[Feature] Better dtype coverage

[vmoens]

closes #833

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}19$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.3220μs	17.6406μs	56.6876 KOps/s	60.9173 KOps/s	$\textbf{\color{#d91a1a}-6.94\%}$
test_plain_set_stack_nested	61.2940μs	17.6439μs	56.6769 KOps/s	60.9882 KOps/s	$\textbf{\color{#d91a1a}-7.07\%}$
test_plain_set_nested_inplace	57.4980μs	19.7057μs	50.7468 KOps/s	54.0795 KOps/s	$\textbf{\color{#d91a1a}-6.16\%}$
test_plain_set_stack_nested_inplace	54.9920μs	19.7685μs	50.5856 KOps/s	54.2124 KOps/s	$\textbf{\color{#d91a1a}-6.69\%}$
test_items	22.8330μs	2.5405μs	393.6167 KOps/s	399.8147 KOps/s	$\color{#d91a1a}-1.55\%$
test_items_nested	1.2688ms	0.2663ms	3.7552 KOps/s	3.7904 KOps/s	$\color{#d91a1a}-0.93\%$
test_items_nested_locked	0.5090ms	0.2671ms	3.7445 KOps/s	3.7474 KOps/s	$\color{#d91a1a}-0.08\%$
test_items_nested_leaf	0.1418ms	76.8830μs	13.0068 KOps/s	13.0226 KOps/s	$\color{#d91a1a}-0.12\%$
test_items_stack_nested	0.5029ms	0.2692ms	3.7142 KOps/s	3.7283 KOps/s	$\color{#d91a1a}-0.38\%$
test_items_stack_nested_leaf	0.3378ms	80.7692μs	12.3810 KOps/s	13.2810 KOps/s	$\textbf{\color{#d91a1a}-6.78\%}$
test_items_stack_nested_locked	1.2876ms	0.2667ms	3.7496 KOps/s	3.7524 KOps/s	$\color{#d91a1a}-0.07\%$
test_keys	35.8570μs	3.9396μs	253.8346 KOps/s	262.4316 KOps/s	$\color{#d91a1a}-3.28\%$
test_keys_nested	0.2227ms	0.1361ms	7.3486 KOps/s	7.2301 KOps/s	$\color{#35bf28}+1.64\%$
test_keys_nested_locked	0.7091ms	0.1419ms	7.0463 KOps/s	6.9539 KOps/s	$\color{#35bf28}+1.33\%$
test_keys_nested_leaf	0.3413ms	0.1165ms	8.5861 KOps/s	8.5687 KOps/s	$\color{#35bf28}+0.20\%$
test_keys_stack_nested	0.2234ms	0.1372ms	7.2896 KOps/s	7.3642 KOps/s	$\color{#d91a1a}-1.01\%$
test_keys_stack_nested_leaf	0.2395ms	0.1166ms	8.5789 KOps/s	8.4965 KOps/s	$\color{#35bf28}+0.97\%$
test_keys_stack_nested_locked	0.2529ms	0.1420ms	7.0406 KOps/s	7.1424 KOps/s	$\color{#d91a1a}-1.43\%$
test_values	10.7225μs	1.1710μs	853.9843 KOps/s	872.0107 KOps/s	$\color{#d91a1a}-2.07\%$
test_values_nested	0.1051ms	50.3682μs	19.8538 KOps/s	19.6896 KOps/s	$\color{#35bf28}+0.83\%$
test_values_nested_locked	0.1158ms	50.3826μs	19.8481 KOps/s	19.6532 KOps/s	$\color{#35bf28}+0.99\%$
test_values_nested_leaf	83.0650μs	45.3897μs	22.0314 KOps/s	21.8534 KOps/s	$\color{#35bf28}+0.81\%$
test_values_stack_nested	0.1201ms	51.7158μs	19.3364 KOps/s	18.9330 KOps/s	$\color{#35bf28}+2.13\%$
test_values_stack_nested_leaf	84.9490μs	46.0438μs	21.7184 KOps/s	21.6871 KOps/s	$\color{#35bf28}+0.14\%$
test_values_stack_nested_locked	0.1064ms	51.5299μs	19.4062 KOps/s	19.0250 KOps/s	$\color{#35bf28}+2.00\%$
test_membership	33.9630μs	1.3502μs	740.6242 KOps/s	730.3180 KOps/s	$\color{#35bf28}+1.41\%$
test_membership_nested	29.4950μs	3.4416μs	290.5616 KOps/s	287.7178 KOps/s	$\color{#35bf28}+0.99\%$
test_membership_nested_leaf	27.1100μs	3.4653μs	288.5752 KOps/s	291.9922 KOps/s	$\color{#d91a1a}-1.17\%$
test_membership_stacked_nested	34.4040μs	3.4014μs	293.9966 KOps/s	289.7410 KOps/s	$\color{#35bf28}+1.47\%$
test_membership_stacked_nested_leaf	27.2210μs	3.4481μs	290.0159 KOps/s	249.2783 KOps/s	$\textbf{\color{#35bf28}+16.34\%}$
test_membership_nested_last	28.5640μs	4.1524μs	240.8225 KOps/s	233.7240 KOps/s	$\color{#35bf28}+3.04\%$
test_membership_nested_leaf_last	36.2370μs	4.1696μs	239.8301 KOps/s	236.7697 KOps/s	$\color{#35bf28}+1.29\%$
test_membership_stacked_nested_last	36.7290μs	4.1297μs	242.1458 KOps/s	75.9188 KOps/s	$\textbf{\color{#35bf28}+218.95\%}$
test_membership_stacked_nested_leaf_last	28.5540μs	4.1516μs	240.8715 KOps/s	75.5629 KOps/s	$\textbf{\color{#35bf28}+218.77\%}$
test_nested_getleaf	44.3720μs	10.4991μs	95.2467 KOps/s	95.2474 KOps/s	$-0.00\%$
