[Feature] ProbabilisticTensorDictModule.num_samples

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] Expose WrapModule #1118
• -> [Feature] ProbabilisticTensorDictModule.num_samples #1117
• [Feature] Make ProbabilisticTensorDictSequential account for more than one distribution #1114

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}23$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.8130μs	18.2524μs	54.7872 KOps/s	59.5307 KOps/s	$\textbf{\color{#d91a1a}-7.97\%}$
test_plain_set_stack_nested	43.2700μs	18.0443μs	55.4193 KOps/s	58.5082 KOps/s	$\textbf{\color{#d91a1a}-5.28\%}$
test_plain_set_nested_inplace	71.8440μs	19.4496μs	51.4150 KOps/s	53.7465 KOps/s	$\color{#d91a1a}-4.34\%$
test_plain_set_stack_nested_inplace	69.1290μs	19.5045μs	51.2703 KOps/s	53.8101 KOps/s	$\color{#d91a1a}-4.72\%$
test_items	25.3370μs	4.2051μs	237.8049 KOps/s	244.0356 KOps/s	$\color{#d91a1a}-2.55\%$
test_items_nested	0.7224ms	0.3998ms	2.5015 KOps/s	2.3411 KOps/s	$\textbf{\color{#35bf28}+6.85\%}$
test_items_nested_locked	0.5838ms	0.3961ms	2.5245 KOps/s	2.3399 KOps/s	$\textbf{\color{#35bf28}+7.89\%}$
test_items_nested_leaf	0.1318ms	70.8946μs	14.1054 KOps/s	13.8725 KOps/s	$\color{#35bf28}+1.68\%$
test_items_stack_nested	0.5253ms	0.3971ms	2.5185 KOps/s	2.3377 KOps/s	$\textbf{\color{#35bf28}+7.73\%}$
test_items_stack_nested_leaf	0.1339ms	72.6385μs	13.7668 KOps/s	13.5332 KOps/s	$\color{#35bf28}+1.73\%$
test_items_stack_nested_locked	0.5819ms	0.3992ms	2.5049 KOps/s	2.3333 KOps/s	$\textbf{\color{#35bf28}+7.35\%}$
test_keys	36.6080μs	3.4956μs	286.0704 KOps/s	283.7522 KOps/s	$\color{#35bf28}+0.82\%$
test_keys_nested	0.1962ms	0.1404ms	7.1224 KOps/s	7.2584 KOps/s	$\color{#d91a1a}-1.87\%$
test_keys_nested_locked	1.9273ms	0.1456ms	6.8692 KOps/s	6.9430 KOps/s	$\color{#d91a1a}-1.06\%$
test_keys_nested_leaf	0.1746ms	0.1212ms	8.2485 KOps/s	8.4845 KOps/s	$\color{#d91a1a}-2.78\%$
test_keys_stack_nested	0.2384ms	0.1413ms	7.0777 KOps/s	7.2652 KOps/s	$\color{#d91a1a}-2.58\%$
test_keys_stack_nested_leaf	0.1820ms	0.1209ms	8.2702 KOps/s	8.4353 KOps/s	$\color{#d91a1a}-1.96\%$
test_keys_stack_nested_locked	0.2532ms	0.1463ms	6.8360 KOps/s	6.9681 KOps/s	$\color{#d91a1a}-1.89\%$
test_values	9.5752μs	1.0269μs	973.8070 KOps/s	955.0587 KOps/s	$\color{#35bf28}+1.96\%$
test_values_nested	0.1104ms	54.3817μs	18.3885 KOps/s	17.9714 KOps/s	$\color{#35bf28}+2.32\%$
test_values_nested_locked	0.1046ms	54.9767μs	18.1895 KOps/s	18.2341 KOps/s	$\color{#d91a1a}-0.24\%$
test_values_nested_leaf	0.1316ms	60.1315μs	16.6302 KOps/s	16.1195 KOps/s	$\color{#35bf28}+3.17\%$
test_values_stack_nested	0.1052ms	54.9453μs	18.1999 KOps/s	17.9018 KOps/s	$\color{#35bf28}+1.67\%$
test_values_stack_nested_leaf	0.1294ms	59.9279μs	16.6867 KOps/s	16.4702 KOps/s	$\color{#35bf28}+1.31\%$
test_values_stack_nested_locked	0.1029ms	55.0845μs	18.1539 KOps/s	17.8535 KOps/s	$\color{#35bf28}+1.68\%$
test_membership	2.5352μs	0.7009μs	1.4268 MOps/s	1.3003 MOps/s	$\textbf{\color{#35bf28}+9.72\%}$
test_membership_nested	22.9630μs	2.8711μs	348.2987 KOps/s	328.9494 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_membership_nested_leaf	41.9980μs	2.8952μs	345.3958 KOps/s	326.6830 KOps/s	$\textbf{\color{#35bf28}+5.73\%}$
test_membership_stacked_nested	24.0450μs	2.8842μs	346.7203 KOps/s	322.3288 KOps/s	$\textbf{\color{#35bf28}+7.57\%}$
test_membership_stacked_nested_leaf	42.3390μs	2.8887μs	346.1791 KOps/s	336.5287 KOps/s	$\color{#35bf28}+2.87\%$
test_membership_nested_last	35.9070μs	4.1289μs	242.1971 KOps/s	233.7822 KOps/s	$\color{#35bf28}+3.60\%$
test_membership_nested_leaf_last	49.5920μs	4.1566μs	240.5801 KOps/s	230.7274 KOps/s	$\color{#35bf28}+4.27\%$
test_membership_stacked_nested_last	29.2950μs	4.2142μs	237.2952 KOps/s	200.6822 KOps/s	$\textbf{\color{#35bf28}+18.24\%}$
test_membership_stacked_nested_leaf_last	24.3160μs	4.1495μs	240.9909 KOps/s	200.5599 KOps/s	$\textbf{\color{#35bf28}+20.16\%}$
test_nested_getleaf	54.4710μs	10.5203μs	95.0546 KOps/s	92.3385 KOps/s	$\color{#35bf28}+2.94\%$
test_nested_get	48.5400μs	10.0593μs	99.4108 KOps/s	95.2812 KOps/s	$\color{#35bf28}+4.33\%$
test_stacked_getleaf	53.7400μs	10.5088μs	95.1581 KOps/s	92.7577 KOps/s	$\color{#35bf28}+2.59\%$
test_stacked_get	34.6150μs	10.0469μs	99.5330 KOps/s	95.4647 KOps/s	$\color{#35bf28}+4.26\%$
test_nested_getitemleaf	46.9580μs	11.0127μs	90.8041 KOps/s	90.3196 KOps/s	$\color{#35bf28}+0.54\%$
test_nested_getitem	42.0780μs	10.2715μs	97.3566 KOps/s	91.4464 KOps/s	$\textbf{\color{#35bf28}+6.46\%}$
test_stacked_getitemleaf	38.6350μs	10.9249μs	91.5337 KOps/s	86.9850 KOps/s	$\textbf{\color{#35bf28}+5.23\%}$
test_stacked_getitem	50.2140μs	10.2674μs	97.3959 KOps/s	93.6195 KOps/s	$\color{#35bf28}+4.03\%$
test_lock_nested	3.3934ms	0.4415ms	2.2648 KOps/s	2.2835 KOps/s	$\color{#d91a1a}-0.82\%$
test_lock_stack_nested	0.8127ms	0.4090ms	2.4448 KOps/s	2.4547 KOps/s	$\color{#d91a1a}-0.41\%$
test_unlock_nested	0.6912ms	0.3554ms	2.8137 KOps/s	2.8259 KOps/s	$\color{#d91a1a}-0.43\%$
test_unlock_stack_nested	0.4938ms	0.3292ms	3.0372 KOps/s	3.0803 KOps/s	$\color{#d91a1a}-1.40\%$
test_flatten_speed	0.1749ms	94.8840μs	10.5392 KOps/s	10.5460 KOps/s	$\color{#d91a1a}-0.06\%$
test_unflatten_speed	0.6622ms	0.4929ms	2.0288 KOps/s	2.0153 KOps/s	$\color{#35bf28}+0.67\%$
test_common_ops	4.4752ms	0.7721ms	1.2951 KOps/s	1.3823 KOps/s	$\textbf{\color{#d91a1a}-6.31\%}$
test_creation	14.6680μs	2.0689μs	483.3409 KOps/s	482.7608 KOps/s	$\color{#35bf28}+0.12\%$
test_creation_empty	59.5780μs	10.8541μs	92.1313 KOps/s	111.6712 KOps/s	$\textbf{\color{#d91a1a}-17.50\%}$
test_creation_nested_1	44.0220μs	13.7161μs	72.9070 KOps/s	83.7736 KOps/s	$\textbf{\color{#d91a1a}-12.97\%}$
test_creation_nested_2	73.2090μs	17.9946μs	55.5722 KOps/s	62.4367 KOps/s	$\textbf{\color{#d91a1a}-10.99\%}$
test_clone	55.8740μs	12.6136μs	79.2792 KOps/s	76.1416 KOps/s	$\color{#35bf28}+4.12\%$
test_getitem[int]	1.4250ms	12.3882μs	80.7218 KOps/s	79.0219 KOps/s	$\color{#35bf28}+2.15\%$
