[BugFix] Fix unitary ops for tensorclass

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] broadcast pointwise ops for tensor/tensordict mixed inputs #1166
• [Feature] TensorDict.clamp #1165
• -> [BugFix] Fix unitary ops for tensorclass #1164
• [Feature] TensorDict.softmax #1163
• [Feature] TensorDict.logsumexp #1162

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}40$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	45.9160μs	21.5347μs	46.4366 KOps/s	48.7287 KOps/s	$\color{#d91a1a}-4.70\%$
test_plain_set_stack_nested	57.0570μs	21.5723μs	46.3558 KOps/s	48.6383 KOps/s	$\color{#d91a1a}-4.69\%$
test_plain_set_nested_inplace	73.6680μs	23.5956μs	42.3808 KOps/s	44.9699 KOps/s	$\textbf{\color{#d91a1a}-5.76\%}$
test_plain_set_stack_nested_inplace	70.0580μs	23.3657μs	42.7978 KOps/s	45.5762 KOps/s	$\textbf{\color{#d91a1a}-6.10\%}$
test_items	23.6150μs	4.2643μs	234.5036 KOps/s	242.1309 KOps/s	$\color{#d91a1a}-3.15\%$
test_items_nested	0.8422ms	0.4019ms	2.4881 KOps/s	2.4851 KOps/s	$\color{#35bf28}+0.12\%$
test_items_nested_locked	0.5562ms	0.4017ms	2.4891 KOps/s	2.4955 KOps/s	$\color{#d91a1a}-0.26\%$
test_items_nested_leaf	0.1374ms	77.7329μs	12.8646 KOps/s	12.5505 KOps/s	$\color{#35bf28}+2.50\%$
test_items_stack_nested	0.5415ms	0.4040ms	2.4751 KOps/s	2.4692 KOps/s	$\color{#35bf28}+0.24\%$
test_items_stack_nested_leaf	0.1439ms	80.4928μs	12.4235 KOps/s	12.6349 KOps/s	$\color{#d91a1a}-1.67\%$
test_items_stack_nested_locked	0.5771ms	0.4034ms	2.4788 KOps/s	2.4689 KOps/s	$\color{#35bf28}+0.40\%$
test_keys	42.4890μs	3.9302μs	254.4393 KOps/s	283.4070 KOps/s	$\textbf{\color{#d91a1a}-10.22\%}$
test_keys_nested	0.2673ms	0.1645ms	6.0809 KOps/s	5.9793 KOps/s	$\color{#35bf28}+1.70\%$
test_keys_nested_locked	0.7049ms	0.1728ms	5.7879 KOps/s	5.7798 KOps/s	$\color{#35bf28}+0.14\%$
test_keys_nested_leaf	0.2002ms	0.1439ms	6.9486 KOps/s	6.8030 KOps/s	$\color{#35bf28}+2.14\%$
test_keys_stack_nested	0.2955ms	0.1635ms	6.1167 KOps/s	5.9742 KOps/s	$\color{#35bf28}+2.39\%$
test_keys_stack_nested_leaf	0.2386ms	0.1421ms	7.0358 KOps/s	6.9128 KOps/s	$\color{#35bf28}+1.78\%$
test_keys_stack_nested_locked	0.3280ms	0.1689ms	5.9198 KOps/s	5.6847 KOps/s	$\color{#35bf28}+4.14\%$
test_values	8.3556μs	1.0398μs	961.7021 KOps/s	956.4555 KOps/s	$\color{#35bf28}+0.55\%$
test_values_nested	0.1470ms	64.3737μs	15.5343 KOps/s	15.8201 KOps/s	$\color{#d91a1a}-1.81\%$
test_values_nested_locked	0.1154ms	63.7035μs	15.6977 KOps/s	15.3685 KOps/s	$\color{#35bf28}+2.14\%$
test_values_nested_leaf	0.1599ms	73.0381μs	13.6915 KOps/s	13.8578 KOps/s	$\color{#d91a1a}-1.20\%$
test_values_stack_nested	0.1109ms	64.9679μs	15.3922 KOps/s	15.7209 KOps/s	$\color{#d91a1a}-2.09\%$
test_values_stack_nested_leaf	0.1385ms	73.0556μs	13.6882 KOps/s	13.7864 KOps/s	$\color{#d91a1a}-0.71\%$
test_values_stack_nested_locked	0.1163ms	64.8984μs	15.4087 KOps/s	15.7904 KOps/s	$\color{#d91a1a}-2.42\%$
test_membership	4.7160μs	0.7192μs	1.3904 MOps/s	1.3574 MOps/s	$\color{#35bf28}+2.43\%$
test_membership_nested	33.2420μs	2.8777μs	347.5045 KOps/s	335.4113 KOps/s	$\color{#35bf28}+3.61\%$
test_membership_nested_leaf	42.7600μs	2.9247μs	341.9186 KOps/s	346.2525 KOps/s	$\color{#d91a1a}-1.25\%$
test_membership_stacked_nested	27.8220μs	2.9174μs	342.7656 KOps/s	342.9114 KOps/s	$\color{#d91a1a}-0.04\%$
test_membership_stacked_nested_leaf	33.4030μs	2.9000μs	344.8218 KOps/s	340.0150 KOps/s	$\color{#35bf28}+1.41\%$
test_membership_nested_last	31.5600μs	4.3187μs	231.5511 KOps/s	231.6628 KOps/s	$\color{#d91a1a}-0.05\%$
test_membership_nested_leaf_last	30.8780μs	4.3681μs	228.9329 KOps/s	230.5081 KOps/s	$\color{#d91a1a}-0.68\%$
test_membership_stacked_nested_last	33.5730μs	7.0679μs	141.4845 KOps/s	194.0921 KOps/s	$\textbf{\color{#d91a1a}-27.10\%}$
test_membership_stacked_nested_leaf_last	37.7710μs	7.0140μs	142.5720 KOps/s	195.4212 KOps/s	$\textbf{\color{#d91a1a}-27.04\%}$
test_nested_getleaf	47.0180μs	10.7790μs	92.7729 KOps/s	93.0242 KOps/s	$\color{#d91a1a}-0.27\%$
