test_keys_stack_nested_locked	1.0866ms	0.6842ms	1.4615 KOps/s	1.4680 KOps/s	$\color{#d91a1a}-0.44\%$
test_values	10.2108μs	1.1519μs	868.0985 KOps/s	867.5810 KOps/s	$\color{#35bf28}+0.06\%$
test_values_nested	97.3110μs	49.4640μs	20.2167 KOps/s	20.1619 KOps/s	$\color{#35bf28}+0.27\%$
test_values_nested_locked	0.1373ms	51.6213μs	19.3719 KOps/s	20.2466 KOps/s	$\color{#d91a1a}-4.32\%$
test_values_nested_leaf	95.5280μs	44.0281μs	22.7128 KOps/s	22.3491 KOps/s	$\color{#35bf28}+1.63\%$
test_values_stack_nested	2.4841ms	1.2180ms	820.9870 Ops/s	832.0677 Ops/s	$\color{#d91a1a}-1.33\%$
test_values_stack_nested_leaf	2.4862ms	1.2010ms	832.6388 Ops/s	839.1190 Ops/s	$\color{#d91a1a}-0.77\%$
test_values_stack_nested_locked	1.0461ms	0.5228ms	1.9128 KOps/s	1.9605 KOps/s	$\color{#d91a1a}-2.43\%$
test_membership	22.3620μs	1.3626μs	733.9076 KOps/s	737.8256 KOps/s	$\color{#d91a1a}-0.53\%$
test_membership_nested	26.6800μs	2.8507μs	350.7963 KOps/s	351.0434 KOps/s	$\color{#d91a1a}-0.07\%$
test_membership_nested_leaf	39.3940μs	2.8595μs	349.7079 KOps/s	352.8535 KOps/s	$\color{#d91a1a}-0.89\%$
test_membership_stacked_nested	52.1170μs	11.7096μs	85.4001 KOps/s	85.0173 KOps/s	$\color{#35bf28}+0.45\%$
test_membership_stacked_nested_leaf	40.9570μs	11.6725μs	85.6714 KOps/s	84.6896 KOps/s	$\color{#35bf28}+1.16\%$
test_membership_nested_last	18.7750μs	5.9497μs	168.0746 KOps/s	161.4871 KOps/s	$\color{#35bf28}+4.08\%$
test_membership_nested_leaf_last	43.1900μs	6.0400μs	165.5636 KOps/s	167.3318 KOps/s	$\color{#d91a1a}-1.06\%$
test_membership_stacked_nested_last	0.2362ms	0.1668ms	5.9965 KOps/s	5.9380 KOps/s	$\color{#35bf28}+0.99\%$
test_membership_stacked_nested_leaf_last	38.0810μs	13.7370μs	72.7963 KOps/s	72.6581 KOps/s	$\color{#35bf28}+0.19\%$
test_nested_getleaf	49.0120μs	10.8907μs	91.8212 KOps/s	94.2273 KOps/s	$\color{#d91a1a}-2.55\%$
test_nested_get	29.7250μs	10.2917μs	97.1655 KOps/s	98.6443 KOps/s	$\color{#d91a1a}-1.50\%$
test_stacked_getleaf	1.0957ms	0.6365ms	1.5712 KOps/s	1.5500 KOps/s	$\color{#35bf28}+1.37\%$
test_stacked_get	0.7282ms	0.6024ms	1.6601 KOps/s	1.6163 KOps/s	$\color{#35bf28}+2.71\%$
test_nested_getitemleaf	32.3500μs	11.0069μs	90.8523 KOps/s	93.4337 KOps/s	$\color{#d91a1a}-2.76\%$
test_nested_getitem	36.9890μs	10.4809μs	95.4116 KOps/s	97.4099 KOps/s	$\color{#d91a1a}-2.05\%$
test_stacked_getitemleaf	1.3457ms	0.6400ms	1.5625 KOps/s	1.5435 KOps/s	$\color{#35bf28}+1.23\%$
test_stacked_getitem	0.7782ms	0.6052ms	1.6524 KOps/s	1.6164 KOps/s	$\color{#35bf28}+2.23\%$
test_lock_nested	55.9243ms	0.4715ms	2.1207 KOps/s	1.7532 KOps/s	$\textbf{\color{#35bf28}+20.96\%}$
test_lock_stack_nested	71.1813ms	6.4306ms	155.5076 Ops/s	194.7076 Ops/s	$\textbf{\color{#d91a1a}-20.13\%}$
test_unlock_nested	1.0946ms	0.4282ms	2.3355 KOps/s	2.2183 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_unlock_stack_nested	75.7449ms	6.1912ms	161.5185 Ops/s	141.9828 Ops/s	$\textbf{\color{#35bf28}+13.76\%}$