test_getitem[slice_int]	0.1446ms	23.8752μs	41.8844 KOps/s	40.6862 KOps/s	$\color{#35bf28}+2.94\%$
test_getitem[range]	0.1667ms	47.8813μs	20.8850 KOps/s	21.3262 KOps/s	$\color{#d91a1a}-2.07\%$
test_getitem[tuple]	0.1276ms	19.6911μs	50.7843 KOps/s	50.5768 KOps/s	$\color{#35bf28}+0.41\%$
test_getitem[list]	0.1589ms	43.5905μs	22.9408 KOps/s	24.1395 KOps/s	$\color{#d91a1a}-4.97\%$
test_setitem_dim[int]	59.7920μs	25.4597μs	39.2778 KOps/s	41.6447 KOps/s	$\textbf{\color{#d91a1a}-5.68\%}$
test_setitem_dim[slice_int]	86.1910μs	52.8998μs	18.9037 KOps/s	19.8143 KOps/s	$\color{#d91a1a}-4.60\%$
test_setitem_dim[range]	0.1247ms	74.2751μs	13.4635 KOps/s	13.8407 KOps/s	$\color{#d91a1a}-2.73\%$
test_setitem_dim[tuple]	62.4060μs	41.7061μs	23.9773 KOps/s	25.0621 KOps/s	$\color{#d91a1a}-4.33\%$
test_setitem	72.2550μs	19.5952μs	51.0330 KOps/s	52.8904 KOps/s	$\color{#d91a1a}-3.51\%$
test_set	68.4580μs	19.3141μs	51.7755 KOps/s	55.7450 KOps/s	$\textbf{\color{#d91a1a}-7.12\%}$
test_set_shared	4.1360ms	0.1671ms	5.9834 KOps/s	6.0633 KOps/s	$\color{#d91a1a}-1.32\%$
test_update	0.1209ms	22.2422μs	44.9596 KOps/s	51.0783 KOps/s	$\textbf{\color{#d91a1a}-11.98\%}$
test_update_nested	84.7380μs	31.9217μs	31.3267 KOps/s	34.2815 KOps/s	$\textbf{\color{#d91a1a}-8.62\%}$
test_update__nested	1.0390ms	31.5517μs	31.6940 KOps/s	31.9920 KOps/s	$\color{#d91a1a}-0.93\%$
test_set_nested	88.1150μs	21.2252μs	47.1137 KOps/s	46.4948 KOps/s	$\color{#35bf28}+1.33\%$
test_set_nested_new	84.3670μs	25.7068μs	38.9002 KOps/s	41.2082 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_select	91.5110μs	43.2512μs	23.1208 KOps/s	24.6368 KOps/s	$\textbf{\color{#d91a1a}-6.15\%}$
test_select_nested	94.6870μs	59.4393μs	16.8239 KOps/s	16.6718 KOps/s	$\color{#35bf28}+0.91\%$
test_exclude_nested	0.1439ms	79.1256μs	12.6381 KOps/s	12.3891 KOps/s	$\color{#35bf28}+2.01\%$
test_empty[True]	0.5395ms	0.3872ms	2.5828 KOps/s	2.6001 KOps/s	$\color{#d91a1a}-0.66\%$
test_empty[False]	10.6573μs	1.2286μs	813.9233 KOps/s	834.9057 KOps/s	$\color{#d91a1a}-2.51\%$
test_unbind_speed	0.3662ms	0.2572ms	3.8878 KOps/s	3.8885 KOps/s	$\color{#d91a1a}-0.02\%$
test_unbind_speed_stack0	0.4439ms	0.2573ms	3.8869 KOps/s	3.9457 KOps/s	$\color{#d91a1a}-1.49\%$
test_unbind_speed_stack1	97.8296ms	0.7619ms	1.3126 KOps/s	1.4431 KOps/s	$\textbf{\color{#d91a1a}-9.04\%}$
test_split	97.4792ms	1.6947ms	590.0741 Ops/s	593.4278 Ops/s	$\color{#d91a1a}-0.57\%$
test_chunk	87.9642ms	1.6717ms	598.1798 Ops/s	589.9872 Ops/s	$\color{#35bf28}+1.39\%$
test_consolidate_njt[False-None]	8.2199ms	7.9504ms	125.7805 Ops/s	123.4681 Ops/s	$\color{#35bf28}+1.87\%$
test_creation[device0]	0.2287ms	89.7671μs	11.1399 KOps/s	11.1468 KOps/s	$\color{#d91a1a}-0.06\%$
test_creation_from_tensor	3.3382ms	93.1760μs	10.7324 KOps/s	10.6512 KOps/s	$\color{#35bf28}+0.76\%$
test_add_one[memmap_tensor0]	0.1694ms	5.0436μs	198.2715 KOps/s	205.8027 KOps/s	$\color{#d91a1a}-3.66\%$
test_contiguous[memmap_tensor0]	19.5760μs	0.5182μs	1.9298 MOps/s	1.9479 MOps/s	$\color{#d91a1a}-0.93\%$
test_stack[memmap_tensor0]	27.7720μs	3.4644μs	288.6464 KOps/s	301.3731 KOps/s	$\color{#d91a1a}-4.22\%$
test_memmaptd_index	1.0750ms	0.2310ms	4.3284 KOps/s	4.3668 KOps/s	$\color{#d91a1a}-0.88\%$
test_memmaptd_index_astensor	0.5692ms	0.3101ms	3.2245 KOps/s	3.2743 KOps/s	$\color{#d91a1a}-1.52\%$
test_memmaptd_index_op	0.9665ms	0.5655ms	1.7685 KOps/s	1.8775 KOps/s	$\textbf{\color{#d91a1a}-5.81\%}$
test_serialize_model	0.1188s	0.1146s	8.7256 Ops/s	7.5592 Ops/s	$\textbf{\color{#35bf28}+15.43\%}$
test_serialize_model_pickle	0.4972s	0.4023s	2.4854 Ops/s	2.5497 Ops/s	$\color{#d91a1a}-2.52\%$
test_serialize_weights	0.2147s	0.1282s	7.8014 Ops/s	8.9413 Ops/s	$\textbf{\color{#d91a1a}-12.75\%}$
test_serialize_weights_returnearly	0.1781s	0.1611s	6.2086 Ops/s	6.5059 Ops/s	$\color{#d91a1a}-4.57\%$
test_serialize_weights_pickle	0.6227s	0.4369s	2.2891 Ops/s	2.4428 Ops/s	$\textbf{\color{#d91a1a}-6.29\%}$
test_serialize_weights_filesystem	0.1501s	0.1422s	7.0347 Ops/s	6.4511 Ops/s	$\textbf{\color{#35bf28}+9.05\%}$
test_serialize_model_filesystem	0.1552s	0.1442s	6.9351 Ops/s	6.7147 Ops/s	$\color{#35bf28}+3.28\%$
test_reshape_pytree	60.5730μs	26.3182μs	37.9965 KOps/s	37.7020 KOps/s	$\color{#35bf28}+0.78\%$
test_reshape_td	78.2950μs	32.1377μs	31.1161 KOps/s	30.3092 KOps/s	$\color{#35bf28}+2.66\%$
test_view_pytree	58.5890μs	26.1650μs	38.2190 KOps/s	37.3686 KOps/s	$\color{#35bf28}+2.28\%$
test_view_td	89.4770μs	36.5136μs	27.3871 KOps/s	26.2341 KOps/s	$\color{#35bf28}+4.39\%$
test_unbind_pytree	87.0320μs	29.8989μs	33.4461 KOps/s	33.5391 KOps/s	$\color{#d91a1a}-0.28\%$
test_unbind_td	0.3852ms	38.1282μs	26.2273 KOps/s	26.1764 KOps/s	$\color{#35bf28}+0.19\%$
test_split_pytree	82.6160μs	29.1199μs	34.3408 KOps/s	34.0736 KOps/s	$\color{#35bf28}+0.78\%$
test_split_td	0.2059ms	42.9501μs	23.2829 KOps/s	22.8723 KOps/s	$\color{#35bf28}+1.80\%$
test_add_pytree	79.7690μs	35.4336μs	28.2218 KOps/s	28.2234 KOps/s	$-0.01\%$
test_add_td	0.1335ms	55.8890μs	17.8926 KOps/s	19.5790 KOps/s	$\textbf{\color{#d91a1a}-8.61\%}$
test_compile_add_one_nested[tensordict-compile]	0.1413ms	61.3440μs	16.3015 KOps/s	16.1544 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_add_one_nested[tensordict-eager]	0.3538ms	0.1604ms	6.2347 KOps/s	6.2130 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_add_one_nested[pytree-compile]	0.1127ms	45.2792μs	22.0852 KOps/s	22.0988 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_add_one_nested[pytree-eager]	0.2426ms	0.1185ms	8.4357 KOps/s	8.3438 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_copy_nested[tensordict-compile]	86.5520μs	26.0546μs	38.3810 KOps/s	38.7968 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_copy_nested[tensordict-eager]	0.1319ms	51.8233μs	19.2963 KOps/s	18.3065 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_compile_copy_nested[pytree-compile]	0.1793ms	76.5367μs	13.0656 KOps/s	12.5359 KOps/s	$\color{#35bf28}+4.23\%$