test_nested_get	42.7400μs	10.1160μs	98.8537 KOps/s	98.4610 KOps/s	$\color{#35bf28}+0.40\%$
test_stacked_getleaf	41.2870μs	10.8084μs	92.5209 KOps/s	92.9894 KOps/s	$\color{#d91a1a}-0.50\%$
test_stacked_get	33.5230μs	10.1399μs	98.6203 KOps/s	98.8342 KOps/s	$\color{#d91a1a}-0.22\%$
test_nested_getitemleaf	50.3640μs	11.1105μs	90.0047 KOps/s	88.7372 KOps/s	$\color{#35bf28}+1.43\%$
test_nested_getitem	36.6390μs	10.4825μs	95.3966 KOps/s	94.6494 KOps/s	$\color{#35bf28}+0.79\%$
test_stacked_getitemleaf	36.1080μs	11.0526μs	90.4768 KOps/s	89.2396 KOps/s	$\color{#35bf28}+1.39\%$
test_stacked_getitem	45.6660μs	10.4680μs	95.5296 KOps/s	96.5254 KOps/s	$\color{#d91a1a}-1.03\%$
test_lock_nested	0.9208ms	0.4573ms	2.1866 KOps/s	2.2306 KOps/s	$\color{#d91a1a}-1.97\%$
test_lock_stack_nested	0.6606ms	0.4273ms	2.3405 KOps/s	2.3594 KOps/s	$\color{#d91a1a}-0.80\%$
test_unlock_nested	0.7811ms	0.3789ms	2.6393 KOps/s	2.6971 KOps/s	$\color{#d91a1a}-2.14\%$
test_unlock_stack_nested	0.5276ms	0.3432ms	2.9136 KOps/s	2.9307 KOps/s	$\color{#d91a1a}-0.58\%$
test_flatten_speed	0.3060ms	0.1059ms	9.4438 KOps/s	10.0688 KOps/s	$\textbf{\color{#d91a1a}-6.21\%}$
test_unflatten_speed	1.2180ms	0.5331ms	1.8757 KOps/s	1.8836 KOps/s	$\color{#d91a1a}-0.42\%$
test_common_ops	2.0489ms	0.8411ms	1.1889 KOps/s	1.3270 KOps/s	$\textbf{\color{#d91a1a}-10.41\%}$
test_creation	19.3860μs	2.5396μs	393.7627 KOps/s	354.3792 KOps/s	$\textbf{\color{#35bf28}+11.11\%}$
test_creation_empty	41.2870μs	13.0814μs	76.4442 KOps/s	90.8394 KOps/s	$\textbf{\color{#d91a1a}-15.85\%}$
test_creation_nested_1	60.9240μs	16.0096μs	62.4627 KOps/s	71.4434 KOps/s	$\textbf{\color{#d91a1a}-12.57\%}$
test_creation_nested_2	60.4330μs	20.3441μs	49.1543 KOps/s	54.7174 KOps/s	$\textbf{\color{#d91a1a}-10.17\%}$
test_clone	68.9590μs	13.9708μs	71.5777 KOps/s	75.0323 KOps/s	$\color{#d91a1a}-4.60\%$
test_getitem[int]	1.2968ms	13.0549μs	76.5998 KOps/s	77.4360 KOps/s	$\color{#d91a1a}-1.08\%$
test_getitem[slice_int]	0.1400ms	25.1611μs	39.7439 KOps/s	40.9997 KOps/s	$\color{#d91a1a}-3.06\%$
test_getitem[range]	0.1716ms	51.2746μs	19.5029 KOps/s	20.8608 KOps/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_getitem[tuple]	0.1528ms	20.5354μs	48.6965 KOps/s	49.7644 KOps/s	$\color{#d91a1a}-2.15\%$
test_getitem[list]	0.2454ms	46.7496μs	21.3906 KOps/s	22.9559 KOps/s	$\textbf{\color{#d91a1a}-6.82\%}$
test_setitem_dim[int]	58.0390μs	26.3207μs	37.9929 KOps/s	39.8511 KOps/s	$\color{#d91a1a}-4.66\%$
test_setitem_dim[slice_int]	90.2690μs	52.1139μs	19.1888 KOps/s	19.7096 KOps/s	$\color{#d91a1a}-2.64\%$
test_setitem_dim[range]	0.1330ms	74.8685μs	13.3567 KOps/s	13.7820 KOps/s	$\color{#d91a1a}-3.09\%$
test_setitem_dim[tuple]	77.3650μs	41.2586μs	24.2374 KOps/s	24.9901 KOps/s	$\color{#d91a1a}-3.01\%$
test_setitem	68.9490μs	22.3723μs	44.6980 KOps/s	51.1501 KOps/s	$\textbf{\color{#d91a1a}-12.61\%}$
test_set	0.2297ms	21.8401μs	45.7874 KOps/s	51.9641 KOps/s	$\textbf{\color{#d91a1a}-11.89\%}$
test_set_shared	1.1692ms	0.1686ms	5.9303 KOps/s	5.8341 KOps/s	$\color{#35bf28}+1.65\%$
test_update	0.2100ms	25.4277μs	39.3273 KOps/s	45.4886 KOps/s	$\textbf{\color{#d91a1a}-13.54\%}$
test_update_nested	0.1857ms	36.5535μs	27.3571 KOps/s	31.1857 KOps/s	$\textbf{\color{#d91a1a}-12.28\%}$
test_update__nested	0.7188ms	35.5680μs	28.1152 KOps/s	29.7958 KOps/s	$\textbf{\color{#d91a1a}-5.64\%}$
test_set_nested	0.1318ms	23.9003μs	41.8405 KOps/s	47.2281 KOps/s	$\textbf{\color{#d91a1a}-11.41\%}$
test_set_nested_new	0.1547ms	28.6634μs	34.8876 KOps/s	38.4418 KOps/s	$\textbf{\color{#d91a1a}-9.25\%}$
test_select	95.0180μs	44.2952μs	22.5758 KOps/s	23.7434 KOps/s	$\color{#d91a1a}-4.92\%$
test_select_nested	0.1472ms	62.5114μs	15.9971 KOps/s	15.9159 KOps/s	$\color{#35bf28}+0.51\%$
test_exclude_nested	0.1715ms	80.5936μs	12.4079 KOps/s	12.2701 KOps/s	$\color{#35bf28}+1.12\%$
test_empty[True]	0.7411ms	0.4126ms	2.4237 KOps/s	2.4340 KOps/s	$\color{#d91a1a}-0.42\%$
test_empty[False]	12.9197μs	1.3802μs	724.5389 KOps/s	713.9013 KOps/s	$\color{#35bf28}+1.49\%$