test_flatten_speed	0.3474ms	0.2679ms	3.7333 KOps/s	3.7229 KOps/s	$\color{#35bf28}+0.28\%$
test_unflatten_speed	0.7638ms	0.4549ms	2.1983 KOps/s	2.1530 KOps/s	$\color{#35bf28}+2.10\%$
test_common_ops	1.2004ms	0.6457ms	1.5487 KOps/s	1.4891 KOps/s	$\color{#35bf28}+4.00\%$
test_creation	17.6630μs	2.0157μs	496.0936 KOps/s	406.6412 KOps/s	$\textbf{\color{#35bf28}+22.00\%}$
test_creation_empty	24.3350μs	7.5692μs	132.1141 KOps/s	121.0339 KOps/s	$\textbf{\color{#35bf28}+9.15\%}$
test_creation_nested_1	30.5370μs	10.4050μs	96.1074 KOps/s	87.0459 KOps/s	$\textbf{\color{#35bf28}+10.41\%}$
test_creation_nested_2	40.1750μs	14.0774μs	71.0358 KOps/s	65.3953 KOps/s	$\textbf{\color{#35bf28}+8.63\%}$
test_clone	0.3264ms	12.2028μs	81.9485 KOps/s	72.8508 KOps/s	$\textbf{\color{#35bf28}+12.49\%}$
test_getitem[int]	35.6160μs	12.2242μs	81.8049 KOps/s	75.1274 KOps/s	$\textbf{\color{#35bf28}+8.89\%}$
test_getitem[slice_int]	69.2390μs	23.9282μs	41.7917 KOps/s	38.7669 KOps/s	$\textbf{\color{#35bf28}+7.80\%}$
test_getitem[range]	95.0680μs	41.3224μs	24.2000 KOps/s	21.5938 KOps/s	$\textbf{\color{#35bf28}+12.07\%}$
test_getitem[tuple]	53.2490μs	19.1933μs	52.1015 KOps/s	48.7479 KOps/s	$\textbf{\color{#35bf28}+6.88\%}$
test_getitem[list]	93.9750μs	36.8019μs	27.1725 KOps/s	25.0990 KOps/s	$\textbf{\color{#35bf28}+8.26\%}$
test_setitem_dim[int]	45.2440μs	27.6648μs	36.1470 KOps/s	35.5712 KOps/s	$\color{#35bf28}+1.62\%$
test_setitem_dim[slice_int]	90.9600μs	52.2335μs	19.1448 KOps/s	18.7400 KOps/s	$\color{#35bf28}+2.16\%$
test_setitem_dim[range]	0.1411ms	71.1494μs	14.0549 KOps/s	13.7763 KOps/s	$\color{#35bf28}+2.02\%$
test_setitem_dim[tuple]	77.7550μs	40.8217μs	24.4968 KOps/s	23.7869 KOps/s	$\color{#35bf28}+2.98\%$
test_setitem	0.3093ms	17.0114μs	58.7842 KOps/s	53.5859 KOps/s	$\textbf{\color{#35bf28}+9.70\%}$
test_set	0.3457ms	16.1540μs	61.9042 KOps/s	55.9394 KOps/s	$\textbf{\color{#35bf28}+10.66\%}$
test_set_shared	2.5650ms	0.1406ms	7.1144 KOps/s	6.9959 KOps/s	$\color{#35bf28}+1.69\%$
test_update	0.3720ms	18.1153μs	55.2018 KOps/s	52.2696 KOps/s	$\textbf{\color{#35bf28}+5.61\%}$
test_update_nested	0.1675ms	25.1964μs	39.6882 KOps/s	37.8148 KOps/s	$\color{#35bf28}+4.95\%$
test_set_nested	0.2813ms	18.0567μs	55.3811 KOps/s	49.4111 KOps/s	$\textbf{\color{#35bf28}+12.08\%}$
test_set_nested_new	0.3022ms	22.0835μs	45.2827 KOps/s	39.4164 KOps/s	$\textbf{\color{#35bf28}+14.88\%}$
test_select	0.1078ms	45.3186μs	22.0660 KOps/s	19.7527 KOps/s	$\textbf{\color{#35bf28}+11.71\%}$
test_unbind_speed	0.4500ms	0.3443ms	2.9046 KOps/s	2.6621 KOps/s	$\textbf{\color{#35bf28}+9.11\%}$
test_unbind_speed_stack0	62.3335ms	4.1732ms	239.6271 Ops/s	219.2663 Ops/s	$\textbf{\color{#35bf28}+9.29\%}$