test_compile_copy_nested[pytree-eager]	0.1326ms	65.8795μs	15.1792 KOps/s	14.4202 KOps/s	$\textbf{\color{#35bf28}+5.26\%}$
test_compile_add_one_flat[tensordict-compile]	0.1778ms	0.1035ms	9.6577 KOps/s	9.6261 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_add_one_flat[tensordict-eager]	0.4123ms	0.2035ms	4.9141 KOps/s	5.0838 KOps/s	$\color{#d91a1a}-3.34\%$
test_compile_add_one_flat[tensorclass-compile]	0.1047ms	43.7201μs	22.8728 KOps/s	22.8524 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_add_one_flat[tensorclass-eager]	0.4973ms	62.2330μs	16.0686 KOps/s	16.4488 KOps/s	$\color{#d91a1a}-2.31\%$
test_compile_add_one_flat[pytree-compile]	0.1861ms	0.1052ms	9.5012 KOps/s	9.9428 KOps/s	$\color{#d91a1a}-4.44\%$
test_compile_add_one_flat[pytree-eager]	0.3874ms	0.2013ms	4.9682 KOps/s	4.8880 KOps/s	$\color{#35bf28}+1.64\%$
test_compile_add_self_flat[tensordict-eager]	0.4237ms	0.2128ms	4.6998 KOps/s	4.7581 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_add_self_flat[tensordict-compile]	0.2489ms	0.1048ms	9.5465 KOps/s	9.6542 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_add_self_flat[tensorclass-eager]	0.1250ms	54.3471μs	18.4002 KOps/s	18.7853 KOps/s	$\color{#d91a1a}-2.05\%$
test_compile_add_self_flat[tensorclass-compile]	95.0580μs	45.4906μs	21.9825 KOps/s	22.0079 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_add_self_flat[pytree-eager]	0.6027ms	0.1600ms	6.2499 KOps/s	6.2028 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_add_self_flat[pytree-compile]	0.2002ms	0.1029ms	9.7144 KOps/s	9.8346 KOps/s	$\color{#d91a1a}-1.22\%$
test_compile_copy_flat[tensordict-compile]	64.8410μs	21.2628μs	47.0304 KOps/s	48.1316 KOps/s	$\color{#d91a1a}-2.29\%$
test_compile_copy_flat[tensordict-eager]	0.1271ms	58.1761μs	17.1892 KOps/s	16.5246 KOps/s	$\color{#35bf28}+4.02\%$
test_compile_copy_flat[pytree-compile]	0.1618ms	80.2892μs	12.4550 KOps/s	12.3034 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_copy_flat[pytree-eager]	0.1458ms	68.7332μs	14.5490 KOps/s	14.1638 KOps/s	$\color{#35bf28}+2.72\%$
test_compile_assign_and_add[tensordict-compile]	0.3115ms	0.2091ms	4.7829 KOps/s	4.9536 KOps/s	$\color{#d91a1a}-3.45\%$
test_compile_assign_and_add[tensordict-eager]	1.4729ms	1.2738ms	785.0589 Ops/s	790.7157 Ops/s	$\color{#d91a1a}-0.72\%$
test_compile_assign_and_add[pytree-compile]	0.3857ms	0.2043ms	4.8948 KOps/s	5.0448 KOps/s	$\color{#d91a1a}-2.97\%$
test_compile_assign_and_add[pytree-eager]	0.9289ms	0.7728ms	1.2941 KOps/s	1.3054 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_assign_and_add_stack[compile]	0.6155ms	0.4528ms	2.2086 KOps/s	2.2075 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_assign_and_add_stack[eager]	4.7657ms	2.6528ms	376.9601 Ops/s	409.4325 Ops/s	$\textbf{\color{#d91a1a}-7.93\%}$
test_compile_indexing[tensor-tensordict-compile]	90.4880μs	34.9334μs	28.6259 KOps/s	28.8112 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_indexing[tensor-tensordict-eager]	0.4919ms	33.2559μs	30.0699 KOps/s	31.6532 KOps/s	$\textbf{\color{#d91a1a}-5.00\%}$
test_compile_indexing[tensor-tensorclass-compile]	80.0900μs	28.9171μs	34.5816 KOps/s	34.8967 KOps/s	$\color{#d91a1a}-0.90\%$
test_compile_indexing[tensor-tensorclass-eager]	70.0000μs	23.4024μs	42.7307 KOps/s	42.8303 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_indexing[tensor-pytree-compile]	78.2960μs	29.7591μs	33.6031 KOps/s	33.2838 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_indexing[tensor-pytree-eager]	98.4030μs	23.4975μs	42.5577 KOps/s	42.7414 KOps/s	$\color{#d91a1a}-0.43\%$
test_compile_indexing[slice-tensordict-compile]	94.5560μs	49.8975μs	20.0411 KOps/s	19.6790 KOps/s	$\color{#35bf28}+1.84\%$
test_compile_indexing[slice-tensordict-eager]	0.5086ms	19.2547μs	51.9353 KOps/s	50.4218 KOps/s	$\color{#35bf28}+3.00\%$
test_compile_indexing[slice-tensorclass-compile]	91.1100μs	42.9004μs	23.3098 KOps/s	22.4385 KOps/s	$\color{#35bf28}+3.88\%$
test_compile_indexing[slice-tensorclass-eager]	63.1880μs	18.4328μs	54.2511 KOps/s	52.0288 KOps/s	$\color{#35bf28}+4.27\%$
test_compile_indexing[slice-pytree-compile]	0.1086ms	43.9466μs	22.7549 KOps/s	22.1690 KOps/s	$\color{#35bf28}+2.64\%$
test_compile_indexing[slice-pytree-eager]	52.0170μs	18.3939μs	54.3657 KOps/s	51.5644 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_compile_indexing[int-tensordict-compile]	0.1182ms	51.0917μs	19.5726 KOps/s	19.2716 KOps/s	$\color{#35bf28}+1.56\%$
test_compile_indexing[int-tensordict-eager]	1.0338ms	19.2690μs	51.8969 KOps/s	50.3346 KOps/s	$\color{#35bf28}+3.10\%$
test_compile_indexing[int-tensorclass-compile]	0.1772ms	44.5571μs	22.4431 KOps/s	22.1694 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_indexing[int-tensorclass-eager]	69.4600μs	18.2895μs	54.6763 KOps/s	52.4223 KOps/s	$\color{#35bf28}+4.30\%$
test_compile_indexing[int-pytree-compile]	0.1024ms	45.0256μs	22.2096 KOps/s	22.4714 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_indexing[int-pytree-eager]	50.6250μs	18.0372μs	55.4408 KOps/s	52.7943 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_mod_add[eager]	75.0100μs	34.3369μs	29.1232 KOps/s	30.7822 KOps/s	$\textbf{\color{#d91a1a}-5.39\%}$
test_mod_add[compile]	89.6270μs	47.5171μs	21.0450 KOps/s	21.3551 KOps/s	$\color{#d91a1a}-1.45\%$
test_mod_add[compile-overhead]	0.1272ms	47.1411μs	21.2129 KOps/s	21.3232 KOps/s	$\color{#d91a1a}-0.52\%$
test_mod_wrap[eager]	0.3610ms	0.2222ms	4.5012 KOps/s	4.6039 KOps/s	$\color{#d91a1a}-2.23\%$
test_mod_wrap[compile]	0.6947ms	0.2109ms	4.7420 KOps/s	4.9249 KOps/s	$\color{#d91a1a}-3.71\%$
test_mod_wrap[compile-overhead]	0.3020ms	0.2039ms	4.9033 KOps/s	4.9888 KOps/s	$\color{#d91a1a}-1.71\%$
test_mod_wrap_and_backward[eager]	15.7557ms	11.6828ms	85.5959 Ops/s	94.2527 Ops/s	$\textbf{\color{#d91a1a}-9.18\%}$
test_mod_wrap_and_backward[compile]	15.9781ms	12.5261ms	79.8335 Ops/s	74.5663 Ops/s	$\textbf{\color{#35bf28}+7.06\%}$