test_unbind_speed	0.5727ms	0.2679ms	3.7323 KOps/s	3.6984 KOps/s	$\color{#35bf28}+0.92\%$
test_unbind_speed_stack0	0.4098ms	0.2636ms	3.7932 KOps/s	3.7603 KOps/s	$\color{#35bf28}+0.87\%$
test_unbind_speed_stack1	97.1881ms	0.7752ms	1.2899 KOps/s	1.3741 KOps/s	$\textbf{\color{#d91a1a}-6.13\%}$
test_split	1.7245ms	1.5980ms	625.7881 Ops/s	570.7525 Ops/s	$\textbf{\color{#35bf28}+9.64\%}$
test_chunk	0.1011s	1.9399ms	515.5011 Ops/s	566.9413 Ops/s	$\textbf{\color{#d91a1a}-9.07\%}$
test_consolidate_njt[False-None]	8.7853ms	8.1935ms	122.0475 Ops/s	122.9360 Ops/s	$\color{#d91a1a}-0.72\%$
test_creation[device0]	4.1043ms	92.3630μs	10.8268 KOps/s	11.0654 KOps/s	$\color{#d91a1a}-2.16\%$
test_creation_from_tensor	0.2553ms	92.9740μs	10.7557 KOps/s	10.5423 KOps/s	$\color{#35bf28}+2.02\%$
test_add_one[memmap_tensor0]	0.2091ms	5.1132μs	195.5711 KOps/s	209.3116 KOps/s	$\textbf{\color{#d91a1a}-6.56\%}$
test_contiguous[memmap_tensor0]	19.4560μs	0.5160μs	1.9378 MOps/s	1.9454 MOps/s	$\color{#d91a1a}-0.39\%$
test_stack[memmap_tensor0]	28.0930μs	3.4068μs	293.5309 KOps/s	303.7055 KOps/s	$\color{#d91a1a}-3.35\%$
test_memmaptd_index	0.9609ms	0.2478ms	4.0348 KOps/s	4.1473 KOps/s	$\color{#d91a1a}-2.71\%$
test_memmaptd_index_astensor	0.8328ms	0.3370ms	2.9675 KOps/s	3.0311 KOps/s	$\color{#d91a1a}-2.10\%$
test_memmaptd_index_op	1.1723ms	0.6387ms	1.5657 KOps/s	1.7584 KOps/s	$\textbf{\color{#d91a1a}-10.96\%}$
test_serialize_model	0.1247s	0.1142s	8.7548 Ops/s	8.5664 Ops/s	$\color{#35bf28}+2.20\%$
test_serialize_model_pickle	0.4430s	0.3858s	2.5921 Ops/s	2.5505 Ops/s	$\color{#35bf28}+1.63\%$
test_serialize_weights	0.2137s	0.1282s	7.7978 Ops/s	7.7421 Ops/s	$\color{#35bf28}+0.72\%$
test_serialize_weights_returnearly	0.1733s	0.1593s	6.2769 Ops/s	6.6238 Ops/s	$\textbf{\color{#d91a1a}-5.24\%}$
test_serialize_weights_pickle	0.5405s	0.4513s	2.2158 Ops/s	1.1100 Ops/s	$\textbf{\color{#35bf28}+99.62\%}$
test_serialize_weights_filesystem	0.1431s	0.1393s	7.1795 Ops/s	7.1615 Ops/s	$\color{#35bf28}+0.25\%$
test_serialize_model_filesystem	0.2337s	0.1557s	6.4207 Ops/s	6.4001 Ops/s	$\color{#35bf28}+0.32\%$
test_reshape_pytree	60.5230μs	26.3098μs	38.0086 KOps/s	37.8387 KOps/s	$\color{#35bf28}+0.45\%$
test_reshape_td	82.3840μs	33.7398μs	29.6386 KOps/s	30.4706 KOps/s	$\color{#d91a1a}-2.73\%$
test_view_pytree	82.0640μs	26.4710μs	37.7772 KOps/s	37.6683 KOps/s	$\color{#35bf28}+0.29\%$
test_view_td	82.0440μs	39.3637μs	25.4041 KOps/s	25.9240 KOps/s	$\color{#d91a1a}-2.01\%$
test_unbind_pytree	83.3960μs	29.9709μs	33.3657 KOps/s	33.8430 KOps/s	$\color{#d91a1a}-1.41\%$
test_unbind_td	0.3551ms	39.4115μs	25.3733 KOps/s	25.3580 KOps/s	$\color{#35bf28}+0.06\%$
test_split_pytree	75.6910μs	29.1812μs	34.2687 KOps/s	34.3841 KOps/s	$\color{#d91a1a}-0.34\%$
test_split_td	0.5176ms	45.3547μs	22.0484 KOps/s	21.9783 KOps/s	$\color{#35bf28}+0.32\%$
test_add_pytree	79.8390μs	35.7160μs	27.9987 KOps/s	28.8904 KOps/s	$\color{#d91a1a}-3.09\%$
test_add_td	0.1300ms	62.2248μs	16.0708 KOps/s	17.8986 KOps/s	$\textbf{\color{#d91a1a}-10.21\%}$
test_compile_add_one_nested[tensordict-compile]	0.1207ms	61.7820μs	16.1859 KOps/s	16.7744 KOps/s	$\color{#d91a1a}-3.51\%$
test_compile_add_one_nested[tensordict-eager]	0.3621ms	0.1707ms	5.8580 KOps/s	5.8768 KOps/s	$\color{#d91a1a}-0.32\%$
test_compile_add_one_nested[pytree-compile]	0.1025ms	45.3846μs	22.0339 KOps/s	22.5176 KOps/s	$\color{#d91a1a}-2.15\%$
test_compile_add_one_nested[pytree-eager]	0.4079ms	0.1201ms	8.3291 KOps/s	8.4607 KOps/s	$\color{#d91a1a}-1.56\%$
test_compile_copy_nested[tensordict-compile]	66.3540μs	25.7245μs	38.8734 KOps/s	39.7519 KOps/s	$\color{#d91a1a}-2.21\%$
test_compile_copy_nested[tensordict-eager]	0.1098ms	58.6877μs	17.0393 KOps/s	17.3810 KOps/s	$\color{#d91a1a}-1.97\%$
test_compile_copy_nested[pytree-compile]	0.1651ms	78.0993μs	12.8042 KOps/s	12.6305 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_copy_nested[pytree-eager]	0.2620ms	67.2934μs	14.8603 KOps/s	14.5013 KOps/s	$\color{#35bf28}+2.48\%$