test_unbind_speed_stack1	1.7813μs	0.6498μs	1.5390 MOps/s	1.6011 MOps/s	$\color{#d91a1a}-3.88\%$
test_split	56.3081ms	1.6727ms	597.8299 Ops/s	552.5667 Ops/s	$\textbf{\color{#35bf28}+8.19\%}$
test_chunk	60.2103ms	1.6613ms	601.9490 Ops/s	561.6808 Ops/s	$\textbf{\color{#35bf28}+7.17\%}$
test_creation[device0]	3.5025ms	0.2939ms	3.4022 KOps/s	3.3705 KOps/s	$\color{#35bf28}+0.94\%$
test_creation_from_tensor	54.6701ms	0.3513ms	2.8465 KOps/s	2.9894 KOps/s	$\color{#d91a1a}-4.78\%$
test_add_one[memmap_tensor0]	70.7920μs	25.5057μs	39.2069 KOps/s	39.4725 KOps/s	$\color{#d91a1a}-0.67\%$
test_contiguous[memmap_tensor0]	23.3940μs	5.6552μs	176.8285 KOps/s	170.0438 KOps/s	$\color{#35bf28}+3.99\%$
test_stack[memmap_tensor0]	64.5500μs	18.7136μs	53.4372 KOps/s	50.4549 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_memmaptd_index	0.4096ms	0.1996ms	5.0099 KOps/s	4.9349 KOps/s	$\color{#35bf28}+1.52\%$
test_memmaptd_index_astensor	0.3333ms	0.2556ms	3.9125 KOps/s	3.7663 KOps/s	$\color{#35bf28}+3.88\%$
test_memmaptd_index_op	1.0171ms	0.4916ms	2.0341 KOps/s	1.9759 KOps/s	$\color{#35bf28}+2.94\%$
test_reshape_pytree	55.2430μs	22.9986μs	43.4809 KOps/s	43.4009 KOps/s	$\color{#35bf28}+0.18\%$
test_reshape_td	67.1250μs	30.2667μs	33.0396 KOps/s	30.9197 KOps/s	$\textbf{\color{#35bf28}+6.86\%}$
test_view_pytree	53.2600μs	22.7209μs	44.0123 KOps/s	42.7509 KOps/s	$\color{#35bf28}+2.95\%$
test_view_td	21.1390μs	4.8715μs	205.2737 KOps/s	201.8139 KOps/s	$\color{#35bf28}+1.71\%$
test_unbind_pytree	68.8590μs	26.5443μs	37.6728 KOps/s	37.4140 KOps/s	$\color{#35bf28}+0.69\%$
test_unbind_td	0.1351ms	56.2104μs	17.7903 KOps/s	16.5611 KOps/s	$\textbf{\color{#35bf28}+7.42\%}$
test_split_pytree	57.4070μs	26.1611μs	38.2247 KOps/s	37.6205 KOps/s	$\color{#35bf28}+1.61\%$
test_split_td	85.9200μs	43.9035μs	22.7772 KOps/s	21.1466 KOps/s	$\textbf{\color{#35bf28}+7.71\%}$
test_add_pytree	0.1027ms	31.9582μs	31.2909 KOps/s	30.5244 KOps/s	$\color{#35bf28}+2.51\%$
test_add_td	0.1225ms	42.7492μs	23.3922 KOps/s	21.7109 KOps/s	$\textbf{\color{#35bf28}+7.74\%}$
test_distributed	41.2270μs	5.9761μs	167.3325 KOps/s	159.1965 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_tdmodule	1.7342ms	22.6303μs	44.1886 KOps/s	47.4603 KOps/s	$\textbf{\color{#d91a1a}-6.89\%}$
test_tdmodule_dispatch	0.1876ms	38.4107μs	26.0344 KOps/s	25.1418 KOps/s	$\color{#35bf28}+3.55\%$
test_tdseq	42.9300μs	24.2149μs	41.2968 KOps/s	40.6520 KOps/s	$\color{#35bf28}+1.59\%$
test_tdseq_dispatch	0.1308ms	41.3597μs	24.1781 KOps/s	22.8769 KOps/s	$\textbf{\color{#35bf28}+5.69\%}$
test_instantiation_functorch	1.4233ms	1.3106ms	763.0255 Ops/s	771.2395 Ops/s	$\color{#d91a1a}-1.07\%$
test_instantiation_td	1.5688ms	1.0214ms	979.0323 Ops/s	969.5362 Ops/s	$\color{#35bf28}+0.98\%$
test_exec_functorch	0.2370ms	0.1594ms	6.2737 KOps/s	6.2222 KOps/s	$\color{#35bf28}+0.83\%$