test_mod_wrap_and_backward[compile-overhead]	18.4507ms	11.3747ms	87.9147 Ops/s	77.5154 Ops/s	$\textbf{\color{#35bf28}+13.42\%}$
test_seq_add[eager]	0.2586ms	0.1114ms	8.9768 KOps/s	9.2359 KOps/s	$\color{#d91a1a}-2.80\%$
test_seq_add[compile]	0.1461ms	62.6942μs	15.9504 KOps/s	16.8221 KOps/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_seq_add[compile-overhead]	0.1382ms	60.3633μs	16.5664 KOps/s	17.3029 KOps/s	$\color{#d91a1a}-4.26\%$
test_seq_wrap[eager]	0.6173ms	0.4441ms	2.2516 KOps/s	2.3518 KOps/s	$\color{#d91a1a}-4.26\%$
test_seq_wrap[compile]	0.4069ms	0.2250ms	4.4440 KOps/s	4.3953 KOps/s	$\color{#35bf28}+1.11\%$
test_seq_wrap[compile-overhead]	0.4344ms	0.2242ms	4.4612 KOps/s	4.4185 KOps/s	$\color{#35bf28}+0.97\%$
test_func_call_runtime[False-eager]	0.8441ms	0.5530ms	1.8082 KOps/s	1.8799 KOps/s	$\color{#d91a1a}-3.81\%$
test_func_call_runtime[False-compile]	0.7867ms	0.4260ms	2.3472 KOps/s	2.3793 KOps/s	$\color{#d91a1a}-1.35\%$
test_func_call_runtime[False-compile-overhead]	0.7327ms	0.4266ms	2.3442 KOps/s	2.3773 KOps/s	$\color{#d91a1a}-1.39\%$
test_func_call_runtime[True-eager]	1.2342ms	0.7712ms	1.2966 KOps/s	1.3501 KOps/s	$\color{#d91a1a}-3.96\%$
test_func_call_runtime[True-compile]	0.8553ms	0.4644ms	2.1532 KOps/s	2.1910 KOps/s	$\color{#d91a1a}-1.72\%$
test_func_call_runtime[True-compile-overhead]	0.7829ms	0.4634ms	2.1580 KOps/s	2.1774 KOps/s	$\color{#d91a1a}-0.89\%$
test_func_call_cm_runtime[False-eager]	0.9664ms	0.5482ms	1.8241 KOps/s	1.9000 KOps/s	$\color{#d91a1a}-3.99\%$
test_func_call_cm_runtime[False-compile]	0.6743ms	0.4265ms	2.3444 KOps/s	2.3739 KOps/s	$\color{#d91a1a}-1.24\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6031ms	0.4232ms	2.3631 KOps/s	2.3828 KOps/s	$\color{#d91a1a}-0.83\%$
test_func_call_cm_runtime[True-eager]	1.0198ms	0.8990ms	1.1124 KOps/s	1.1430 KOps/s	$\color{#d91a1a}-2.68\%$
test_func_call_cm_runtime[True-compile]	0.7435ms	0.4857ms	2.0587 KOps/s	2.0672 KOps/s	$\color{#d91a1a}-0.41\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6396ms	0.4869ms	2.0540 KOps/s	2.0570 KOps/s	$\color{#d91a1a}-0.15\%$
test_vmap_func_call_cm_runtime[eager]	2.4709ms	1.8671ms	535.5815 Ops/s	534.4212 Ops/s	$\color{#35bf28}+0.22\%$
test_vmap_func_call_cm_runtime[compile]	1.0066ms	0.5121ms	1.9526 KOps/s	1.9475 KOps/s	$\color{#35bf28}+0.26\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6993ms	0.5090ms	1.9647 KOps/s	1.9352 KOps/s	$\color{#35bf28}+1.52\%$
test_distributed	0.2502ms	0.1259ms	7.9439 KOps/s	7.8636 KOps/s	$\color{#35bf28}+1.02\%$
test_tdmodule	44.2930μs	25.4773μs	39.2507 KOps/s	40.9623 KOps/s	$\color{#d91a1a}-4.18\%$
test_tdmodule_dispatch	68.2270μs	47.1190μs	21.2228 KOps/s	22.2375 KOps/s	$\color{#d91a1a}-4.56\%$
test_tdseq	45.0650μs	25.5288μs	39.1714 KOps/s	39.8798 KOps/s	$\color{#d91a1a}-1.78\%$
test_tdseq_dispatch	71.2530μs	50.4563μs	19.8191 KOps/s	21.1636 KOps/s	$\textbf{\color{#d91a1a}-6.35\%}$
test_instantiation_functorch	2.5837ms	1.5423ms	648.3636 Ops/s	651.2866 Ops/s	$\color{#d91a1a}-0.45\%$
test_exec_functorch	0.4440ms	0.1818ms	5.5006 KOps/s	5.6647 KOps/s	$\color{#d91a1a}-2.90\%$
test_exec_functional_call	0.2913ms	0.1720ms	5.8155 KOps/s	5.8082 KOps/s	$\color{#35bf28}+0.13\%$
test_exec_td_decorator	0.8032ms	0.2309ms	4.3308 KOps/s	4.4008 KOps/s	$\color{#d91a1a}-1.59\%$
test_vmap_mlp_speed_decorator[True-True]	0.9037ms	0.6432ms	1.5548 KOps/s	1.5344 KOps/s	$\color{#35bf28}+1.33\%$
test_vmap_mlp_speed_decorator[True-False]	1.1195ms	0.6465ms	1.5468 KOps/s	1.5495 KOps/s	$\color{#d91a1a}-0.17\%$
test_vmap_mlp_speed_decorator[False-True]	0.9020ms	0.5204ms	1.9214 KOps/s	1.9036 KOps/s	$\color{#35bf28}+0.94\%$
test_vmap_mlp_speed_decorator[False-False]	0.7543ms	0.5214ms	1.9181 KOps/s	1.9089 KOps/s	$\color{#35bf28}+0.48\%$
test_to_module_speed[True]	1.6342ms	1.2780ms	782.4980 Ops/s	773.5001 Ops/s	$\color{#35bf28}+1.16\%$
test_to_module_speed[False]	1.8639ms	1.2505ms	799.6731 Ops/s	797.9767 Ops/s	$\color{#35bf28}+0.21\%$
test_tc_init	83.1250μs	44.9311μs	22.2563 KOps/s	22.5224 KOps/s	$\color{#d91a1a}-1.18\%$
test_tc_init_nested	0.1624ms	89.7266μs	11.1450 KOps/s	11.2434 KOps/s	$\color{#d91a1a}-0.88\%$
test_tc_first_layer_tensor	16.2700μs	1.5059μs	664.0336 KOps/s	665.3972 KOps/s	$\color{#d91a1a}-0.20\%$
test_tc_first_layer_nontensor	43.1680μs	4.6738μs	213.9590 KOps/s	209.4003 KOps/s	$\color{#35bf28}+2.18\%$
test_tc_second_layer_tensor	28.8260μs	2.7465μs	364.0979 KOps/s	356.0032 KOps/s	$\color{#35bf28}+2.27\%$
test_tc_second_layer_nontensor	25.5280μs	5.9156μs	169.0443 KOps/s	161.9859 KOps/s	$\color{#35bf28}+4.36\%$
test_unbind	0.2219s	12.5637ms	79.5942 Ops/s	80.2995 Ops/s	$\color{#d91a1a}-0.88\%$
test_full_like	8.9333ms	7.6589ms	130.5664 Ops/s	129.6677 Ops/s	$\color{#35bf28}+0.69\%$
test_zeros_like	3.5093ms	2.9387ms	340.2870 Ops/s	316.8009 Ops/s	$\textbf{\color{#35bf28}+7.41\%}$
test_ones_like	4.0139ms	3.3555ms	298.0215 Ops/s	295.7131 Ops/s	$\color{#35bf28}+0.78\%$
test_clone	6.3453ms	5.5272ms	180.9238 Ops/s	171.6295 Ops/s	$\textbf{\color{#35bf28}+5.42\%}$
test_squeeze	58.2280μs	11.9024μs	84.0167 KOps/s	85.7733 KOps/s	$\color{#d91a1a}-2.05\%$
test_unsqueeze	0.2926ms	88.0655μs	11.3552 KOps/s	11.3051 KOps/s	$\color{#35bf28}+0.44\%$
test_split	0.3124ms	0.1887ms	5.2983 KOps/s	5.1691 KOps/s	$\color{#35bf28}+2.50\%$
test_permute	0.3445ms	0.2146ms	4.6603 KOps/s	4.4971 KOps/s	$\color{#35bf28}+3.63\%$
test_stack	31.2211ms	27.5893ms	36.2460 Ops/s	36.4065 Ops/s	$\color{#d91a1a}-0.44\%$
test_cat	32.0271ms	27.5466ms	36.3021 Ops/s	37.8247 Ops/s	$\color{#d91a1a}-4.03\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}20$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.3510μs	9.9357μs	100.6476 KOps/s	92.3390 KOps/s	$\textbf{\color{#35bf28}+9.00\%}$
test_plain_set_stack_nested	32.2500μs	9.9517μs	100.4850 KOps/s	92.7040 KOps/s	$\textbf{\color{#35bf28}+8.39\%}$
test_plain_set_nested_inplace	35.3600μs	10.9052μs	91.6997 KOps/s	85.4620 KOps/s	$\textbf{\color{#35bf28}+7.30\%}$