test_nested_get	58.7480μs	9.9796μs	100.2046 KOps/s	99.8959 KOps/s	$\color{#35bf28}+0.31\%$
test_stacked_getleaf	51.1150μs	10.5903μs	94.4263 KOps/s	94.8685 KOps/s	$\color{#d91a1a}-0.47\%$
test_stacked_get	45.3950μs	9.8193μs	101.8398 KOps/s	100.0066 KOps/s	$\color{#35bf28}+1.83\%$
test_nested_getitemleaf	45.0540μs	11.0718μs	90.3197 KOps/s	87.8632 KOps/s	$\color{#35bf28}+2.80\%$
test_nested_getitem	49.7830μs	10.1566μs	98.4584 KOps/s	97.0905 KOps/s	$\color{#35bf28}+1.41\%$
test_stacked_getitemleaf	42.8400μs	11.0851μs	90.2110 KOps/s	90.2928 KOps/s	$\color{#d91a1a}-0.09\%$
test_stacked_getitem	30.0160μs	10.0602μs	99.4016 KOps/s	97.3083 KOps/s	$\color{#35bf28}+2.15\%$
test_lock_nested	49.0483ms	0.3864ms	2.5882 KOps/s	2.9567 KOps/s	$\textbf{\color{#d91a1a}-12.46\%}$
test_lock_stack_nested	0.4900ms	0.3096ms	3.2298 KOps/s	3.3749 KOps/s	$\color{#d91a1a}-4.30\%$
test_unlock_nested	0.9955ms	0.3537ms	2.8272 KOps/s	2.9057 KOps/s	$\color{#d91a1a}-2.70\%$
test_unlock_stack_nested	0.6848ms	0.3196ms	3.1290 KOps/s	3.2976 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_flatten_speed	0.1755ms	96.1303μs	10.4025 KOps/s	10.3257 KOps/s	$\color{#35bf28}+0.74\%$
test_unflatten_speed	0.9068ms	0.4113ms	2.4314 KOps/s	2.4551 KOps/s	$\color{#d91a1a}-0.97\%$
test_common_ops	4.2398ms	0.7244ms	1.3805 KOps/s	1.4414 KOps/s	$\color{#d91a1a}-4.22\%$
test_creation	15.7290μs	1.9030μs	525.4857 KOps/s	504.2000 KOps/s	$\color{#35bf28}+4.22\%$
test_creation_empty	37.7010μs	11.1548μs	89.6473 KOps/s	108.5557 KOps/s	$\textbf{\color{#d91a1a}-17.42\%}$
test_creation_nested_1	52.4380μs	13.9047μs	71.9183 KOps/s	83.1157 KOps/s	$\textbf{\color{#d91a1a}-13.47\%}$
test_creation_nested_2	0.1788ms	17.4905μs	57.1740 KOps/s	65.8842 KOps/s	$\textbf{\color{#d91a1a}-13.22\%}$
test_clone	0.1182ms	13.4154μs	74.5411 KOps/s	73.1633 KOps/s	$\color{#35bf28}+1.88\%$
test_getitem[int]	45.2540μs	11.1878μs	89.3827 KOps/s	86.1007 KOps/s	$\color{#35bf28}+3.81\%$
test_getitem[slice_int]	85.5290μs	22.4031μs	44.6367 KOps/s	43.1471 KOps/s	$\color{#35bf28}+3.45\%$
test_getitem[range]	80.6200μs	59.2124μs	16.8884 KOps/s	16.5402 KOps/s	$\color{#35bf28}+2.10\%$
test_getitem[tuple]	49.2510μs	18.8450μs	53.0645 KOps/s	52.2920 KOps/s	$\color{#35bf28}+1.48\%$
test_getitem[list]	0.1398ms	39.5393μs	25.2913 KOps/s	24.0768 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_setitem_dim[int]	61.5140μs	34.4614μs	29.0180 KOps/s	31.5197 KOps/s	$\textbf{\color{#d91a1a}-7.94\%}$
test_setitem_dim[slice_int]	0.1154ms	60.9041μs	16.4192 KOps/s	16.7584 KOps/s	$\color{#d91a1a}-2.02\%$
test_setitem_dim[range]	0.1385ms	82.4719μs	12.1253 KOps/s	12.1462 KOps/s	$\color{#d91a1a}-0.17\%$
test_setitem_dim[tuple]	0.1010ms	50.0891μs	19.9644 KOps/s	20.9771 KOps/s	$\color{#d91a1a}-4.83\%$
test_setitem	57.7970μs	20.6999μs	48.3095 KOps/s	50.3892 KOps/s	$\color{#d91a1a}-4.13\%$
test_set	96.2790μs	20.3214μs	49.2092 KOps/s	46.6368 KOps/s	$\textbf{\color{#35bf28}+5.52\%}$
test_set_shared	1.5681ms	0.1383ms	7.2298 KOps/s	7.0194 KOps/s	$\color{#35bf28}+3.00\%$
test_update	85.1490μs	23.2378μs	43.0334 KOps/s	47.4416 KOps/s	$\textbf{\color{#d91a1a}-9.29\%}$
test_update_nested	76.3220μs	32.2410μs	31.0165 KOps/s	32.8176 KOps/s	$\textbf{\color{#d91a1a}-5.49\%}$
test_update__nested	83.9570μs	25.3262μs	39.4847 KOps/s	39.5709 KOps/s	$\color{#d91a1a}-0.22\%$
test_set_nested	66.0530μs	22.2481μs	44.9476 KOps/s	47.0241 KOps/s	$\color{#d91a1a}-4.42\%$
test_set_nested_new	84.3370μs	26.1634μs	38.2214 KOps/s	38.9203 KOps/s	$\color{#d91a1a}-1.80\%$
test_select	0.1120ms	40.8958μs	24.4524 KOps/s	24.5734 KOps/s	$\color{#d91a1a}-0.49\%$
test_select_nested	0.1246ms	59.9362μs	16.6844 KOps/s	16.4051 KOps/s	$\color{#35bf28}+1.70\%$