test_exec_functional_call	0.2092ms	0.1473ms	6.7867 KOps/s	6.7478 KOps/s	$\color{#35bf28}+0.58\%$
test_exec_td	0.2239ms	0.1425ms	7.0181 KOps/s	6.7405 KOps/s	$\color{#35bf28}+4.12\%$
test_exec_td_decorator	1.0429ms	0.1744ms	5.7326 KOps/s	5.5939 KOps/s	$\color{#35bf28}+2.48\%$
test_vmap_mlp_speed[True-True]	1.2892ms	0.9071ms	1.1024 KOps/s	1.0648 KOps/s	$\color{#35bf28}+3.53\%$
test_vmap_mlp_speed[True-False]	0.9075ms	0.4686ms	2.1339 KOps/s	2.1052 KOps/s	$\color{#35bf28}+1.36\%$
test_vmap_mlp_speed[False-True]	1.2041ms	0.7815ms	1.2796 KOps/s	1.2527 KOps/s	$\color{#35bf28}+2.15\%$
test_vmap_mlp_speed[False-False]	0.6113ms	0.3857ms	2.5928 KOps/s	2.5281 KOps/s	$\color{#35bf28}+2.56\%$
test_vmap_mlp_speed_decorator[True-True]	2.3594ms	1.7741ms	563.6644 Ops/s	556.1261 Ops/s	$\color{#35bf28}+1.36\%$
test_vmap_mlp_speed_decorator[True-False]	0.9721ms	0.5107ms	1.9581 KOps/s	1.8846 KOps/s	$\color{#35bf28}+3.90\%$
test_vmap_mlp_speed_decorator[False-True]	2.2575ms	1.4916ms	670.3991 Ops/s	657.2390 Ops/s	$\color{#35bf28}+2.00\%$
test_vmap_mlp_speed_decorator[False-False]	0.7826ms	0.3969ms	2.5195 KOps/s	2.4528 KOps/s	$\color{#35bf28}+2.72\%$

[github-actions]

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

Total Benchmarks: 127. Improved: $\large\color{#35bf28}27$. Worsened: $\large\color{#d91a1a}1$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	83.7210μs	12.7621μs	78.3569 KOps/s	79.0512 KOps/s	$\color{#d91a1a}-0.88\%$
test_plain_set_stack_nested	0.1560ms	0.1156ms	8.6526 KOps/s	8.3593 KOps/s	$\color{#35bf28}+3.51\%$
test_plain_set_nested_inplace	34.7810μs	13.9954μs	71.4520 KOps/s	71.3493 KOps/s	$\color{#35bf28}+0.14\%$
test_plain_set_stack_nested_inplace	0.1697ms	0.1409ms	7.0960 KOps/s	7.0245 KOps/s	$\color{#35bf28}+1.02\%$
test_items	30.6010μs	4.6541μs	214.8644 KOps/s	209.9038 KOps/s	$\color{#35bf28}+2.36\%$
test_items_nested	0.3657ms	0.3381ms	2.9581 KOps/s	2.9364 KOps/s	$\color{#35bf28}+0.74\%$
test_items_nested_locked	0.4329ms	0.3417ms	2.9269 KOps/s	2.9146 KOps/s	$\color{#35bf28}+0.42\%$
test_items_nested_leaf	0.2525ms	0.1982ms	5.0446 KOps/s	4.9906 KOps/s	$\color{#35bf28}+1.08\%$
test_items_stack_nested	1.4995ms	1.4476ms	690.7909 Ops/s	685.3963 Ops/s	$\color{#35bf28}+0.79\%$
test_items_stack_nested_leaf	1.3183ms	1.2691ms	787.9822 Ops/s	777.2556 Ops/s	$\color{#35bf28}+1.38\%$
test_items_stack_nested_locked	2.3143ms	0.8125ms	1.2308 KOps/s	1.2379 KOps/s	$\color{#d91a1a}-0.58\%$
test_keys	21.2410μs	4.6094μs	216.9457 KOps/s	214.7811 KOps/s	$\color{#35bf28}+1.01\%$
test_keys_nested	0.4943ms	90.0926μs	11.0997 KOps/s	11.0436 KOps/s	$\color{#35bf28}+0.51\%$
test_keys_nested_locked	0.1601ms	89.4548μs	11.1788 KOps/s	11.1478 KOps/s	$\color{#35bf28}+0.28\%$
test_keys_nested_leaf	42.1992ms	87.5710μs	11.4193 KOps/s	12.2694 KOps/s	$\textbf{\color{#d91a1a}-6.93\%}$