test_plain_set_stack_nested_inplace	38.5010μs	10.8359μs	92.2857 KOps/s	86.4009 KOps/s	$\textbf{\color{#35bf28}+6.81\%}$
test_items	28.9900μs	2.8977μs	345.1036 KOps/s	340.2388 KOps/s	$\color{#35bf28}+1.43\%$
test_items_nested	0.4131ms	0.3578ms	2.7951 KOps/s	2.8347 KOps/s	$\color{#d91a1a}-1.40\%$
test_items_nested_locked	0.4421ms	0.3589ms	2.7862 KOps/s	2.8012 KOps/s	$\color{#d91a1a}-0.54\%$
test_items_nested_leaf	0.2225ms	58.1874μs	17.1859 KOps/s	17.2315 KOps/s	$\color{#d91a1a}-0.26\%$
test_items_stack_nested	0.3879ms	0.3577ms	2.7956 KOps/s	2.8084 KOps/s	$\color{#d91a1a}-0.46\%$
test_items_stack_nested_leaf	0.2406ms	59.3634μs	16.8454 KOps/s	17.1401 KOps/s	$\color{#d91a1a}-1.72\%$
test_items_stack_nested_locked	0.5282ms	0.3621ms	2.7620 KOps/s	2.7883 KOps/s	$\color{#d91a1a}-0.94\%$
test_keys	0.1855ms	3.4931μs	286.2802 KOps/s	290.1901 KOps/s	$\color{#d91a1a}-1.35\%$
test_keys_nested	0.1790ms	70.4208μs	14.2003 KOps/s	14.3998 KOps/s	$\color{#d91a1a}-1.39\%$
test_keys_nested_locked	0.7846ms	75.5123μs	13.2429 KOps/s	13.2308 KOps/s	$\color{#35bf28}+0.09\%$
test_keys_nested_leaf	95.4010μs	61.7106μs	16.2047 KOps/s	16.2991 KOps/s	$\color{#d91a1a}-0.58\%$
test_keys_stack_nested	0.1079ms	71.0212μs	14.0803 KOps/s	14.1966 KOps/s	$\color{#d91a1a}-0.82\%$
test_keys_stack_nested_leaf	88.5610μs	61.6142μs	16.2300 KOps/s	16.3191 KOps/s	$\color{#d91a1a}-0.55\%$
test_keys_stack_nested_locked	0.1258ms	76.2583μs	13.1133 KOps/s	13.2059 KOps/s	$\color{#d91a1a}-0.70\%$
test_values	4.7083μs	0.8540μs	1.1710 MOps/s	1.1781 MOps/s	$\color{#d91a1a}-0.60\%$
test_values_nested	72.5810μs	31.2869μs	31.9622 KOps/s	32.0912 KOps/s	$\color{#d91a1a}-0.40\%$
test_values_nested_locked	60.0810μs	33.8407μs	29.5502 KOps/s	30.5312 KOps/s	$\color{#d91a1a}-3.21\%$
test_values_nested_leaf	0.1557ms	33.3287μs	30.0042 KOps/s	29.8605 KOps/s	$\color{#35bf28}+0.48\%$
test_values_stack_nested	63.5510μs	31.8468μs	31.4003 KOps/s	31.6706 KOps/s	$\color{#d91a1a}-0.85\%$
test_values_stack_nested_leaf	60.0810μs	34.1037μs	29.3223 KOps/s	29.7010 KOps/s	$\color{#d91a1a}-1.27\%$
test_values_stack_nested_locked	63.1010μs	34.0368μs	29.3800 KOps/s	30.2765 KOps/s	$\color{#d91a1a}-2.96\%$
test_membership	1.6325μs	0.5071μs	1.9720 MOps/s	1.9555 MOps/s	$\color{#35bf28}+0.84\%$
test_membership_nested	13.9400μs	1.9930μs	501.7553 KOps/s	498.7240 KOps/s	$\color{#35bf28}+0.61\%$
test_membership_nested_leaf	17.6055μs	2.0358μs	491.2117 KOps/s	492.9482 KOps/s	$\color{#d91a1a}-0.35\%$
test_membership_stacked_nested	33.0410μs	2.0992μs	476.3685 KOps/s	485.6215 KOps/s	$\color{#d91a1a}-1.91\%$
test_membership_stacked_nested_leaf	23.7200μs	2.0879μs	478.9611 KOps/s	481.8746 KOps/s	$\color{#d91a1a}-0.60\%$
test_membership_nested_last	33.4110μs	2.9549μs	338.4218 KOps/s	339.0604 KOps/s	$\color{#d91a1a}-0.19\%$
test_membership_nested_leaf_last	0.1583ms	2.9145μs	343.1118 KOps/s	341.1922 KOps/s	$\color{#35bf28}+0.56\%$
test_membership_stacked_nested_last	80.6310μs	2.9241μs	341.9890 KOps/s	336.5874 KOps/s	$\color{#35bf28}+1.60\%$
test_membership_stacked_nested_leaf_last	38.5600μs	2.9230μs	342.1176 KOps/s	339.4754 KOps/s	$\color{#35bf28}+0.78\%$
test_nested_getleaf	28.7500μs	6.1448μs	162.7404 KOps/s	162.4165 KOps/s	$\color{#35bf28}+0.20\%$
test_nested_get	29.7400μs	5.8266μs	171.6263 KOps/s	171.1119 KOps/s	$\color{#35bf28}+0.30\%$
test_stacked_getleaf	35.0000μs	6.1295μs	163.1461 KOps/s	162.4668 KOps/s	$\color{#35bf28}+0.42\%$
test_stacked_get	33.5700μs	5.8697μs	170.3668 KOps/s	172.5168 KOps/s	$\color{#d91a1a}-1.25\%$
test_nested_getitemleaf	34.0100μs	6.2346μs	160.3958 KOps/s	160.7467 KOps/s	$\color{#d91a1a}-0.22\%$
test_nested_getitem	34.5700μs	5.9395μs	168.3635 KOps/s	168.6220 KOps/s	$\color{#d91a1a}-0.15\%$
test_stacked_getitemleaf	34.5710μs	6.2002μs	161.2864 KOps/s	161.0827 KOps/s	$\color{#35bf28}+0.13\%$
test_stacked_getitem	29.3000μs	5.9389μs	168.3816 KOps/s	168.8606 KOps/s	$\color{#d91a1a}-0.28\%$
test_lock_nested	9.6404ms	0.3825ms	2.6144 KOps/s	2.7177 KOps/s	$\color{#d91a1a}-3.80\%$
test_lock_stack_nested	0.4591ms	0.3420ms	2.9236 KOps/s	2.9926 KOps/s	$\color{#d91a1a}-2.31\%$
test_unlock_nested	0.6509ms	0.3128ms	3.1969 KOps/s	3.3110 KOps/s	$\color{#d91a1a}-3.45\%$
test_unlock_stack_nested	0.3156ms	0.2806ms	3.5639 KOps/s	3.6798 KOps/s	$\color{#d91a1a}-3.15\%$
test_flatten_speed	0.1789ms	74.8866μs	13.3535 KOps/s	13.2200 KOps/s	$\color{#35bf28}+1.01\%$
test_unflatten_speed	0.3487ms	0.3098ms	3.2276 KOps/s	3.2951 KOps/s	$\color{#d91a1a}-2.05\%$
test_common_ops	1.6314ms	0.5787ms	1.7281 KOps/s	1.6706 KOps/s	$\color{#35bf28}+3.44\%$
test_creation	78.2310μs	1.4902μs	671.0701 KOps/s	689.2597 KOps/s	$\color{#d91a1a}-2.64\%$
test_creation_empty	30.3910μs	6.0088μs	166.4232 KOps/s	128.6397 KOps/s	$\textbf{\color{#35bf28}+29.37\%}$
test_creation_nested_1	34.7600μs	7.6306μs	131.0508 KOps/s	107.7846 KOps/s	$\textbf{\color{#35bf28}+21.59\%}$
test_creation_nested_2	34.3310μs	10.1901μs	98.1344 KOps/s	84.2769 KOps/s	$\textbf{\color{#35bf28}+16.44\%}$
test_clone	0.1233ms	11.5017μs	86.9434 KOps/s	96.3121 KOps/s	$\textbf{\color{#d91a1a}-9.73\%}$
test_getitem[int]	1.5979ms	10.9437μs	91.3769 KOps/s	94.9589 KOps/s	$\color{#d91a1a}-3.77\%$
test_getitem[slice_int]	93.3793ms	30.5357μs	32.7485 KOps/s	48.1302 KOps/s	$\textbf{\color{#d91a1a}-31.96\%}$
test_getitem[range]	0.1882ms	38.7750μs	25.7898 KOps/s	26.6128 KOps/s	$\color{#d91a1a}-3.09\%$
test_getitem[tuple]	0.1124ms	19.0837μs	52.4006 KOps/s	55.1520 KOps/s	$\color{#d91a1a}-4.99\%$
test_getitem[list]	0.1973ms	34.6528μs	28.8577 KOps/s	30.4330 KOps/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_setitem_dim[int]	0.1293ms	20.2406μs	49.4057 KOps/s	54.0073 KOps/s	$\textbf{\color{#d91a1a}-8.52\%}$
test_setitem_dim[slice_int]	97.2710μs	39.6492μs	25.2212 KOps/s	26.2745 KOps/s	$\color{#d91a1a}-4.01\%$
test_setitem_dim[range]	0.1628ms	54.8799μs	18.2216 KOps/s	18.9993 KOps/s	$\color{#d91a1a}-4.09\%$