test_compile_add_one_flat[tensordict-compile]	0.1823ms	0.1040ms	9.6143 KOps/s	9.6180 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_add_one_flat[tensordict-eager]	0.4827ms	0.2165ms	4.6182 KOps/s	4.6168 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_add_one_flat[tensorclass-compile]	97.7740μs	44.6002μs	22.4214 KOps/s	23.3097 KOps/s	$\color{#d91a1a}-3.81\%$
test_compile_add_one_flat[tensorclass-eager]	0.5078ms	66.1122μs	15.1258 KOps/s	15.7837 KOps/s	$\color{#d91a1a}-4.17\%$
test_compile_add_one_flat[pytree-compile]	0.2197ms	0.1032ms	9.6907 KOps/s	9.8218 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_add_one_flat[pytree-eager]	0.6620ms	0.2004ms	4.9909 KOps/s	5.0155 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_add_self_flat[tensordict-eager]	0.4371ms	0.2352ms	4.2509 KOps/s	4.2811 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_add_self_flat[tensordict-compile]	0.4004ms	0.1111ms	9.0031 KOps/s	9.6507 KOps/s	$\textbf{\color{#d91a1a}-6.71\%}$
test_compile_add_self_flat[tensorclass-eager]	0.2694ms	61.7303μs	16.1995 KOps/s	17.4790 KOps/s	$\textbf{\color{#d91a1a}-7.32\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1699ms	45.5328μs	21.9622 KOps/s	22.7918 KOps/s	$\color{#d91a1a}-3.64\%$
test_compile_add_self_flat[pytree-eager]	0.6018ms	0.1566ms	6.3864 KOps/s	6.2409 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_add_self_flat[pytree-compile]	0.4189ms	0.1032ms	9.6859 KOps/s	9.6991 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_copy_flat[tensordict-compile]	56.7070μs	21.1680μs	47.2411 KOps/s	46.2371 KOps/s	$\color{#35bf28}+2.17\%$
test_compile_copy_flat[tensordict-eager]	0.1505ms	66.7893μs	14.9725 KOps/s	15.1008 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_copy_flat[pytree-compile]	0.1595ms	78.5067μs	12.7378 KOps/s	12.6608 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_copy_flat[pytree-eager]	0.1405ms	66.5822μs	15.0190 KOps/s	14.6611 KOps/s	$\color{#35bf28}+2.44\%$
test_compile_assign_and_add[tensordict-compile]	0.3063ms	0.2043ms	4.8942 KOps/s	4.8532 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_assign_and_add[tensordict-eager]	1.5127ms	1.3095ms	763.6507 Ops/s	747.6470 Ops/s	$\color{#35bf28}+2.14\%$
test_compile_assign_and_add[pytree-compile]	0.2843ms	0.2023ms	4.9429 KOps/s	4.9467 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_assign_and_add[pytree-eager]	1.3826ms	0.7722ms	1.2951 KOps/s	1.3008 KOps/s	$\color{#d91a1a}-0.44\%$
test_compile_assign_and_add_stack[compile]	0.8197ms	0.4654ms	2.1489 KOps/s	2.1825 KOps/s	$\color{#d91a1a}-1.54\%$
test_compile_assign_and_add_stack[eager]	2.9731ms	2.8395ms	352.1694 Ops/s	380.1681 Ops/s	$\textbf{\color{#d91a1a}-7.36\%}$
test_compile_indexing[tensor-tensordict-compile]	93.2450μs	35.9194μs	27.8401 KOps/s	29.1301 KOps/s	$\color{#d91a1a}-4.43\%$
test_compile_indexing[tensor-tensordict-eager]	0.7802ms	34.6507μs	28.8594 KOps/s	30.5642 KOps/s	$\textbf{\color{#d91a1a}-5.58\%}$
test_compile_indexing[tensor-tensorclass-compile]	74.4990μs	28.9757μs	34.5116 KOps/s	35.0006 KOps/s	$\color{#d91a1a}-1.40\%$
test_compile_indexing[tensor-tensorclass-eager]	64.5910μs	23.3219μs	42.8782 KOps/s	43.0911 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_indexing[tensor-pytree-compile]	92.2630μs	30.2134μs	33.0979 KOps/s	33.7431 KOps/s	$\color{#d91a1a}-1.91\%$
test_compile_indexing[tensor-pytree-eager]	91.4110μs	22.8565μs	43.7512 KOps/s	43.9227 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_indexing[slice-tensordict-compile]	0.1136ms	50.7104μs	19.7198 KOps/s	19.5846 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_indexing[slice-tensordict-eager]	0.3884ms	20.2524μs	49.3769 KOps/s	49.2554 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_indexing[slice-tensorclass-compile]	96.8210μs	43.8131μs	22.8242 KOps/s	22.7377 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_indexing[slice-tensorclass-eager]	64.4810μs	18.9792μs	52.6891 KOps/s	53.4690 KOps/s	$\color{#d91a1a}-1.46\%$