test_exclude_nested	0.2380ms	0.1194ms	8.3719 KOps/s	8.1861 KOps/s	$\color{#35bf28}+2.27\%$
test_empty[True]	0.7267ms	0.3976ms	2.5152 KOps/s	2.5755 KOps/s	$\color{#d91a1a}-2.34\%$
test_empty[False]	6.3468μs	1.1661μs	857.5635 KOps/s	856.1258 KOps/s	$\color{#35bf28}+0.17\%$
test_unbind_speed	0.4118ms	0.2546ms	3.9283 KOps/s	3.9633 KOps/s	$\color{#d91a1a}-0.88\%$
test_unbind_speed_stack0	0.5002ms	0.2502ms	3.9968 KOps/s	4.1264 KOps/s	$\color{#d91a1a}-3.14\%$
test_unbind_speed_stack1	0.9197ms	0.6367ms	1.5706 KOps/s	1.4340 KOps/s	$\textbf{\color{#35bf28}+9.53\%}$
test_split	70.0719ms	1.5988ms	625.4621 Ops/s	583.3214 Ops/s	$\textbf{\color{#35bf28}+7.22\%}$
test_chunk	71.5994ms	1.6002ms	624.9150 Ops/s	630.7567 Ops/s	$\color{#d91a1a}-0.93\%$
test_creation[device0]	0.2521ms	85.3757μs	11.7129 KOps/s	11.7124 KOps/s	$+0.00\%$
test_creation_from_tensor	3.4505ms	86.2165μs	11.5987 KOps/s	11.9709 KOps/s	$\color{#d91a1a}-3.11\%$
test_add_one[memmap_tensor0]	89.4460μs	5.2436μs	190.7082 KOps/s	175.2390 KOps/s	$\textbf{\color{#35bf28}+8.83\%}$
test_contiguous[memmap_tensor0]	21.7410μs	0.6321μs	1.5820 MOps/s	1.5813 MOps/s	$\color{#35bf28}+0.04\%$
test_stack[memmap_tensor0]	16.9220μs	3.5596μs	280.9334 KOps/s	265.2076 KOps/s	$\textbf{\color{#35bf28}+5.93\%}$
test_memmaptd_index	0.9549ms	0.2538ms	3.9407 KOps/s	3.9607 KOps/s	$\color{#d91a1a}-0.51\%$
test_memmaptd_index_astensor	0.7716ms	0.3266ms	3.0621 KOps/s	3.0830 KOps/s	$\color{#d91a1a}-0.68\%$
test_memmaptd_index_op	0.8960ms	0.6149ms	1.6263 KOps/s	1.6772 KOps/s	$\color{#d91a1a}-3.03\%$
test_serialize_model	0.1750s	0.1112s	8.9918 Ops/s	8.9016 Ops/s	$\color{#35bf28}+1.01\%$
test_serialize_model_pickle	0.4663s	0.3850s	2.5972 Ops/s	2.6222 Ops/s	$\color{#d91a1a}-0.95\%$
test_serialize_weights	0.1079s	0.1018s	9.8209 Ops/s	8.7553 Ops/s	$\textbf{\color{#35bf28}+12.17\%}$
test_serialize_weights_returnearly	0.1930s	0.1354s	7.3837 Ops/s	7.3855 Ops/s	$\color{#d91a1a}-0.02\%$
test_serialize_weights_pickle	0.8475s	0.4971s	2.0115 Ops/s	1.5489 Ops/s	$\textbf{\color{#35bf28}+29.87\%}$
test_serialize_weights_filesystem	0.1503s	97.4455ms	10.2621 Ops/s	10.2794 Ops/s	$\color{#d91a1a}-0.17\%$
test_serialize_model_filesystem	0.1004s	94.7147ms	10.5580 Ops/s	10.9260 Ops/s	$\color{#d91a1a}-3.37\%$
test_reshape_pytree	54.7930μs	25.3869μs	39.3904 KOps/s	38.0322 KOps/s	$\color{#35bf28}+3.57\%$
test_reshape_td	0.1096ms	34.1862μs	29.2515 KOps/s	28.8925 KOps/s	$\color{#35bf28}+1.24\%$
test_view_pytree	55.3230μs	25.2340μs	39.6290 KOps/s	39.2903 KOps/s	$\color{#35bf28}+0.86\%$
test_view_td	0.1017ms	38.4660μs	25.9970 KOps/s	25.0416 KOps/s	$\color{#35bf28}+3.81\%$
test_unbind_pytree	75.5810μs	29.6453μs	33.7322 KOps/s	34.0307 KOps/s	$\color{#d91a1a}-0.88\%$
test_unbind_td	0.4276ms	37.6249μs	26.5781 KOps/s	26.5193 KOps/s	$\color{#35bf28}+0.22\%$
test_split_pytree	71.2120μs	29.0923μs	34.3733 KOps/s	34.5544 KOps/s	$\color{#d91a1a}-0.52\%$
test_split_td	0.5844ms	40.3137μs	24.8055 KOps/s	24.3541 KOps/s	$\color{#35bf28}+1.85\%$
test_add_pytree	0.1137ms	34.2533μs	29.1943 KOps/s	28.5863 KOps/s	$\color{#35bf28}+2.13\%$
test_add_td	0.1191ms	53.9339μs	18.5412 KOps/s	18.9175 KOps/s	$\color{#d91a1a}-1.99\%$
test_distributed	0.2556ms	0.1025ms	9.7558 KOps/s	9.8269 KOps/s	$\color{#d91a1a}-0.72\%$
test_tdmodule	86.9020μs	18.1419μs	55.1209 KOps/s	58.6647 KOps/s	$\textbf{\color{#d91a1a}-6.04\%}$
test_tdmodule_dispatch	71.0320μs	34.8932μs	28.6588 KOps/s	29.5248 KOps/s	$\color{#d91a1a}-2.93\%$
test_tdseq	39.3130μs	20.8895μs	47.8710 KOps/s	42.9639 KOps/s	$\textbf{\color{#35bf28}+11.42\%}$
test_tdseq_dispatch	79.9090μs	41.4587μs	24.1204 KOps/s	25.5743 KOps/s	$\textbf{\color{#d91a1a}-5.69\%}$
test_instantiation_functorch	1.5655ms	1.3058ms	765.7923 Ops/s	760.6286 Ops/s	$\color{#35bf28}+0.68\%$