test_setitem_dim[tuple]	90.3510μs	33.6151μs	29.7485 KOps/s	31.1464 KOps/s	$\color{#d91a1a}-4.49\%$
test_setitem	76.9510μs	15.0856μs	66.2886 KOps/s	66.6526 KOps/s	$\color{#d91a1a}-0.55\%$
test_set	86.6810μs	14.3895μs	69.4950 KOps/s	69.8493 KOps/s	$\color{#d91a1a}-0.51\%$
test_set_shared	1.6198ms	0.1481ms	6.7524 KOps/s	6.7704 KOps/s	$\color{#d91a1a}-0.27\%$
test_update	0.3093ms	16.2501μs	61.5382 KOps/s	57.9157 KOps/s	$\textbf{\color{#35bf28}+6.25\%}$
test_update_nested	85.2410μs	21.5123μs	46.4850 KOps/s	45.8645 KOps/s	$\color{#35bf28}+1.35\%$
test_update__nested	0.5295ms	26.0403μs	38.4020 KOps/s	40.9848 KOps/s	$\textbf{\color{#d91a1a}-6.30\%}$
test_set_nested	76.2810μs	15.8286μs	63.1767 KOps/s	64.1168 KOps/s	$\color{#d91a1a}-1.47\%$
test_set_nested_new	0.1941ms	18.0566μs	55.3813 KOps/s	55.1027 KOps/s	$\color{#35bf28}+0.51\%$
test_select	0.2218ms	28.8441μs	34.6691 KOps/s	33.1489 KOps/s	$\color{#35bf28}+4.59\%$
test_select_nested	0.2245ms	41.8267μs	23.9082 KOps/s	24.3637 KOps/s	$\color{#d91a1a}-1.87\%$
test_exclude_nested	0.2537ms	62.4361μs	16.0164 KOps/s	16.4189 KOps/s	$\color{#d91a1a}-2.45\%$
test_empty[True]	0.4620ms	0.2762ms	3.6205 KOps/s	3.6261 KOps/s	$\color{#d91a1a}-0.15\%$
test_empty[False]	20.0593μs	0.7421μs	1.3475 MOps/s	1.3453 MOps/s	$\color{#35bf28}+0.16\%$
test_to	87.4710μs	56.2907μs	17.7649 KOps/s	15.9360 KOps/s	$\textbf{\color{#35bf28}+11.48\%}$
test_to_nonblocking	0.2161ms	47.9837μs	20.8404 KOps/s	21.5588 KOps/s	$\color{#d91a1a}-3.33\%$
test_unbind_speed	0.3298ms	0.2367ms	4.2249 KOps/s	4.3929 KOps/s	$\color{#d91a1a}-3.83\%$
test_unbind_speed_stack0	0.4129ms	0.2361ms	4.2348 KOps/s	4.3819 KOps/s	$\color{#d91a1a}-3.36\%$
test_unbind_speed_stack1	95.1182ms	0.6578ms	1.5202 KOps/s	1.5523 KOps/s	$\color{#d91a1a}-2.07\%$
test_split	93.3774ms	1.6297ms	613.6267 Ops/s	641.3184 Ops/s	$\color{#d91a1a}-4.32\%$
test_chunk	96.3623ms	1.6415ms	609.1941 Ops/s	588.1018 Ops/s	$\color{#35bf28}+3.59\%$
test_consolidate[False-None]	96.0018ms	2.8700ms	348.4279 Ops/s	379.7752 Ops/s	$\textbf{\color{#d91a1a}-8.25\%}$
test_consolidate[default-None]	1.8802ms	1.7122ms	584.0436 Ops/s	595.4387 Ops/s	$\color{#d91a1a}-1.91\%$
test_consolidate[reduce-overhead-None]	1.9046ms	1.7511ms	571.0728 Ops/s	579.8443 Ops/s	$\color{#d91a1a}-1.51\%$
test_consolidate_njt[False-None]	6.9470ms	6.6726ms	149.8661 Ops/s	153.2501 Ops/s	$\color{#d91a1a}-2.21\%$
test_to[False-False-None]	1.9328ms	1.7640ms	566.8854 Ops/s	585.1548 Ops/s	$\color{#d91a1a}-3.12\%$
test_to[True-False-None]	1.6291ms	1.3631ms	733.6110 Ops/s	755.1963 Ops/s	$\color{#d91a1a}-2.86\%$
test_to[within-False-None]	4.4668ms	4.1340ms	241.8991 Ops/s	245.5670 Ops/s	$\color{#d91a1a}-1.49\%$
test_to[True-default-None]	6.0380ms	5.4006ms	185.1633 Ops/s	177.4829 Ops/s	$\color{#35bf28}+4.33\%$
test_to_njt[False-False-None]	7.3587ms	7.0436ms	141.9724 Ops/s	136.7390 Ops/s	$\color{#35bf28}+3.83\%$
test_to_njt[True-False-None]	5.8612ms	5.5694ms	179.5519 Ops/s	179.9506 Ops/s	$\color{#d91a1a}-0.22\%$
test_to_njt[within-False-None]	13.3366ms	13.0496ms	76.6309 Ops/s	81.6099 Ops/s	$\textbf{\color{#d91a1a}-6.10\%}$
test_creation[device0]	0.5312ms	83.2703μs	12.0091 KOps/s	12.3222 KOps/s	$\color{#d91a1a}-2.54\%$
test_creation_from_tensor	0.6001ms	86.8578μs	11.5131 KOps/s	12.0431 KOps/s	$\color{#d91a1a}-4.40\%$
test_add_one[memmap_tensor0]	0.4332ms	7.3347μs	136.3379 KOps/s	146.0011 KOps/s	$\textbf{\color{#d91a1a}-6.62\%}$
test_contiguous[memmap_tensor0]	2.0255μs	0.4146μs	2.4120 MOps/s	2.4296 MOps/s	$\color{#d91a1a}-0.72\%$
test_stack[memmap_tensor0]	0.1519ms	4.7768μs	209.3468 KOps/s	223.5894 KOps/s	$\textbf{\color{#d91a1a}-6.37\%}$
test_memmaptd_index	1.6332ms	0.2573ms	3.8869 KOps/s	3.9526 KOps/s	$\color{#d91a1a}-1.66\%$
test_memmaptd_index_astensor	0.8041ms	0.3130ms	3.1951 KOps/s	3.2183 KOps/s	$\color{#d91a1a}-0.72\%$
test_memmaptd_index_op	1.0412ms	0.5799ms	1.7243 KOps/s	1.6813 KOps/s	$\color{#35bf28}+2.56\%$
test_serialize_model	0.1313s	0.1306s	7.6590 Ops/s	7.6983 Ops/s	$\color{#d91a1a}-0.51\%$
test_serialize_model_pickle	1.3500s	1.2118s	0.8252 Ops/s	0.8410 Ops/s	$\color{#d91a1a}-1.87\%$
test_serialize_weights	0.1313s	0.1299s	7.7012 Ops/s	7.6912 Ops/s	$\color{#35bf28}+0.13\%$
test_serialize_weights_returnearly	0.4676s	69.8374ms	14.3190 Ops/s	14.0725 Ops/s	$\color{#35bf28}+1.75\%$
test_serialize_weights_pickle	1.3760s	1.2215s	0.8187 Ops/s	0.8207 Ops/s	$\color{#d91a1a}-0.25\%$
test_reshape_pytree	0.1325ms	22.6313μs	44.1866 KOps/s	44.1087 KOps/s	$\color{#35bf28}+0.18\%$
test_reshape_td	54.0800μs	27.0730μs	36.9372 KOps/s	37.2632 KOps/s	$\color{#d91a1a}-0.87\%$
test_view_pytree	0.1695ms	22.4896μs	44.4650 KOps/s	44.5666 KOps/s	$\color{#d91a1a}-0.23\%$
test_view_td	0.1319ms	30.4743μs	32.8146 KOps/s	32.7224 KOps/s	$\color{#35bf28}+0.28\%$
test_unbind_pytree	0.1176ms	28.3572μs	35.2645 KOps/s	35.2949 KOps/s	$\color{#d91a1a}-0.09\%$
test_unbind_td	0.8115ms	36.2245μs	27.6057 KOps/s	27.1138 KOps/s	$\color{#35bf28}+1.81\%$
test_split_pytree	0.1711ms	30.5260μs	32.7589 KOps/s	32.8573 KOps/s	$\color{#d91a1a}-0.30\%$
test_split_td	1.0086ms	40.2191μs	24.8638 KOps/s	25.9432 KOps/s	$\color{#d91a1a}-4.16\%$
test_add_pytree	0.1486ms	35.9928μs	27.7833 KOps/s	28.7524 KOps/s	$\color{#d91a1a}-3.37\%$
test_add_td	0.1363ms	46.6569μs	21.4331 KOps/s	21.0855 KOps/s	$\color{#35bf28}+1.65\%$
test_compile_add_one_nested[tensordict-compile]	0.2714ms	0.1227ms	8.1477 KOps/s	7.9790 KOps/s	$\color{#35bf28}+2.12\%$
test_compile_add_one_nested[tensordict-eager]	0.2761ms	0.1249ms	8.0044 KOps/s	8.0364 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_add_one_nested[pytree-compile]	0.2463ms	97.1054μs	10.2981 KOps/s	10.0614 KOps/s	$\color{#35bf28}+2.35\%$
test_compile_add_one_nested[pytree-eager]	0.3021ms	0.1551ms	6.4486 KOps/s	6.6432 KOps/s	$\color{#d91a1a}-2.93\%$
test_compile_copy_nested[tensordict-compile]	0.1728ms	23.7380μs	42.1265 KOps/s	45.6348 KOps/s	$\textbf{\color{#d91a1a}-7.69\%}$