test_compile_indexing[slice-pytree-compile]	0.1333ms	44.8176μs	22.3127 KOps/s	22.3032 KOps/s	$\color{#35bf28}+0.04\%$
test_compile_indexing[slice-pytree-eager]	81.3220μs	18.4043μs	54.3350 KOps/s	52.7068 KOps/s	$\color{#35bf28}+3.09\%$
test_compile_indexing[int-tensordict-compile]	0.1420ms	52.4638μs	19.0608 KOps/s	19.1199 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_indexing[int-tensordict-eager]	1.1865ms	20.5354μs	48.6963 KOps/s	48.9697 KOps/s	$\color{#d91a1a}-0.56\%$
test_compile_indexing[int-tensorclass-compile]	93.8760μs	44.7127μs	22.3650 KOps/s	22.0618 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_indexing[int-tensorclass-eager]	84.9800μs	18.5362μs	53.9484 KOps/s	53.9186 KOps/s	$\color{#35bf28}+0.06\%$
test_compile_indexing[int-pytree-compile]	0.1211ms	44.9389μs	22.2525 KOps/s	22.4236 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_indexing[int-pytree-eager]	78.8080μs	19.9540μs	50.1153 KOps/s	54.0156 KOps/s	$\textbf{\color{#d91a1a}-7.22\%}$
test_mod_add[eager]	99.7470μs	36.2041μs	27.6212 KOps/s	29.7191 KOps/s	$\textbf{\color{#d91a1a}-7.06\%}$
test_mod_add[compile]	0.1239ms	47.0391μs	21.2589 KOps/s	21.4165 KOps/s	$\color{#d91a1a}-0.74\%$
test_mod_add[compile-overhead]	0.1165ms	46.6396μs	21.4410 KOps/s	21.3739 KOps/s	$\color{#35bf28}+0.31\%$
test_mod_wrap[eager]	0.4957ms	0.2217ms	4.5105 KOps/s	4.5306 KOps/s	$\color{#d91a1a}-0.44\%$
test_mod_wrap[compile]	0.3779ms	0.2040ms	4.9025 KOps/s	4.7010 KOps/s	$\color{#35bf28}+4.29\%$
test_mod_wrap[compile-overhead]	0.3794ms	0.2041ms	4.8998 KOps/s	4.8180 KOps/s	$\color{#35bf28}+1.70\%$
test_mod_wrap_and_backward[eager]	19.7789ms	13.4508ms	74.3450 Ops/s	91.2975 Ops/s	$\textbf{\color{#d91a1a}-18.57\%}$
test_mod_wrap_and_backward[compile]	17.9690ms	13.7586ms	72.6816 Ops/s	91.3368 Ops/s	$\textbf{\color{#d91a1a}-20.42\%}$
test_mod_wrap_and_backward[compile-overhead]	16.1809ms	12.7986ms	78.1336 Ops/s	90.0115 Ops/s	$\textbf{\color{#d91a1a}-13.20\%}$
test_seq_add[eager]	0.2535ms	0.1190ms	8.4064 KOps/s	9.0148 KOps/s	$\textbf{\color{#d91a1a}-6.75\%}$
test_seq_add[compile]	0.1436ms	62.9323μs	15.8901 KOps/s	16.6019 KOps/s	$\color{#d91a1a}-4.29\%$
test_seq_add[compile-overhead]	0.1323ms	60.3165μs	16.5792 KOps/s	17.0709 KOps/s	$\color{#d91a1a}-2.88\%$
test_seq_wrap[eager]	0.7574ms	0.4453ms	2.2455 KOps/s	2.2775 KOps/s	$\color{#d91a1a}-1.40\%$
test_seq_wrap[compile]	0.4276ms	0.2288ms	4.3702 KOps/s	4.3542 KOps/s	$\color{#35bf28}+0.37\%$
test_seq_wrap[compile-overhead]	0.3624ms	0.2263ms	4.4199 KOps/s	4.3979 KOps/s	$\color{#35bf28}+0.50\%$
test_func_call_runtime[False-eager]	0.9382ms	0.5495ms	1.8197 KOps/s	1.8877 KOps/s	$\color{#d91a1a}-3.60\%$
test_func_call_runtime[False-compile]	0.8049ms	0.4349ms	2.2993 KOps/s	2.3602 KOps/s	$\color{#d91a1a}-2.58\%$
test_func_call_runtime[False-compile-overhead]	0.8050ms	0.4362ms	2.2924 KOps/s	2.3540 KOps/s	$\color{#d91a1a}-2.62\%$
test_func_call_runtime[True-eager]	0.9598ms	0.7739ms	1.2921 KOps/s	1.3224 KOps/s	$\color{#d91a1a}-2.29\%$
test_func_call_runtime[True-compile]	0.8744ms	0.4769ms	2.0971 KOps/s	2.1481 KOps/s	$\color{#d91a1a}-2.38\%$
test_func_call_runtime[True-compile-overhead]	0.5757ms	0.4726ms	2.1161 KOps/s	2.1426 KOps/s	$\color{#d91a1a}-1.24\%$
test_func_call_cm_runtime[False-eager]	0.9843ms	0.5693ms	1.7566 KOps/s	1.8798 KOps/s	$\textbf{\color{#d91a1a}-6.56\%}$
test_func_call_cm_runtime[False-compile]	0.7758ms	0.4301ms	2.3248 KOps/s	2.3226 KOps/s	$\color{#35bf28}+0.10\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5814ms	0.4287ms	2.3329 KOps/s	2.3626 KOps/s	$\color{#d91a1a}-1.26\%$
test_func_call_cm_runtime[True-eager]	1.3737ms	0.9173ms	1.0901 KOps/s	1.1275 KOps/s	$\color{#d91a1a}-3.31\%$
test_func_call_cm_runtime[True-compile]	0.6033ms	0.4950ms	2.0200 KOps/s	2.0337 KOps/s	$\color{#d91a1a}-0.67\%$
test_func_call_cm_runtime[True-compile-overhead]	0.7711ms	0.4961ms	2.0158 KOps/s	2.0465 KOps/s	$\color{#d91a1a}-1.50\%$