test_keys_stack_nested	1.3364ms	1.2578ms	795.0387 Ops/s	796.8540 Ops/s	$\color{#d91a1a}-0.23\%$
test_keys_stack_nested_leaf	1.3343ms	1.2466ms	802.1824 Ops/s	799.9001 Ops/s	$\color{#35bf28}+0.29\%$
test_keys_stack_nested_locked	0.7180ms	0.6085ms	1.6435 KOps/s	1.6309 KOps/s	$\color{#35bf28}+0.77\%$
test_values	30.1040μs	1.8781μs	532.4393 KOps/s	531.6429 KOps/s	$\color{#35bf28}+0.15\%$
test_values_nested	60.8400μs	42.5472μs	23.5033 KOps/s	23.0152 KOps/s	$\color{#35bf28}+2.12\%$
test_values_nested_locked	97.8110μs	45.0755μs	22.1850 KOps/s	22.0072 KOps/s	$\color{#35bf28}+0.81\%$
test_values_nested_leaf	58.5500μs	37.0522μs	26.9889 KOps/s	26.4622 KOps/s	$\color{#35bf28}+1.99\%$
test_values_stack_nested	1.1571ms	1.0965ms	912.0059 Ops/s	897.8887 Ops/s	$\color{#35bf28}+1.57\%$
test_values_stack_nested_leaf	1.1542ms	1.0849ms	921.7077 Ops/s	908.8319 Ops/s	$\color{#35bf28}+1.42\%$
test_values_stack_nested_locked	0.5322ms	0.4809ms	2.0795 KOps/s	2.0480 KOps/s	$\color{#35bf28}+1.54\%$
test_membership	5.1962μs	0.9448μs	1.0585 MOps/s	1.0590 MOps/s	$\color{#d91a1a}-0.05\%$
test_membership_nested	20.3390μs	2.1715μs	460.5208 KOps/s	450.4476 KOps/s	$\color{#35bf28}+2.24\%$
test_membership_nested_leaf	17.3305μs	2.0963μs	477.0325 KOps/s	472.5390 KOps/s	$\color{#35bf28}+0.95\%$
test_membership_stacked_nested	32.2100μs	10.9410μs	91.3996 KOps/s	91.2098 KOps/s	$\color{#35bf28}+0.21\%$
test_membership_stacked_nested_leaf	27.5200μs	10.9551μs	91.2819 KOps/s	91.4715 KOps/s	$\color{#d91a1a}-0.21\%$
test_membership_nested_last	19.8190μs	4.6035μs	217.2263 KOps/s	215.4742 KOps/s	$\color{#35bf28}+0.81\%$
test_membership_nested_leaf_last	26.5200μs	4.6650μs	214.3637 KOps/s	216.0180 KOps/s	$\color{#d91a1a}-0.77\%$
test_membership_stacked_nested_last	0.2082ms	0.1326ms	7.5406 KOps/s	7.4226 KOps/s	$\color{#35bf28}+1.59\%$
test_membership_stacked_nested_leaf_last	25.4410μs	12.7372μs	78.5100 KOps/s	78.8421 KOps/s	$\color{#d91a1a}-0.42\%$
test_nested_getleaf	24.4000μs	8.3178μs	120.2242 KOps/s	119.3984 KOps/s	$\color{#35bf28}+0.69\%$
test_nested_get	21.3500μs	7.8835μs	126.8473 KOps/s	125.9135 KOps/s	$\color{#35bf28}+0.74\%$
test_stacked_getleaf	0.6088ms	0.5502ms	1.8176 KOps/s	1.7929 KOps/s	$\color{#35bf28}+1.38\%$
test_stacked_get	0.5942ms	0.5182ms	1.9298 KOps/s	1.8975 KOps/s	$\color{#35bf28}+1.70\%$
test_nested_getitemleaf	29.8800μs	8.4548μs	118.2766 KOps/s	118.5289 KOps/s	$\color{#d91a1a}-0.21\%$
test_nested_getitem	20.9600μs	7.9955μs	125.0703 KOps/s	125.0805 KOps/s	$-0.01\%$
test_stacked_getitemleaf	0.5949ms	0.5619ms	1.7797 KOps/s	1.7860 KOps/s	$\color{#d91a1a}-0.35\%$
test_stacked_getitem	0.5698ms	0.5235ms	1.9101 KOps/s	1.9053 KOps/s	$\color{#35bf28}+0.26\%$
test_lock_nested	1.5763ms	0.4128ms	2.4224 KOps/s	1.8121 KOps/s	$\textbf{\color{#35bf28}+33.68\%}$