test_compile_copy_nested[tensordict-eager]	0.1469ms	27.2133μs	36.7467 KOps/s	37.0797 KOps/s	$\color{#d91a1a}-0.90\%$
test_compile_copy_nested[pytree-compile]	0.4069ms	64.9340μs	15.4002 KOps/s	15.3493 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_copy_nested[pytree-eager]	99.0020μs	49.3752μs	20.2531 KOps/s	20.1735 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_add_one_flat[tensordict-compile]	0.2996ms	0.1438ms	6.9533 KOps/s	6.8993 KOps/s	$\color{#35bf28}+0.78\%$
test_compile_add_one_flat[tensordict-eager]	0.4136ms	0.2113ms	4.7317 KOps/s	4.7990 KOps/s	$\color{#d91a1a}-1.40\%$
test_compile_add_one_flat[tensorclass-compile]	0.2459ms	99.7705μs	10.0230 KOps/s	10.1241 KOps/s	$\color{#d91a1a}-1.00\%$
test_compile_add_one_flat[tensorclass-eager]	0.2164ms	54.2096μs	18.4469 KOps/s	18.7563 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_add_one_flat[pytree-compile]	0.2774ms	0.1375ms	7.2713 KOps/s	6.9900 KOps/s	$\color{#35bf28}+4.02\%$
test_compile_add_one_flat[pytree-eager]	0.6610ms	0.4998ms	2.0009 KOps/s	1.9252 KOps/s	$\color{#35bf28}+3.93\%$
test_compile_add_self_flat[tensordict-eager]	0.3983ms	0.2504ms	3.9944 KOps/s	4.0148 KOps/s	$\color{#d91a1a}-0.51\%$
test_compile_add_self_flat[tensordict-compile]	0.2890ms	0.1458ms	6.8585 KOps/s	6.7291 KOps/s	$\color{#35bf28}+1.92\%$
test_compile_add_self_flat[tensorclass-eager]	0.2241ms	65.4544μs	15.2778 KOps/s	15.7814 KOps/s	$\color{#d91a1a}-3.19\%$
test_compile_add_self_flat[tensorclass-compile]	0.2859ms	0.1058ms	9.4544 KOps/s	9.7288 KOps/s	$\color{#d91a1a}-2.82\%$
test_compile_add_self_flat[pytree-eager]	0.5821ms	0.4204ms	2.3789 KOps/s	2.4204 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_add_self_flat[pytree-compile]	0.3317ms	0.1439ms	6.9508 KOps/s	7.3461 KOps/s	$\textbf{\color{#d91a1a}-5.38\%}$
test_compile_copy_flat[tensordict-compile]	0.2073ms	23.0498μs	43.3844 KOps/s	52.5312 KOps/s	$\textbf{\color{#d91a1a}-17.41\%}$
test_compile_copy_flat[tensordict-eager]	52.8610μs	26.6543μs	37.5173 KOps/s	37.1144 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_copy_flat[pytree-compile]	0.1155ms	69.9801μs	14.2898 KOps/s	14.3461 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_copy_flat[pytree-eager]	0.1602ms	51.5460μs	19.4002 KOps/s	19.3244 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_assign_and_add[tensordict-compile]	1.6252ms	0.3930ms	2.5448 KOps/s	2.0202 KOps/s	$\textbf{\color{#35bf28}+25.96\%}$
test_compile_assign_and_add[tensordict-eager]	3.0373ms	2.7189ms	367.8008 Ops/s	380.7693 Ops/s	$\color{#d91a1a}-3.41\%$
test_compile_assign_and_add[pytree-compile]	1.6038ms	0.4359ms	2.2939 KOps/s	2.2049 KOps/s	$\color{#35bf28}+4.03\%$
test_compile_assign_and_add[pytree-eager]	3.1994ms	2.8375ms	352.4265 Ops/s	373.9704 Ops/s	$\textbf{\color{#d91a1a}-5.76\%}$
test_compile_indexing[tensor-tensordict-compile]	0.3487ms	0.1227ms	8.1513 KOps/s	8.2499 KOps/s	$\color{#d91a1a}-1.20\%$
test_compile_indexing[tensor-tensordict-eager]	0.5680ms	83.7381μs	11.9420 KOps/s	11.8900 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5303ms	0.1132ms	8.8347 KOps/s	8.8832 KOps/s	$\color{#d91a1a}-0.55\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2493ms	73.9601μs	13.5208 KOps/s	13.7213 KOps/s	$\color{#d91a1a}-1.46\%$
test_compile_indexing[tensor-pytree-compile]	0.2844ms	0.1128ms	8.8614 KOps/s	8.8051 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_indexing[tensor-pytree-eager]	0.2393ms	71.1557μs	14.0537 KOps/s	13.8153 KOps/s	$\color{#35bf28}+1.73\%$
test_compile_indexing[slice-tensordict-compile]	0.2523ms	0.1033ms	9.6798 KOps/s	9.3712 KOps/s	$\color{#35bf28}+3.29\%$
test_compile_indexing[slice-tensordict-eager]	0.1651ms	18.6784μs	53.5379 KOps/s	56.4221 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_compile_indexing[slice-tensorclass-compile]	0.2473ms	98.7546μs	10.1261 KOps/s	10.0901 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_indexing[slice-tensorclass-eager]	0.1480ms	16.6734μs	59.9759 KOps/s	61.5076 KOps/s	$\color{#d91a1a}-2.49\%$
test_compile_indexing[slice-pytree-compile]	0.2559ms	98.9866μs	10.1024 KOps/s	9.6738 KOps/s	$\color{#35bf28}+4.43\%$
test_compile_indexing[slice-pytree-eager]	0.1461ms	16.6022μs	60.2331 KOps/s	62.6721 KOps/s	$\color{#d91a1a}-3.89\%$
test_compile_indexing[int-tensordict-compile]	0.3008ms	0.1082ms	9.2442 KOps/s	9.4827 KOps/s	$\color{#d91a1a}-2.52\%$
test_compile_indexing[int-tensordict-eager]	0.6451ms	18.1102μs	55.2175 KOps/s	57.3037 KOps/s	$\color{#d91a1a}-3.64\%$
test_compile_indexing[int-tensorclass-compile]	0.2622ms	0.1019ms	9.8153 KOps/s	9.7036 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_indexing[int-tensorclass-eager]	0.2559ms	17.2026μs	58.1308 KOps/s	62.8080 KOps/s	$\textbf{\color{#d91a1a}-7.45\%}$
test_compile_indexing[int-pytree-compile]	0.2502ms	98.9855μs	10.1025 KOps/s	9.8276 KOps/s	$\color{#35bf28}+2.80\%$
test_compile_indexing[int-pytree-eager]	0.1606ms	16.9901μs	58.8579 KOps/s	63.1049 KOps/s	$\textbf{\color{#d91a1a}-6.73\%}$
test_mod_add[eager]	0.1871ms	37.3469μs	26.7760 KOps/s	26.6565 KOps/s	$\color{#35bf28}+0.45\%$
test_mod_add[compile]	0.3115ms	82.5076μs	12.1201 KOps/s	12.1346 KOps/s	$\color{#d91a1a}-0.12\%$
test_mod_add[compile-overhead]	0.3211ms	0.1668ms	5.9964 KOps/s	5.6405 KOps/s	$\textbf{\color{#35bf28}+6.31\%}$
test_mod_wrap[eager]	0.3993ms	0.2534ms	3.9461 KOps/s	3.9124 KOps/s	$\color{#35bf28}+0.86\%$
test_mod_wrap[compile]	0.4418ms	0.2895ms	3.4547 KOps/s	3.4350 KOps/s	$\color{#35bf28}+0.57\%$
test_mod_wrap[compile-overhead]	7.5301ms	3.8304ms	261.0681 Ops/s	262.3746 Ops/s	$\color{#d91a1a}-0.50\%$
test_mod_wrap_and_backward[eager]	1.5950ms	1.3933ms	717.7305 Ops/s	675.6480 Ops/s	$\textbf{\color{#35bf28}+6.23\%}$
test_mod_wrap_and_backward[compile]	1.6460ms	1.2953ms	772.0297 Ops/s	706.0070 Ops/s	$\textbf{\color{#35bf28}+9.35\%}$
test_mod_wrap_and_backward[compile-overhead]	1.4025ms	0.9375ms	1.0667 KOps/s	925.3949 Ops/s	$\textbf{\color{#35bf28}+15.26\%}$