test_vmap_func_call_cm_runtime[eager]	2.4427ms	1.8987ms	526.6645 Ops/s	516.3873 Ops/s	$\color{#35bf28}+1.99\%$
test_vmap_func_call_cm_runtime[compile]	1.0666ms	0.5196ms	1.9246 KOps/s	1.9060 KOps/s	$\color{#35bf28}+0.97\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7428ms	0.5161ms	1.9375 KOps/s	1.9095 KOps/s	$\color{#35bf28}+1.46\%$
test_distributed	0.2587ms	0.1255ms	7.9692 KOps/s	7.8066 KOps/s	$\color{#35bf28}+2.08\%$
test_tdmodule	55.4840μs	26.3921μs	37.8901 KOps/s	39.3800 KOps/s	$\color{#d91a1a}-3.78\%$
test_tdmodule_dispatch	75.8820μs	48.1950μs	20.7491 KOps/s	19.7307 KOps/s	$\textbf{\color{#35bf28}+5.16\%}$
test_tdseq	47.0590μs	29.7543μs	33.6086 KOps/s	34.8220 KOps/s	$\color{#d91a1a}-3.48\%$
test_tdseq_dispatch	84.2180μs	54.9245μs	18.2068 KOps/s	19.0674 KOps/s	$\color{#d91a1a}-4.51\%$
test_instantiation_functorch	2.8232ms	1.5573ms	642.1433 Ops/s	646.1144 Ops/s	$\color{#d91a1a}-0.61\%$
test_exec_functorch	0.3311ms	0.1813ms	5.5154 KOps/s	5.5769 KOps/s	$\color{#d91a1a}-1.10\%$
test_exec_functional_call	0.4311ms	0.1753ms	5.7037 KOps/s	5.8307 KOps/s	$\color{#d91a1a}-2.18\%$
test_exec_td_decorator	0.4446ms	0.2341ms	4.2722 KOps/s	4.2527 KOps/s	$\color{#35bf28}+0.46\%$
test_vmap_mlp_speed_decorator[True-True]	0.7653ms	0.6514ms	1.5352 KOps/s	1.5174 KOps/s	$\color{#35bf28}+1.17\%$
test_vmap_mlp_speed_decorator[True-False]	0.8656ms	0.6592ms	1.5171 KOps/s	1.5239 KOps/s	$\color{#d91a1a}-0.45\%$
test_vmap_mlp_speed_decorator[False-True]	0.8028ms	0.5285ms	1.8920 KOps/s	1.8436 KOps/s	$\color{#35bf28}+2.63\%$
test_vmap_mlp_speed_decorator[False-False]	0.8091ms	0.5311ms	1.8828 KOps/s	1.8605 KOps/s	$\color{#35bf28}+1.20\%$
test_to_module_speed[True]	1.6530ms	1.3324ms	750.5155 Ops/s	731.5214 Ops/s	$\color{#35bf28}+2.60\%$
test_to_module_speed[False]	1.8099ms	1.2999ms	769.2620 Ops/s	757.6721 Ops/s	$\color{#35bf28}+1.53\%$
test_tc_init	81.9640μs	49.1734μs	20.3362 KOps/s	21.3715 KOps/s	$\color{#d91a1a}-4.84\%$
test_tc_init_nested	0.1911ms	96.8871μs	10.3213 KOps/s	10.6348 KOps/s	$\color{#d91a1a}-2.95\%$
test_tc_first_layer_tensor	17.1220μs	1.5608μs	640.6984 KOps/s	665.0398 KOps/s	$\color{#d91a1a}-3.66\%$
test_tc_first_layer_nontensor	24.4750μs	4.8530μs	206.0595 KOps/s	213.0473 KOps/s	$\color{#d91a1a}-3.28\%$
test_tc_second_layer_tensor	34.5140μs	2.9004μs	344.7837 KOps/s	349.1528 KOps/s	$\color{#d91a1a}-1.25\%$
test_tc_second_layer_nontensor	26.5700μs	6.2665μs	159.5797 KOps/s	165.6032 KOps/s	$\color{#d91a1a}-3.64\%$
test_unbind	0.2159s	13.2058ms	75.7244 Ops/s	77.8041 Ops/s	$\color{#d91a1a}-2.67\%$
test_full_like	19.2066ms	10.9520ms	91.3076 Ops/s	132.1902 Ops/s	$\textbf{\color{#d91a1a}-30.93\%}$
test_zeros_like	12.3861ms	7.2682ms	137.5851 Ops/s	315.3309 Ops/s	$\textbf{\color{#d91a1a}-56.37\%}$
test_ones_like	12.0888ms	7.2139ms	138.6220 Ops/s	162.3271 Ops/s	$\textbf{\color{#d91a1a}-14.60\%}$
test_clone	16.1671ms	8.7794ms	113.9027 Ops/s	125.3611 Ops/s	$\textbf{\color{#d91a1a}-9.14\%}$
test_squeeze	58.7910μs	12.2639μs	81.5400 KOps/s	80.7259 KOps/s	$\color{#35bf28}+1.01\%$
test_unsqueeze	0.2426ms	93.0476μs	10.7472 KOps/s	11.0722 KOps/s	$\color{#d91a1a}-2.94\%$
test_split	0.4347ms	0.2003ms	4.9923 KOps/s	5.1709 KOps/s	$\color{#d91a1a}-3.45\%$
test_permute	0.3053ms	0.2044ms	4.8916 KOps/s	4.9106 KOps/s	$\color{#d91a1a}-0.39\%$
test_stack	26.5248ms	24.2563ms	41.2264 Ops/s	39.9277 Ops/s	$\color{#35bf28}+3.25\%$
test_cat	26.0377ms	23.8779ms	41.8797 Ops/s	40.2992 Ops/s	$\color{#35bf28}+3.92\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}61$. Worsened: $\large\color{#d91a1a}1$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	32.4810μs	11.3650μs	87.9893 KOps/s	76.6110 KOps/s	$\textbf{\color{#35bf28}+14.85\%}$
test_plain_set_stack_nested	38.4210μs	11.6777μs	85.6334 KOps/s	75.0563 KOps/s	$\textbf{\color{#35bf28}+14.09\%}$
test_plain_set_nested_inplace	37.7110μs	12.6588μs	78.9966 KOps/s	69.6893 KOps/s	$\textbf{\color{#35bf28}+13.36\%}$
test_plain_set_stack_nested_inplace	36.7400μs	12.6816μs	78.8547 KOps/s	69.9451 KOps/s	$\textbf{\color{#35bf28}+12.74\%}$