test_instantiation_td	2.0610ms	1.0476ms	954.5771 Ops/s	991.2174 Ops/s	$\color{#d91a1a}-3.70\%$
test_exec_functorch	0.3133ms	0.1588ms	6.2963 KOps/s	6.2404 KOps/s	$\color{#35bf28}+0.90\%$
test_exec_functional_call	0.2787ms	0.1470ms	6.8045 KOps/s	6.6428 KOps/s	$\color{#35bf28}+2.43\%$
test_exec_td	0.2963ms	0.1423ms	7.0270 KOps/s	6.8309 KOps/s	$\color{#35bf28}+2.87\%$
test_exec_td_decorator	0.6566ms	0.2195ms	4.5556 KOps/s	4.5460 KOps/s	$\color{#35bf28}+0.21\%$
test_vmap_mlp_speed[True-True]	0.8162ms	0.5016ms	1.9938 KOps/s	2.0817 KOps/s	$\color{#d91a1a}-4.22\%$
test_vmap_mlp_speed[True-False]	0.5962ms	0.4771ms	2.0960 KOps/s	2.1133 KOps/s	$\color{#d91a1a}-0.82\%$
test_vmap_mlp_speed[False-True]	0.6692ms	0.3881ms	2.5765 KOps/s	2.5759 KOps/s	$\color{#35bf28}+0.02\%$
test_vmap_mlp_speed[False-False]	0.7284ms	0.3928ms	2.5459 KOps/s	2.5677 KOps/s	$\color{#d91a1a}-0.85\%$
test_vmap_mlp_speed_decorator[True-True]	1.0313ms	0.5497ms	1.8190 KOps/s	1.8211 KOps/s	$\color{#d91a1a}-0.11\%$
test_vmap_mlp_speed_decorator[True-False]	1.0375ms	0.5532ms	1.8077 KOps/s	1.8251 KOps/s	$\color{#d91a1a}-0.95\%$
test_vmap_mlp_speed_decorator[False-True]	0.7745ms	0.4519ms	2.2127 KOps/s	2.1619 KOps/s	$\color{#35bf28}+2.35\%$
test_vmap_mlp_speed_decorator[False-False]	0.5844ms	0.4514ms	2.2153 KOps/s	2.2254 KOps/s	$\color{#d91a1a}-0.45\%$
test_to_module_speed[True]	2.6682ms	1.6793ms	595.4995 Ops/s	595.1880 Ops/s	$\color{#35bf28}+0.05\%$
test_to_module_speed[False]	1.8689ms	1.6617ms	601.7841 Ops/s	595.1387 Ops/s	$\color{#35bf28}+1.12\%$
test_tc_init	62.5360μs	30.7900μs	32.4781 KOps/s	37.4213 KOps/s	$\textbf{\color{#d91a1a}-13.21\%}$
test_tc_init_nested	0.1536ms	63.2081μs	15.8208 KOps/s	19.0623 KOps/s	$\textbf{\color{#d91a1a}-17.00\%}$
test_tc_first_layer_tensor	5.1109μs	0.6861μs	1.4575 MOps/s	1.5354 MOps/s	$\textbf{\color{#d91a1a}-5.08\%}$
test_tc_first_layer_nontensor	1.9100μs	0.6722μs	1.4875 MOps/s	1.4861 MOps/s	$\color{#35bf28}+0.10\%$
test_tc_second_layer_tensor	14.0560μs	1.8349μs	544.9962 KOps/s	549.7330 KOps/s	$\color{#d91a1a}-0.86\%$
test_tc_second_layer_nontensor	11.6557μs	1.5101μs	662.2217 KOps/s	673.8859 KOps/s	$\color{#d91a1a}-1.73\%$
test_unbind	74.8064ms	6.1297ms	163.1411 Ops/s	190.4981 Ops/s	$\textbf{\color{#d91a1a}-14.36\%}$
test_full_like	17.7217ms	11.1250ms	89.8879 Ops/s	91.0798 Ops/s	$\color{#d91a1a}-1.31\%$
test_zeros_like	8.5218ms	5.6388ms	177.3441 Ops/s	181.7638 Ops/s	$\color{#d91a1a}-2.43\%$
test_ones_like	13.0622ms	6.2063ms	161.1257 Ops/s	164.3370 Ops/s	$\color{#d91a1a}-1.95\%$
test_clone	11.6775ms	7.4011ms	135.1145 Ops/s	129.8253 Ops/s	$\color{#35bf28}+4.07\%$
test_squeeze	60.1930μs	14.1203μs	70.8202 KOps/s	72.3273 KOps/s	$\color{#d91a1a}-2.08\%$
test_unsqueeze	0.1151ms	60.8481μs	16.4344 KOps/s	16.5537 KOps/s	$\color{#d91a1a}-0.72\%$
test_split	0.2167ms	0.1113ms	8.9834 KOps/s	8.4940 KOps/s	$\textbf{\color{#35bf28}+5.76\%}$
test_permute	0.1893ms	0.1258ms	7.9516 KOps/s	7.9465 KOps/s	$\color{#35bf28}+0.06\%$
test_stack	27.7363ms	20.7999ms	48.0773 Ops/s	44.7744 Ops/s	$\textbf{\color{#35bf28}+7.38\%}$
test_cat	33.9267ms	21.3976ms	46.7343 Ops/s	44.4662 Ops/s	$\textbf{\color{#35bf28}+5.10\%}$

[github-actions]

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

Total Benchmarks: 152. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}28$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.2310μs	13.5662μs	73.7125 KOps/s	83.3950 KOps/s	$\textbf{\color{#d91a1a}-11.61\%}$
test_plain_set_stack_nested	28.6610μs	13.6059μs	73.4976 KOps/s	81.5875 KOps/s	$\textbf{\color{#d91a1a}-9.92\%}$
test_plain_set_nested_inplace	37.9410μs	14.9620μs	66.8359 KOps/s	75.0313 KOps/s	$\textbf{\color{#d91a1a}-10.92\%}$