test_seq_add[eager]	0.2875ms	0.1155ms	8.6545 KOps/s	8.6897 KOps/s	$\color{#d91a1a}-0.41\%$
test_seq_add[compile]	0.2380ms	90.9785μs	10.9916 KOps/s	10.5024 KOps/s	$\color{#35bf28}+4.66\%$
test_seq_add[compile-overhead]	0.3108ms	0.1295ms	7.7191 KOps/s	7.6346 KOps/s	$\color{#35bf28}+1.11\%$
test_seq_wrap[eager]	0.6275ms	0.4396ms	2.2747 KOps/s	2.2280 KOps/s	$\color{#35bf28}+2.09\%$
test_seq_wrap[compile]	0.5040ms	0.3076ms	3.2511 KOps/s	3.2577 KOps/s	$\color{#d91a1a}-0.20\%$
test_seq_wrap[compile-overhead]	0.3805ms	0.2252ms	4.4414 KOps/s	4.4483 KOps/s	$\color{#d91a1a}-0.16\%$
test_func_call_runtime[False-eager]	0.9150ms	0.7558ms	1.3231 KOps/s	1.3135 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_runtime[False-compile]	0.9318ms	0.7603ms	1.3154 KOps/s	1.3176 KOps/s	$\color{#d91a1a}-0.17\%$
test_func_call_runtime[False-compile-overhead]	0.5008ms	0.3647ms	2.7417 KOps/s	2.7543 KOps/s	$\color{#d91a1a}-0.46\%$
test_func_call_runtime[True-eager]	1.0712ms	0.9220ms	1.0846 KOps/s	1.0883 KOps/s	$\color{#d91a1a}-0.34\%$
test_func_call_runtime[True-compile]	0.9345ms	0.7813ms	1.2799 KOps/s	1.2815 KOps/s	$\color{#d91a1a}-0.12\%$
test_func_call_runtime[True-compile-overhead]	0.5334ms	0.3854ms	2.5945 KOps/s	2.5959 KOps/s	$\color{#d91a1a}-0.05\%$
test_func_call_cm_runtime[False-eager]	0.9451ms	0.7573ms	1.3204 KOps/s	1.3105 KOps/s	$\color{#35bf28}+0.75\%$
test_func_call_cm_runtime[False-compile]	0.9173ms	0.7593ms	1.3170 KOps/s	1.3105 KOps/s	$\color{#35bf28}+0.49\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5109ms	0.3661ms	2.7314 KOps/s	2.7276 KOps/s	$\color{#35bf28}+0.14\%$
test_func_call_cm_runtime[True-eager]	1.1849ms	1.0168ms	983.4658 Ops/s	977.7996 Ops/s	$\color{#35bf28}+0.58\%$
test_func_call_cm_runtime[True-compile]	0.9795ms	0.8119ms	1.2317 KOps/s	1.2350 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5724ms	0.4124ms	2.4248 KOps/s	2.4176 KOps/s	$\color{#35bf28}+0.30\%$
test_vmap_func_call_cm_runtime[eager]	2.6368ms	2.1191ms	471.8971 Ops/s	471.5125 Ops/s	$\color{#35bf28}+0.08\%$
test_vmap_func_call_cm_runtime[compile]	1.2258ms	0.8221ms	1.2164 KOps/s	1.2169 KOps/s	$\color{#d91a1a}-0.04\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5796ms	0.4137ms	2.4171 KOps/s	2.4213 KOps/s	$\color{#d91a1a}-0.17\%$
test_distributed	5.4634ms	0.3414ms	2.9294 KOps/s	8.7765 KOps/s	$\textbf{\color{#d91a1a}-66.62\%}$
test_tdmodule	0.1240ms	18.2744μs	54.7215 KOps/s	51.3984 KOps/s	$\textbf{\color{#35bf28}+6.47\%}$
test_tdmodule_dispatch	0.1292ms	32.0881μs	31.1642 KOps/s	29.2498 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_tdseq	0.1975ms	17.8411μs	56.0503 KOps/s	52.8395 KOps/s	$\textbf{\color{#35bf28}+6.08\%}$
test_tdseq_dispatch	53.9710μs	33.7954μs	29.5899 KOps/s	27.3646 KOps/s	$\textbf{\color{#35bf28}+8.13\%}$
test_instantiation_functorch	1.8311ms	1.5689ms	637.4036 Ops/s	640.2591 Ops/s	$\color{#d91a1a}-0.45\%$
test_exec_functorch	0.3442ms	0.1486ms	6.7292 KOps/s	6.9330 KOps/s	$\color{#d91a1a}-2.94\%$
test_exec_functional_call	0.3502ms	0.1435ms	6.9662 KOps/s	7.1333 KOps/s	$\color{#d91a1a}-2.34\%$
test_exec_td_decorator	0.3924ms	0.1909ms	5.2372 KOps/s	5.3581 KOps/s	$\color{#d91a1a}-2.26\%$
test_vmap_mlp_speed_decorator[True-True]	0.8675ms	0.6858ms	1.4581 KOps/s	1.4484 KOps/s	$\color{#35bf28}+0.67\%$
test_vmap_mlp_speed_decorator[True-False]	0.9316ms	0.6972ms	1.4344 KOps/s	1.4292 KOps/s	$\color{#35bf28}+0.36\%$
test_vmap_mlp_speed_decorator[False-True]	0.8004ms	0.6141ms	1.6285 KOps/s	1.6559 KOps/s	$\color{#d91a1a}-1.66\%$
test_vmap_mlp_speed_decorator[False-False]	0.7523ms	0.6018ms	1.6616 KOps/s	1.6563 KOps/s	$\color{#35bf28}+0.32\%$
test_vmap_transformer_speed_decorator[True-True]	20.0857ms	19.5249ms	51.2167 Ops/s	51.3648 Ops/s	$\color{#d91a1a}-0.29\%$
test_vmap_transformer_speed_decorator[True-False]	20.3281ms	19.6504ms	50.8895 Ops/s	51.3163 Ops/s	$\color{#d91a1a}-0.83\%$
test_vmap_transformer_speed_decorator[False-True]	20.1927ms	19.4428ms	51.4329 Ops/s	51.7033 Ops/s	$\color{#d91a1a}-0.52\%$
test_vmap_transformer_speed_decorator[False-False]	20.3224ms	19.4788ms	51.3378 Ops/s	51.7244 Ops/s	$\color{#d91a1a}-0.75\%$
test_to_module_speed[True]	1.1055ms	0.9370ms	1.0672 KOps/s	1.0571 KOps/s	$\color{#35bf28}+0.96\%$
test_to_module_speed[False]	1.3737ms	0.9271ms	1.0787 KOps/s	1.0888 KOps/s	$\color{#d91a1a}-0.93\%$
test_tc_init	92.3110μs	33.6056μs	29.7570 KOps/s	27.6996 KOps/s	$\textbf{\color{#35bf28}+7.43\%}$
test_tc_init_nested	0.1253ms	66.8806μs	14.9520 KOps/s	13.6527 KOps/s	$\textbf{\color{#35bf28}+9.52\%}$
test_tc_first_layer_tensor	10.8701μs	0.6787μs	1.4733 MOps/s	1.4177 MOps/s	$\color{#35bf28}+3.93\%$
test_tc_first_layer_nontensor	24.0400μs	2.2733μs	439.8974 KOps/s	425.0373 KOps/s	$\color{#35bf28}+3.50\%$
test_tc_second_layer_tensor	7.4378μs	1.4105μs	708.9494 KOps/s	694.9952 KOps/s	$\color{#35bf28}+2.01\%$
test_tc_second_layer_nontensor	24.5600μs	3.0618μs	326.6082 KOps/s	328.0017 KOps/s	$\color{#d91a1a}-0.42\%$
test_unbind	0.2424s	10.0200ms	99.8006 Ops/s	150.0311 Ops/s	$\textbf{\color{#d91a1a}-33.48\%}$
test_full_like	12.4830ms	9.6874ms	103.2267 Ops/s	105.4291 Ops/s	$\color{#d91a1a}-2.09\%$
test_zeros_like	6.0053ms	4.3424ms	230.2848 Ops/s	232.9390 Ops/s	$\color{#d91a1a}-1.14\%$
test_ones_like	9.5164ms	7.2944ms	137.0914 Ops/s	235.8939 Ops/s	$\textbf{\color{#d91a1a}-41.88\%}$
test_clone	12.2790ms	9.2723ms	107.8477 Ops/s	152.3798 Ops/s	$\textbf{\color{#d91a1a}-29.22\%}$
test_squeeze	61.2910μs	9.4786μs	105.5013 KOps/s	109.4805 KOps/s	$\color{#d91a1a}-3.63\%$
test_unsqueeze	0.2162ms	72.6871μs	13.7576 KOps/s	13.6228 KOps/s	$\color{#35bf28}+0.99\%$
test_split	0.4004ms	0.1629ms	6.1400 KOps/s	6.2696 KOps/s	$\color{#d91a1a}-2.07\%$
test_permute	0.3328ms	0.1859ms	5.3804 KOps/s	5.5514 KOps/s	$\color{#d91a1a}-3.08\%$
test_stack	53.9124ms	51.7321ms	19.3303 Ops/s	19.4467 Ops/s	$\color{#d91a1a}-0.60\%$
test_cat	53.6635ms	51.3420ms	19.4772 Ops/s	19.5573 Ops/s	$\color{#d91a1a}-0.41\%$