test_lock_stack_nested	62.4200ms	5.8397ms	171.2430 Ops/s	137.8077 Ops/s	$\textbf{\color{#35bf28}+24.26\%}$
test_unlock_nested	0.9092ms	0.4096ms	2.4417 KOps/s	2.3400 KOps/s	$\color{#35bf28}+4.34\%$
test_unlock_stack_nested	61.0779ms	5.9527ms	167.9915 Ops/s	162.7464 Ops/s	$\color{#35bf28}+3.22\%$
test_flatten_speed	0.4697ms	0.1869ms	5.3514 KOps/s	5.3659 KOps/s	$\color{#d91a1a}-0.27\%$
test_unflatten_speed	0.3773ms	0.3522ms	2.8394 KOps/s	2.7979 KOps/s	$\color{#35bf28}+1.48\%$
test_common_ops	1.0376ms	0.5561ms	1.7983 KOps/s	1.6709 KOps/s	$\textbf{\color{#35bf28}+7.62\%}$
test_creation	36.8600μs	1.6103μs	620.9941 KOps/s	472.6666 KOps/s	$\textbf{\color{#35bf28}+31.38\%}$
test_creation_empty	30.2510μs	6.8770μs	145.4130 KOps/s	138.4280 KOps/s	$\textbf{\color{#35bf28}+5.05\%}$
test_creation_nested_1	62.1900μs	8.7545μs	114.2266 KOps/s	105.3313 KOps/s	$\textbf{\color{#35bf28}+8.45\%}$
test_creation_nested_2	30.8710μs	11.4377μs	87.4301 KOps/s	83.4612 KOps/s	$\color{#35bf28}+4.76\%$
test_clone	98.1220μs	12.9280μs	77.3517 KOps/s	70.6149 KOps/s	$\textbf{\color{#35bf28}+9.54\%}$
test_getitem[int]	29.3000μs	10.9160μs	91.6085 KOps/s	81.8123 KOps/s	$\textbf{\color{#35bf28}+11.97\%}$
test_getitem[slice_int]	37.0810μs	21.0090μs	47.5986 KOps/s	44.8231 KOps/s	$\textbf{\color{#35bf28}+6.19\%}$
test_getitem[range]	0.3034ms	36.0542μs	27.7360 KOps/s	24.5839 KOps/s	$\textbf{\color{#35bf28}+12.82\%}$
test_getitem[tuple]	35.6010μs	18.6207μs	53.7036 KOps/s	50.1122 KOps/s	$\textbf{\color{#35bf28}+7.17\%}$
test_getitem[list]	0.3092ms	33.4237μs	29.9189 KOps/s	27.1753 KOps/s	$\textbf{\color{#35bf28}+10.10\%}$
test_setitem_dim[int]	39.6200μs	24.2599μs	41.2203 KOps/s	39.9731 KOps/s	$\color{#35bf28}+3.12\%$
test_setitem_dim[slice_int]	63.9510μs	43.0165μs	23.2469 KOps/s	22.8717 KOps/s	$\color{#35bf28}+1.64\%$
test_setitem_dim[range]	85.9920μs	60.0378μs	16.6562 KOps/s	16.7510 KOps/s	$\color{#d91a1a}-0.57\%$
test_setitem_dim[tuple]	51.9910μs	36.1757μs	27.6429 KOps/s	27.2786 KOps/s	$\color{#35bf28}+1.34\%$
test_setitem	0.1197ms	16.5515μs	60.4176 KOps/s	55.9356 KOps/s	$\textbf{\color{#35bf28}+8.01\%}$
test_set	0.1228ms	16.0256μs	62.4002 KOps/s	57.1606 KOps/s	$\textbf{\color{#35bf28}+9.17\%}$
test_set_shared	3.0495ms	0.1022ms	9.7831 KOps/s	8.7164 KOps/s	$\textbf{\color{#35bf28}+12.24\%}$
test_update	0.1194ms	18.0992μs	55.2511 KOps/s	53.7012 KOps/s	$\color{#35bf28}+2.89\%$
test_update_nested	0.1045ms	23.7728μs	42.0649 KOps/s	39.3632 KOps/s	$\textbf{\color{#35bf28}+6.86\%}$
test_set_nested	0.1093ms	17.2676μs	57.9121 KOps/s	53.1428 KOps/s	$\textbf{\color{#35bf28}+8.97\%}$
test_set_nested_new	0.1031ms	20.4462μs	48.9088 KOps/s	44.3777 KOps/s	$\textbf{\color{#35bf28}+10.21\%}$
test_select	69.4210μs	41.9010μs	23.8658 KOps/s	21.7470 KOps/s	$\textbf{\color{#35bf28}+9.74\%}$
test_to	70.4810μs	51.1620μs	19.5458 KOps/s	19.0235 KOps/s	$\color{#35bf28}+2.75\%$