test_plain_set_stack_nested_inplace	34.5900μs	14.9730μs	66.7867 KOps/s	74.5433 KOps/s	$\textbf{\color{#d91a1a}-10.41\%}$
test_items	25.8000μs	4.6680μs	214.2244 KOps/s	207.9323 KOps/s	$\color{#35bf28}+3.03\%$
test_items_nested	0.3637ms	0.3366ms	2.9713 KOps/s	2.9901 KOps/s	$\color{#d91a1a}-0.63\%$
test_items_nested_locked	0.5182ms	0.3377ms	2.9616 KOps/s	2.9298 KOps/s	$\color{#35bf28}+1.08\%$
test_items_nested_leaf	0.1043ms	83.2033μs	12.0187 KOps/s	12.2562 KOps/s	$\color{#d91a1a}-1.94\%$
test_items_stack_nested	0.3909ms	0.3398ms	2.9429 KOps/s	2.9893 KOps/s	$\color{#d91a1a}-1.55\%$
test_items_stack_nested_leaf	0.1019ms	84.2188μs	11.8738 KOps/s	12.0136 KOps/s	$\color{#d91a1a}-1.16\%$
test_items_stack_nested_locked	0.3906ms	0.3413ms	2.9299 KOps/s	2.9545 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys	19.4690μs	4.3700μs	228.8351 KOps/s	229.4254 KOps/s	$\color{#d91a1a}-0.26\%$
test_keys_nested	99.2510μs	66.8735μs	14.9536 KOps/s	15.0372 KOps/s	$\color{#d91a1a}-0.56\%$
test_keys_nested_locked	2.2080ms	71.3377μs	14.0178 KOps/s	13.9912 KOps/s	$\color{#35bf28}+0.19\%$
test_keys_nested_leaf	79.2300μs	57.2263μs	17.4745 KOps/s	17.4281 KOps/s	$\color{#35bf28}+0.27\%$
test_keys_stack_nested	83.3010μs	66.8179μs	14.9660 KOps/s	14.9979 KOps/s	$\color{#d91a1a}-0.21\%$
test_keys_stack_nested_leaf	75.5520μs	57.6293μs	17.3523 KOps/s	17.4967 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys_stack_nested_locked	94.1510μs	72.0801μs	13.8735 KOps/s	14.1117 KOps/s	$\color{#d91a1a}-1.69\%$
test_values	8.6870μs	1.8163μs	550.5657 KOps/s	554.7816 KOps/s	$\color{#d91a1a}-0.76\%$
test_values_nested	58.9700μs	35.4805μs	28.1845 KOps/s	28.5509 KOps/s	$\color{#d91a1a}-1.28\%$
test_values_nested_locked	62.9400μs	37.5731μs	26.6148 KOps/s	26.7301 KOps/s	$\color{#d91a1a}-0.43\%$
test_values_nested_leaf	50.3900μs	31.8022μs	31.4443 KOps/s	31.7758 KOps/s	$\color{#d91a1a}-1.04\%$
test_values_stack_nested	76.1200μs	36.3199μs	27.5331 KOps/s	27.7163 KOps/s	$\color{#d91a1a}-0.66\%$
test_values_stack_nested_leaf	60.3410μs	32.3118μs	30.9485 KOps/s	30.9990 KOps/s	$\color{#d91a1a}-0.16\%$
test_values_stack_nested_locked	63.4200μs	38.3462μs	26.0782 KOps/s	26.4579 KOps/s	$\color{#d91a1a}-1.44\%$
test_membership	3.3700μs	0.7225μs	1.3842 MOps/s	1.4078 MOps/s	$\color{#d91a1a}-1.68\%$
test_membership_nested	28.7910μs	2.6202μs	381.6528 KOps/s	387.8971 KOps/s	$\color{#d91a1a}-1.61\%$
test_membership_nested_leaf	18.3710μs	2.6361μs	379.3537 KOps/s	385.9314 KOps/s	$\color{#d91a1a}-1.70\%$
test_membership_stacked_nested	19.6900μs	2.6518μs	377.0984 KOps/s	386.1520 KOps/s	$\color{#d91a1a}-2.34\%$
test_membership_stacked_nested_leaf	37.8400μs	2.6489μs	377.5185 KOps/s	384.6815 KOps/s	$\color{#d91a1a}-1.86\%$
test_membership_nested_last	29.5000μs	3.2012μs	312.3845 KOps/s	320.0066 KOps/s	$\color{#d91a1a}-2.38\%$
test_membership_nested_leaf_last	31.9800μs	3.1812μs	314.3473 KOps/s	319.1405 KOps/s	$\color{#d91a1a}-1.50\%$
test_membership_stacked_nested_last	19.7800μs	3.1710μs	315.3555 KOps/s	234.6146 KOps/s	$\textbf{\color{#35bf28}+34.41\%}$
test_membership_stacked_nested_leaf_last	33.8210μs	3.1682μs	315.6403 KOps/s	236.0892 KOps/s	$\textbf{\color{#35bf28}+33.70\%}$
test_nested_getleaf	25.0300μs	8.3633μs	119.5706 KOps/s	120.1045 KOps/s	$\color{#d91a1a}-0.44\%$
test_nested_get	45.6410μs	7.8876μs	126.7810 KOps/s	126.8527 KOps/s	$\color{#d91a1a}-0.06\%$
test_stacked_getleaf	38.0100μs	8.3946μs	119.1246 KOps/s	118.2514 KOps/s	$\color{#35bf28}+0.74\%$
test_stacked_get	40.1410μs	7.8802μs	126.9004 KOps/s	126.0947 KOps/s	$\color{#35bf28}+0.64\%$
test_nested_getitemleaf	35.0000μs	8.5079μs	117.5374 KOps/s	118.2628 KOps/s	$\color{#d91a1a}-0.61\%$