test_items	25.5710μs	2.8688μs	348.5822 KOps/s	343.8163 KOps/s	$\color{#35bf28}+1.39\%$
test_items_nested	0.4174ms	0.3592ms	2.7841 KOps/s	2.8332 KOps/s	$\color{#d91a1a}-1.73\%$
test_items_nested_locked	0.4163ms	0.3596ms	2.7810 KOps/s	2.8239 KOps/s	$\color{#d91a1a}-1.52\%$
test_items_nested_leaf	90.1220μs	58.9224μs	16.9715 KOps/s	17.1641 KOps/s	$\color{#d91a1a}-1.12\%$
test_items_stack_nested	0.4321ms	0.3643ms	2.7451 KOps/s	2.7892 KOps/s	$\color{#d91a1a}-1.58\%$
test_items_stack_nested_leaf	0.1090ms	59.6081μs	16.7763 KOps/s	16.2109 KOps/s	$\color{#35bf28}+3.49\%$
test_items_stack_nested_locked	0.4108ms	0.3642ms	2.7456 KOps/s	2.8090 KOps/s	$\color{#d91a1a}-2.26\%$
test_keys	27.3210μs	3.4383μs	290.8438 KOps/s	290.3650 KOps/s	$\color{#35bf28}+0.16\%$
test_keys_nested	0.1095ms	80.9174μs	12.3583 KOps/s	12.2747 KOps/s	$\color{#35bf28}+0.68\%$
test_keys_nested_locked	0.7334ms	86.6380μs	11.5423 KOps/s	11.4507 KOps/s	$\color{#35bf28}+0.80\%$
test_keys_nested_leaf	0.1131ms	71.6819μs	13.9505 KOps/s	13.8900 KOps/s	$\color{#35bf28}+0.44\%$
test_keys_stack_nested	0.1239ms	81.5864μs	12.2569 KOps/s	12.1315 KOps/s	$\color{#35bf28}+1.03\%$
test_keys_stack_nested_leaf	0.1227ms	73.1953μs	13.6621 KOps/s	13.4952 KOps/s	$\color{#35bf28}+1.24\%$
test_keys_stack_nested_locked	0.1485ms	86.9836μs	11.4964 KOps/s	11.2804 KOps/s	$\color{#35bf28}+1.91\%$
test_values	6.0417μs	0.8493μs	1.1774 MOps/s	1.1815 MOps/s	$\color{#d91a1a}-0.35\%$
test_values_nested	67.7810μs	34.5725μs	28.9247 KOps/s	29.1175 KOps/s	$\color{#d91a1a}-0.66\%$
test_values_nested_locked	64.4110μs	35.6190μs	28.0749 KOps/s	27.5802 KOps/s	$\color{#35bf28}+1.79\%$
test_values_nested_leaf	80.1110μs	39.2807μs	25.4578 KOps/s	25.7134 KOps/s	$\color{#d91a1a}-0.99\%$
test_values_stack_nested	82.3010μs	34.8922μs	28.6597 KOps/s	28.6792 KOps/s	$\color{#d91a1a}-0.07\%$
test_values_stack_nested_leaf	80.2720μs	39.6768μs	25.2036 KOps/s	25.2472 KOps/s	$\color{#d91a1a}-0.17\%$
test_values_stack_nested_locked	85.1120μs	36.2985μs	27.5493 KOps/s	27.4595 KOps/s	$\color{#35bf28}+0.33\%$
test_membership	2.5381μs	0.5091μs	1.9641 MOps/s	1.9502 MOps/s	$\color{#35bf28}+0.71\%$
test_membership_nested	27.5410μs	2.0736μs	482.2471 KOps/s	496.4883 KOps/s	$\color{#d91a1a}-2.87\%$
test_membership_nested_leaf	19.6355μs	1.9605μs	510.0618 KOps/s	491.1853 KOps/s	$\color{#35bf28}+3.84\%$
test_membership_stacked_nested	29.8910μs	2.0670μs	483.7925 KOps/s	466.5242 KOps/s	$\color{#35bf28}+3.70\%$
test_membership_stacked_nested_leaf	27.0200μs	2.0462μs	488.7039 KOps/s	478.7313 KOps/s	$\color{#35bf28}+2.08\%$
test_membership_nested_last	31.7400μs	3.0447μs	328.4439 KOps/s	317.9668 KOps/s	$\color{#35bf28}+3.30\%$
test_membership_nested_leaf_last	32.8300μs	3.0958μs	323.0196 KOps/s	315.7166 KOps/s	$\color{#35bf28}+2.31\%$
test_membership_stacked_nested_last	22.3300μs	3.6266μs	275.7423 KOps/s	274.1322 KOps/s	$\color{#35bf28}+0.59\%$
test_membership_stacked_nested_leaf_last	31.7210μs	3.6066μs	277.2694 KOps/s	271.3986 KOps/s	$\color{#35bf28}+2.16\%$
test_nested_getleaf	37.2710μs	6.1585μs	162.3778 KOps/s	163.4539 KOps/s	$\color{#d91a1a}-0.66\%$
test_nested_get	52.5610μs	5.6868μs	175.8450 KOps/s	170.8357 KOps/s	$\color{#35bf28}+2.93\%$
test_stacked_getleaf	88.8720μs	6.1240μs	163.2919 KOps/s	164.0851 KOps/s	$\color{#d91a1a}-0.48\%$
test_stacked_get	32.1810μs	5.8771μs	170.1507 KOps/s	171.6579 KOps/s	$\color{#d91a1a}-0.88\%$
test_nested_getitemleaf	28.7500μs	6.2077μs	161.0910 KOps/s	157.4166 KOps/s	$\color{#35bf28}+2.33\%$
test_nested_getitem	26.5310μs	6.0075μs	166.4584 KOps/s	164.8710 KOps/s	$\color{#35bf28}+0.96\%$
test_stacked_getitemleaf	36.8710μs	6.2210μs	160.7448 KOps/s	159.4680 KOps/s	$\color{#35bf28}+0.80\%$
test_stacked_getitem	39.6000μs	5.9352μs	168.4876 KOps/s	164.9751 KOps/s	$\color{#35bf28}+2.13\%$
test_lock_nested	0.9056ms	0.3746ms	2.6698 KOps/s	2.5553 KOps/s	$\color{#35bf28}+4.48\%$
test_lock_stack_nested	0.3811ms	0.3452ms	2.8968 KOps/s	2.8492 KOps/s	$\color{#35bf28}+1.67\%$