test_to_nonblocking	65.8310μs	33.0393μs	30.2670 KOps/s	29.3425 KOps/s	$\color{#35bf28}+3.15\%$
test_unbind_speed	0.3753ms	0.3225ms	3.1004 KOps/s	2.8058 KOps/s	$\textbf{\color{#35bf28}+10.50\%}$
test_unbind_speed_stack0	60.4188ms	3.9403ms	253.7892 Ops/s	236.8110 Ops/s	$\textbf{\color{#35bf28}+7.17\%}$
test_unbind_speed_stack1	1.3036μs	0.5208μs	1.9200 MOps/s	1.9167 MOps/s	$\color{#35bf28}+0.17\%$
test_split	53.9667ms	1.6246ms	615.5377 Ops/s	578.6883 Ops/s	$\textbf{\color{#35bf28}+6.37\%}$
test_chunk	53.6742ms	1.6076ms	622.0421 Ops/s	583.7389 Ops/s	$\textbf{\color{#35bf28}+6.56\%}$
test_creation[device0]	0.5060ms	0.3100ms	3.2262 KOps/s	3.2370 KOps/s	$\color{#d91a1a}-0.33\%$
test_creation[device1]	0.8045ms	0.3123ms	3.2025 KOps/s	3.2002 KOps/s	$\color{#35bf28}+0.07\%$
test_creation_from_tensor	0.6975ms	0.3366ms	2.9706 KOps/s	2.9655 KOps/s	$\color{#35bf28}+0.17\%$
test_add_one[memmap_tensor0]	0.1586ms	23.5327μs	42.4941 KOps/s	41.7519 KOps/s	$\color{#35bf28}+1.78\%$
test_add_one[memmap_tensor1]	0.2017ms	71.5607μs	13.9741 KOps/s	13.9496 KOps/s	$\color{#35bf28}+0.18\%$
test_contiguous[memmap_tensor0]	26.0200μs	5.9528μs	167.9890 KOps/s	165.3439 KOps/s	$\color{#35bf28}+1.60\%$
test_contiguous[memmap_tensor1]	50.5500μs	21.1110μs	47.3686 KOps/s	46.1116 KOps/s	$\color{#35bf28}+2.73\%$
test_stack[memmap_tensor0]	42.7600μs	19.2130μs	52.0480 KOps/s	50.0396 KOps/s	$\color{#35bf28}+4.01\%$
test_stack[memmap_tensor1]	0.1048ms	70.7573μs	14.1328 KOps/s	14.0558 KOps/s	$\color{#35bf28}+0.55\%$
test_memmaptd_index	0.2675ms	0.2349ms	4.2563 KOps/s	4.2313 KOps/s	$\color{#35bf28}+0.59\%$
test_memmaptd_index_astensor	0.3203ms	0.2896ms	3.4530 KOps/s	3.4589 KOps/s	$\color{#d91a1a}-0.17\%$
test_memmaptd_index_op	0.5961ms	0.5366ms	1.8635 KOps/s	1.8400 KOps/s	$\color{#35bf28}+1.28\%$
test_reshape_pytree	37.9510μs	20.3916μs	49.0398 KOps/s	47.8656 KOps/s	$\color{#35bf28}+2.45\%$
test_reshape_td	51.7910μs	28.9380μs	34.5567 KOps/s	32.3584 KOps/s	$\textbf{\color{#35bf28}+6.79\%}$
test_view_pytree	87.1210μs	20.3197μs	49.2134 KOps/s	48.4753 KOps/s	$\color{#35bf28}+1.52\%$
test_view_td	17.1490μs	4.0578μs	246.4413 KOps/s	244.7759 KOps/s	$\color{#35bf28}+0.68\%$
test_unbind_pytree	48.6620μs	25.1931μs	39.6934 KOps/s	38.7756 KOps/s	$\color{#35bf28}+2.37\%$
test_unbind_td	85.2520μs	51.3515μs	19.4736 KOps/s	17.7604 KOps/s	$\textbf{\color{#35bf28}+9.65\%}$
test_split_pytree	37.9820μs	23.2735μs	42.9673 KOps/s	42.2550 KOps/s	$\color{#35bf28}+1.69\%$
test_split_td	63.0610μs	38.6434μs	25.8776 KOps/s	23.0997 KOps/s	$\textbf{\color{#35bf28}+12.03\%}$
test_add_pytree	47.9710μs	30.8703μs	32.3936 KOps/s	31.8887 KOps/s	$\color{#35bf28}+1.58\%$
test_add_td	67.2310μs	40.8086μs	24.5046 KOps/s	22.8902 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$