test_nested_getitem	31.0310μs	8.0482μs	124.2519 KOps/s	123.6422 KOps/s	$\color{#35bf28}+0.49\%$
test_stacked_getitemleaf	25.9010μs	8.5920μs	116.3875 KOps/s	116.7206 KOps/s	$\color{#d91a1a}-0.29\%$
test_stacked_getitem	35.5010μs	8.0715μs	123.8930 KOps/s	124.7772 KOps/s	$\color{#d91a1a}-0.71\%$
test_lock_nested	58.8248ms	0.4062ms	2.4616 KOps/s	2.5009 KOps/s	$\color{#d91a1a}-1.58\%$
test_lock_stack_nested	0.3481ms	0.3021ms	3.3097 KOps/s	3.3433 KOps/s	$\color{#d91a1a}-1.00\%$
test_unlock_nested	61.6587ms	0.4115ms	2.4302 KOps/s	2.4679 KOps/s	$\color{#d91a1a}-1.53\%$
test_unlock_stack_nested	0.3529ms	0.3101ms	3.2247 KOps/s	3.2499 KOps/s	$\color{#d91a1a}-0.78\%$
test_flatten_speed	0.3021ms	0.1017ms	9.8376 KOps/s	9.9973 KOps/s	$\color{#d91a1a}-1.60\%$
test_unflatten_speed	0.3578ms	0.2903ms	3.4452 KOps/s	3.3997 KOps/s	$\color{#35bf28}+1.34\%$
test_common_ops	1.0621ms	0.6074ms	1.6464 KOps/s	1.7973 KOps/s	$\textbf{\color{#d91a1a}-8.40\%}$
test_creation	33.2300μs	1.6545μs	604.4077 KOps/s	609.4030 KOps/s	$\color{#d91a1a}-0.82\%$
test_creation_empty	38.4300μs	10.1478μs	98.5435 KOps/s	140.8598 KOps/s	$\textbf{\color{#d91a1a}-30.04\%}$
test_creation_nested_1	34.6300μs	11.9276μs	83.8391 KOps/s	112.2319 KOps/s	$\textbf{\color{#d91a1a}-25.30\%}$
test_creation_nested_2	32.6900μs	14.1901μs	70.4716 KOps/s	89.7790 KOps/s	$\textbf{\color{#d91a1a}-21.51\%}$
test_clone	81.3810μs	11.6947μs	85.5088 KOps/s	84.4036 KOps/s	$\color{#35bf28}+1.31\%$
test_getitem[int]	36.8600μs	10.9560μs	91.2745 KOps/s	92.5361 KOps/s	$\color{#d91a1a}-1.36\%$
test_getitem[slice_int]	52.1310μs	20.6762μs	48.3649 KOps/s	48.5250 KOps/s	$\color{#d91a1a}-0.33\%$
test_getitem[range]	64.0820μs	46.7890μs	21.3725 KOps/s	20.7663 KOps/s	$\color{#35bf28}+2.92\%$
test_getitem[tuple]	56.4010μs	18.7530μs	53.3249 KOps/s	54.3952 KOps/s	$\color{#d91a1a}-1.97\%$
test_getitem[list]	0.1427ms	33.7724μs	29.6099 KOps/s	29.6511 KOps/s	$\color{#d91a1a}-0.14\%$
test_setitem_dim[int]	54.2600μs	32.1359μs	31.1179 KOps/s	37.7243 KOps/s	$\textbf{\color{#d91a1a}-17.51\%}$
test_setitem_dim[slice_int]	76.2310μs	51.9384μs	19.2536 KOps/s	20.9877 KOps/s	$\textbf{\color{#d91a1a}-8.26\%}$
test_setitem_dim[range]	0.1069ms	70.9950μs	14.0855 KOps/s	14.8046 KOps/s	$\color{#d91a1a}-4.86\%$
test_setitem_dim[tuple]	70.8310μs	46.1334μs	21.6763 KOps/s	23.9616 KOps/s	$\textbf{\color{#d91a1a}-9.54\%}$
test_setitem	62.7400μs	17.4831μs	57.1981 KOps/s	64.2658 KOps/s	$\textbf{\color{#d91a1a}-11.00\%}$
test_set	43.5810μs	16.7761μs	59.6087 KOps/s	65.9859 KOps/s	$\textbf{\color{#d91a1a}-9.66\%}$
test_set_shared	1.2277ms	0.1002ms	9.9815 KOps/s	9.4426 KOps/s	$\textbf{\color{#35bf28}+5.71\%}$
test_update	64.9000μs	20.3655μs	49.1026 KOps/s	57.6639 KOps/s	$\textbf{\color{#d91a1a}-14.85\%}$
test_update_nested	59.7620μs	25.5348μs	39.1623 KOps/s	44.1737 KOps/s	$\textbf{\color{#d91a1a}-11.34\%}$
test_update__nested	48.5100μs	22.4008μs	44.6412 KOps/s	43.6775 KOps/s	$\color{#35bf28}+2.21\%$
test_set_nested	48.5910μs	17.8434μs	56.0432 KOps/s	60.3646 KOps/s	$\textbf{\color{#d91a1a}-7.16\%}$
test_set_nested_new	43.9910μs	20.6221μs	48.4918 KOps/s	51.5140 KOps/s	$\textbf{\color{#d91a1a}-5.87\%}$
test_select	63.0320μs	33.2793μs	30.0487 KOps/s	30.7363 KOps/s	$\color{#d91a1a}-2.24\%$
test_select_nested	86.7220μs	55.6506μs	17.9693 KOps/s	18.4641 KOps/s	$\color{#d91a1a}-2.68\%$
test_exclude_nested	0.1517ms	0.1101ms	9.0866 KOps/s	9.1189 KOps/s	$\color{#d91a1a}-0.35\%$
test_empty[True]	0.4127ms	0.3481ms	2.8727 KOps/s	2.9031 KOps/s	$\color{#d91a1a}-1.05\%$
test_empty[False]	3.1670μs	0.9530μs	1.0493 MOps/s	1.0692 MOps/s	$\color{#d91a1a}-1.86\%$
test_to	0.1010ms	76.2662μs	13.1120 KOps/s	13.1711 KOps/s	$\color{#d91a1a}-0.45\%$
test_to_nonblocking	93.3210μs	61.8057μs	16.1797 KOps/s	16.0619 KOps/s	$\color{#35bf28}+0.73\%$
test_unbind_speed	1.4791ms	0.2687ms	3.7216 KOps/s	3.7851 KOps/s	$\color{#d91a1a}-1.68\%$
test_unbind_speed_stack0	0.3159ms	0.2684ms	3.7259 KOps/s	3.8319 KOps/s	$\color{#d91a1a}-2.76\%$
test_unbind_speed_stack1	75.7465ms	0.8042ms	1.2435 KOps/s	1.2626 KOps/s	$\color{#d91a1a}-1.52\%$
test_split	76.4065ms	1.6637ms	601.0695 Ops/s	598.0208 Ops/s	$\color{#35bf28}+0.51\%$
test_chunk	76.1278ms	1.6612ms	601.9825 Ops/s	600.0465 Ops/s	$\color{#35bf28}+0.32\%$
test_creation[device0]	0.1279ms	58.2331μs	17.1724 KOps/s	16.8620 KOps/s	$\color{#35bf28}+1.84\%$
test_creation_from_tensor	0.1286ms	54.7347μs	18.2700 KOps/s	18.4779 KOps/s	$\color{#d91a1a}-1.13\%$
test_add_one[memmap_tensor0]	65.4120μs	6.9965μs	142.9283 KOps/s	143.8599 KOps/s	$\color{#d91a1a}-0.65\%$
test_contiguous[memmap_tensor0]	23.5910μs	0.7143μs	1.4000 MOps/s	1.4758 MOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_stack[memmap_tensor0]	36.6800μs	5.0449μs	198.2192 KOps/s	201.9077 KOps/s	$\color{#d91a1a}-1.83\%$
test_memmaptd_index	0.5136ms	0.2926ms	3.4172 KOps/s	3.4302 KOps/s	$\color{#d91a1a}-0.38\%$
test_memmaptd_index_astensor	0.6573ms	0.3653ms	2.7371 KOps/s	2.7875 KOps/s	$\color{#d91a1a}-1.81\%$
test_memmaptd_index_op	1.2978ms	0.6930ms	1.4429 KOps/s	1.5946 KOps/s	$\textbf{\color{#d91a1a}-9.51\%}$
test_serialize_model	0.1816s	0.1102s	9.0739 Ops/s	8.4651 Ops/s	$\textbf{\color{#35bf28}+7.19\%}$
test_serialize_model_pickle	1.3487s	1.2360s	0.8090 Ops/s	0.8087 Ops/s	$\color{#35bf28}+0.04\%$
test_serialize_weights	0.1803s	0.1088s	9.1936 Ops/s	8.8520 Ops/s	$\color{#35bf28}+3.86\%$
test_serialize_weights_returnearly	0.2729s	0.1039s	9.6218 Ops/s	9.8012 Ops/s	$\color{#d91a1a}-1.83\%$
test_serialize_weights_pickle	1.3868s	1.2525s	0.7984 Ops/s	0.7970 Ops/s	$\color{#35bf28}+0.18\%$
test_reshape_pytree	0.2465ms	26.3664μs	37.9271 KOps/s	37.7697 KOps/s	$\color{#35bf28}+0.42\%$
test_reshape_td	52.7900μs	31.8387μs	31.4084 KOps/s	31.4455 KOps/s	$\color{#d91a1a}-0.12\%$
test_view_pytree	0.2276ms	26.0627μs	38.3691 KOps/s	37.7970 KOps/s	$\color{#35bf28}+1.51\%$
test_view_td	58.7410μs	35.9768μs	27.7957 KOps/s	27.1878 KOps/s	$\color{#35bf28}+2.24\%$
test_unbind_pytree	57.7100μs	32.0707μs	31.1811 KOps/s	30.7119 KOps/s	$\color{#35bf28}+1.53\%$
test_unbind_td	0.4209ms	41.0300μs	24.3724 KOps/s	24.7660 KOps/s	$\color{#d91a1a}-1.59\%$
test_split_pytree	57.4410μs	35.2803μs	28.3444 KOps/s	28.3405 KOps/s	$\color{#35bf28}+0.01\%$
test_split_td	0.1056ms	40.8907μs	24.4555 KOps/s	25.3880 KOps/s	$\color{#d91a1a}-3.67\%$
test_add_pytree	61.8410μs	37.7558μs	26.4860 KOps/s	26.5476 KOps/s	$\color{#d91a1a}-0.23\%$
test_add_td	0.2676ms	56.1302μs	17.8157 KOps/s	22.1084 KOps/s	$\textbf{\color{#d91a1a}-19.42\%}$
test_distributed	0.1929ms	66.5051μs	15.0364 KOps/s	14.8831 KOps/s	$\color{#35bf28}+1.03\%$
test_tdmodule	30.3600μs	15.7616μs	63.4453 KOps/s	70.5356 KOps/s	$\textbf{\color{#d91a1a}-10.05\%}$
test_tdmodule_dispatch	56.1610μs	30.7455μs	32.5251 KOps/s	36.9676 KOps/s	$\textbf{\color{#d91a1a}-12.02\%}$
test_tdseq	40.2000μs	17.6321μs	56.7149 KOps/s	62.5041 KOps/s	$\textbf{\color{#d91a1a}-9.26\%}$
test_tdseq_dispatch	0.2340ms	34.5700μs	28.9269 KOps/s	32.4874 KOps/s	$\textbf{\color{#d91a1a}-10.96\%}$
test_instantiation_functorch	1.7359ms	1.5458ms	646.8973 Ops/s	649.7274 Ops/s	$\color{#d91a1a}-0.44\%$
test_instantiation_td	1.5432ms	1.0417ms	959.9343 Ops/s	953.3234 Ops/s	$\color{#35bf28}+0.69\%$
test_exec_functorch	0.3607ms	0.1517ms	6.5917 KOps/s	6.6281 KOps/s	$\color{#d91a1a}-0.55\%$
test_exec_functional_call	0.1736ms	0.1356ms	7.3730 KOps/s	7.1842 KOps/s	$\color{#35bf28}+2.63\%$
test_exec_td	0.3331ms	0.1365ms	7.3241 KOps/s	7.3611 KOps/s	$\color{#d91a1a}-0.50\%$