test_distributed	17.5400μs	5.4476μs	183.5668 KOps/s	179.7491 KOps/s	$\color{#35bf28}+2.12\%$
test_tdmodule	54.7110μs	16.6838μs	59.9384 KOps/s	59.8398 KOps/s	$\color{#35bf28}+0.16\%$
test_tdmodule_dispatch	0.1373ms	31.7881μs	31.4583 KOps/s	30.2012 KOps/s	$\color{#35bf28}+4.16\%$
test_tdseq	33.9410μs	19.7973μs	50.5119 KOps/s	50.0432 KOps/s	$\color{#35bf28}+0.94\%$
test_tdseq_dispatch	57.8310μs	35.5318μs	28.1438 KOps/s	27.1659 KOps/s	$\color{#35bf28}+3.60\%$
test_instantiation_functorch	1.7219ms	1.6666ms	600.0099 Ops/s	605.9721 Ops/s	$\color{#d91a1a}-0.98\%$
test_instantiation_td	1.7218ms	1.1674ms	856.5825 Ops/s	844.6519 Ops/s	$\color{#35bf28}+1.41\%$
test_exec_functorch	0.2032ms	0.1554ms	6.4343 KOps/s	6.4190 KOps/s	$\color{#35bf28}+0.24\%$
test_exec_functional_call	0.1983ms	0.1473ms	6.7896 KOps/s	6.6548 KOps/s	$\color{#35bf28}+2.03\%$
test_exec_td	0.1710ms	0.1399ms	7.1485 KOps/s	7.1394 KOps/s	$\color{#35bf28}+0.13\%$
test_exec_td_decorator	0.5944ms	0.1766ms	5.6630 KOps/s	5.5151 KOps/s	$\color{#35bf28}+2.68\%$
test_vmap_mlp_speed[True-True]	1.1014ms	1.0423ms	959.3900 Ops/s	939.5073 Ops/s	$\color{#35bf28}+2.12\%$
test_vmap_mlp_speed[True-False]	0.7177ms	0.5936ms	1.6846 KOps/s	1.6403 KOps/s	$\color{#35bf28}+2.70\%$
test_vmap_mlp_speed[False-True]	0.9906ms	0.9540ms	1.0483 KOps/s	1.0345 KOps/s	$\color{#35bf28}+1.34\%$
test_vmap_mlp_speed[False-False]	0.5861ms	0.5225ms	1.9138 KOps/s	1.8752 KOps/s	$\color{#35bf28}+2.06\%$
test_vmap_mlp_speed_decorator[True-True]	2.6130ms	1.9979ms	500.5159 Ops/s	495.4380 Ops/s	$\color{#35bf28}+1.02\%$
test_vmap_mlp_speed_decorator[True-False]	1.0219ms	0.6318ms	1.5827 KOps/s	1.5583 KOps/s	$\color{#35bf28}+1.56\%$
test_vmap_mlp_speed_decorator[False-True]	2.1407ms	1.7259ms	579.4146 Ops/s	563.6227 Ops/s	$\color{#35bf28}+2.80\%$
test_vmap_mlp_speed_decorator[False-False]	0.8272ms	0.5396ms	1.8531 KOps/s	1.8137 KOps/s	$\color{#35bf28}+2.17\%$
test_vmap_transformer_speed[True-True]	12.1691ms	12.1124ms	82.5597 Ops/s	81.8928 Ops/s	$\color{#35bf28}+0.81\%$
test_vmap_transformer_speed[True-False]	7.9535ms	7.8991ms	126.5959 Ops/s	123.1632 Ops/s	$\color{#35bf28}+2.79\%$
test_vmap_transformer_speed[False-True]	12.0650ms	12.0124ms	83.2476 Ops/s	82.7925 Ops/s	$\color{#35bf28}+0.55\%$
test_vmap_transformer_speed[False-False]	7.8841ms	7.7991ms	128.2200 Ops/s	124.7879 Ops/s	$\color{#35bf28}+2.75\%$
test_vmap_transformer_speed_decorator[True-True]	0.1353s	67.2646ms	14.8667 Ops/s	14.5026 Ops/s	$\color{#35bf28}+2.51\%$
test_vmap_transformer_speed_decorator[True-False]	20.8393ms	19.1048ms	52.3428 Ops/s	50.4640 Ops/s	$\color{#35bf28}+3.72\%$
test_vmap_transformer_speed_decorator[False-True]	57.8216ms	56.9590ms	17.5565 Ops/s	16.9333 Ops/s	$\color{#35bf28}+3.68\%$
test_vmap_transformer_speed_decorator[False-False]	20.3741ms	18.6562ms	53.6014 Ops/s	51.9814 Ops/s	$\color{#35bf28}+3.12\%$