test_exec_td_decorator	0.7415ms	0.2073ms	4.8249 KOps/s	4.8525 KOps/s	$\color{#d91a1a}-0.57\%$
test_vmap_mlp_speed[True-True]	0.7720ms	0.5804ms	1.7230 KOps/s	1.7550 KOps/s	$\color{#d91a1a}-1.82\%$
test_vmap_mlp_speed[True-False]	0.8190ms	0.5783ms	1.7292 KOps/s	1.7650 KOps/s	$\color{#d91a1a}-2.03\%$
test_vmap_mlp_speed[False-True]	0.6947ms	0.4982ms	2.0071 KOps/s	1.9639 KOps/s	$\color{#35bf28}+2.20\%$
test_vmap_mlp_speed[False-False]	0.6963ms	0.5030ms	1.9879 KOps/s	1.9255 KOps/s	$\color{#35bf28}+3.24\%$
test_vmap_mlp_speed_decorator[True-True]	1.1316ms	0.6348ms	1.5753 KOps/s	1.5749 KOps/s	$\color{#35bf28}+0.03\%$
test_vmap_mlp_speed_decorator[True-False]	0.8230ms	0.6348ms	1.5753 KOps/s	1.5782 KOps/s	$\color{#d91a1a}-0.19\%$
test_vmap_mlp_speed_decorator[False-True]	0.8814ms	0.5777ms	1.7309 KOps/s	1.7289 KOps/s	$\color{#35bf28}+0.11\%$
test_vmap_mlp_speed_decorator[False-False]	0.8107ms	0.5611ms	1.7822 KOps/s	1.6827 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_vmap_transformer_speed[True-True]	7.6620ms	7.4639ms	133.9790 Ops/s	126.9154 Ops/s	$\textbf{\color{#35bf28}+5.57\%}$
test_vmap_transformer_speed[True-False]	7.6857ms	7.4443ms	134.3303 Ops/s	129.3541 Ops/s	$\color{#35bf28}+3.85\%$
test_vmap_transformer_speed[False-True]	7.5400ms	7.3692ms	135.6991 Ops/s	130.6223 Ops/s	$\color{#35bf28}+3.89\%$
test_vmap_transformer_speed[False-False]	7.5936ms	7.3734ms	135.6232 Ops/s	130.2411 Ops/s	$\color{#35bf28}+4.13\%$
test_vmap_transformer_speed_decorator[True-True]	18.3660ms	18.0961ms	55.2605 Ops/s	53.6315 Ops/s	$\color{#35bf28}+3.04\%$
test_vmap_transformer_speed_decorator[True-False]	18.3861ms	18.1234ms	55.1774 Ops/s	53.6449 Ops/s	$\color{#35bf28}+2.86\%$
test_vmap_transformer_speed_decorator[False-True]	18.1394ms	18.0227ms	55.4856 Ops/s	53.9365 Ops/s	$\color{#35bf28}+2.87\%$
test_vmap_transformer_speed_decorator[False-False]	18.6512ms	17.9843ms	55.6040 Ops/s	53.9914 Ops/s	$\color{#35bf28}+2.99\%$
test_to_module_speed[True]	1.7921ms	1.5360ms	651.0611 Ops/s	635.8809 Ops/s	$\color{#35bf28}+2.39\%$
test_to_module_speed[False]	1.7130ms	1.5138ms	660.5840 Ops/s	646.0900 Ops/s	$\color{#35bf28}+2.24\%$
test_tc_init	48.4000μs	28.0543μs	35.6452 KOps/s	45.1343 KOps/s	$\textbf{\color{#d91a1a}-21.02\%}$
test_tc_init_nested	0.2803ms	57.4651μs	17.4019 KOps/s	22.4039 KOps/s	$\textbf{\color{#d91a1a}-22.33\%}$
test_tc_first_layer_tensor	6.1226μs	0.3565μs	2.8054 MOps/s	2.7745 MOps/s	$\color{#35bf28}+1.11\%$
test_tc_first_layer_nontensor	16.2326μs	0.3849μs	2.5980 MOps/s	2.5569 MOps/s	$\color{#35bf28}+1.61\%$
test_tc_second_layer_tensor	41.2628μs	0.9635μs	1.0379 MOps/s	1.0171 MOps/s	$\color{#35bf28}+2.05\%$
test_tc_second_layer_nontensor	11.2466μs	0.7931μs	1.2609 MOps/s	1.2126 MOps/s	$\color{#35bf28}+3.98\%$
test_unbind	0.1061s	9.1107ms	109.7612 Ops/s	121.2908 Ops/s	$\textbf{\color{#d91a1a}-9.51\%}$
test_full_like	11.5544ms	11.1641ms	89.5728 Ops/s	75.9560 Ops/s	$\textbf{\color{#35bf28}+17.93\%}$
test_zeros_like	8.4378ms	7.9201ms	126.2617 Ops/s	127.1458 Ops/s	$\color{#d91a1a}-0.70\%$
test_ones_like	8.6782ms	7.9518ms	125.7570 Ops/s	126.8979 Ops/s	$\color{#d91a1a}-0.90\%$
test_clone	10.1409ms	9.3540ms	106.9064 Ops/s	106.6972 Ops/s	$\color{#35bf28}+0.20\%$
test_squeeze	71.0400μs	11.2952μs	88.5329 KOps/s	89.5777 KOps/s	$\color{#d91a1a}-1.17\%$
test_unsqueeze	0.3198ms	55.0650μs	18.1604 KOps/s	19.1503 KOps/s	$\textbf{\color{#d91a1a}-5.17\%}$
test_split	0.2027ms	98.1774μs	10.1856 KOps/s	10.1744 KOps/s	$\color{#35bf28}+0.11\%$
test_permute	0.4014ms	0.1134ms	8.8173 KOps/s	9.1131 KOps/s	$\color{#d91a1a}-3.25\%$
test_stack	27.3211ms	26.8643ms	37.2241 Ops/s	37.2306 Ops/s	$\color{#d91a1a}-0.02\%$
test_cat	27.3210ms	26.8892ms	37.1896 Ops/s	37.1868 Ops/s	$+0.01\%$