test_unlock_nested	0.7712ms	0.3159ms	3.1656 KOps/s	3.0832 KOps/s	$\color{#35bf28}+2.67\%$
test_unlock_stack_nested	0.3274ms	0.2825ms	3.5393 KOps/s	3.4419 KOps/s	$\color{#35bf28}+2.83\%$
test_flatten_speed	0.1112ms	75.5551μs	13.2354 KOps/s	13.3455 KOps/s	$\color{#d91a1a}-0.82\%$
test_unflatten_speed	0.3828ms	0.3201ms	3.1241 KOps/s	3.0904 KOps/s	$\color{#35bf28}+1.09\%$
test_common_ops	1.5448ms	0.5900ms	1.6948 KOps/s	1.5169 KOps/s	$\textbf{\color{#35bf28}+11.73\%}$
test_creation	0.1125ms	1.7429μs	573.7458 KOps/s	573.1776 KOps/s	$\color{#35bf28}+0.10\%$
test_creation_empty	32.3810μs	6.9719μs	143.4329 KOps/s	97.9392 KOps/s	$\textbf{\color{#35bf28}+46.45\%}$
test_creation_nested_1	40.4410μs	8.5656μs	116.7463 KOps/s	84.1334 KOps/s	$\textbf{\color{#35bf28}+38.76\%}$
test_creation_nested_2	45.9310μs	11.3412μs	88.1742 KOps/s	68.8724 KOps/s	$\textbf{\color{#35bf28}+28.03\%}$
test_clone	98.6320μs	10.7496μs	93.0266 KOps/s	88.5847 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_getitem[int]	1.3809ms	10.8698μs	91.9984 KOps/s	88.3334 KOps/s	$\color{#35bf28}+4.15\%$
test_getitem[slice_int]	0.1163ms	21.1742μs	47.2273 KOps/s	45.8847 KOps/s	$\color{#35bf28}+2.93\%$
test_getitem[range]	0.1662ms	37.5908μs	26.6023 KOps/s	25.8476 KOps/s	$\color{#35bf28}+2.92\%$
test_getitem[tuple]	0.1083ms	18.2409μs	54.8218 KOps/s	53.3317 KOps/s	$\color{#35bf28}+2.79\%$
test_getitem[list]	0.3338ms	34.1131μs	29.3142 KOps/s	28.7900 KOps/s	$\color{#35bf28}+1.82\%$
test_setitem_dim[int]	41.9510μs	19.3234μs	51.7508 KOps/s	49.7707 KOps/s	$\color{#35bf28}+3.98\%$
test_setitem_dim[slice_int]	63.0010μs	39.3914μs	25.3862 KOps/s	26.3200 KOps/s	$\color{#d91a1a}-3.55\%$
test_setitem_dim[range]	80.7420μs	53.9271μs	18.5435 KOps/s	18.6337 KOps/s	$\color{#d91a1a}-0.48\%$
test_setitem_dim[tuple]	55.7620μs	32.8155μs	30.4734 KOps/s	30.3260 KOps/s	$\color{#35bf28}+0.49\%$
test_setitem	0.1120ms	14.3498μs	69.6875 KOps/s	59.5905 KOps/s	$\textbf{\color{#35bf28}+16.94\%}$
test_set	0.1052ms	13.9320μs	71.7771 KOps/s	60.2722 KOps/s	$\textbf{\color{#35bf28}+19.09\%}$
test_set_shared	1.6281ms	0.1537ms	6.5061 KOps/s	6.5235 KOps/s	$\color{#d91a1a}-0.27\%$
test_update	0.3200ms	16.3776μs	61.0592 KOps/s	50.0020 KOps/s	$\textbf{\color{#35bf28}+22.11\%}$
test_update_nested	0.1041ms	21.4289μs	46.6659 KOps/s	37.9598 KOps/s	$\textbf{\color{#35bf28}+22.94\%}$
test_update__nested	0.6004ms	25.7824μs	38.7861 KOps/s	37.7256 KOps/s	$\color{#35bf28}+2.81\%$
test_set_nested	0.1052ms	15.2590μs	65.5352 KOps/s	56.7069 KOps/s	$\textbf{\color{#35bf28}+15.57\%}$
test_set_nested_new	0.1094ms	17.3387μs	57.6746 KOps/s	50.0871 KOps/s	$\textbf{\color{#35bf28}+15.15\%}$
test_select	0.1145ms	29.4270μs	33.9824 KOps/s	31.6969 KOps/s	$\textbf{\color{#35bf28}+7.21\%}$
test_select_nested	0.1428ms	43.1794μs	23.1592 KOps/s	22.9248 KOps/s	$\color{#35bf28}+1.02\%$
test_exclude_nested	91.4520μs	62.8634μs	15.9075 KOps/s	15.9699 KOps/s	$\color{#d91a1a}-0.39\%$
test_empty[True]	0.3667ms	0.2895ms	3.4545 KOps/s	3.4600 KOps/s	$\color{#d91a1a}-0.16\%$
test_empty[False]	3.3531μs	0.8330μs	1.2005 MOps/s	1.1985 MOps/s	$\color{#35bf28}+0.17\%$
test_to	88.4820μs	56.7011μs	17.6363 KOps/s	17.4906 KOps/s	$\color{#35bf28}+0.83\%$
test_to_nonblocking	97.3020μs	48.3255μs	20.6930 KOps/s	20.5311 KOps/s	$\color{#35bf28}+0.79\%$
test_unbind_speed	0.2638ms	0.2346ms	4.2634 KOps/s	4.1603 KOps/s	$\color{#35bf28}+2.48\%$
test_unbind_speed_stack0	0.3433ms	0.2356ms	4.2438 KOps/s	4.1368 KOps/s	$\color{#35bf28}+2.59\%$
test_unbind_speed_stack1	92.8019ms	0.6620ms	1.5106 KOps/s	1.4951 KOps/s	$\color{#35bf28}+1.04\%$
test_split	93.4185ms	1.6050ms	623.0350 Ops/s	620.4442 Ops/s	$\color{#35bf28}+0.42\%$
test_chunk	93.5184ms	1.6020ms	624.2021 Ops/s	615.4158 Ops/s	$\color{#35bf28}+1.43\%$
test_consolidate[False-None]	96.2327ms	2.9254ms	341.8379 Ops/s	337.4372 Ops/s	$\color{#35bf28}+1.30\%$
test_consolidate[default-None]	1.8709ms	1.7050ms	586.4959 Ops/s	575.1674 Ops/s	$\color{#35bf28}+1.97\%$
test_consolidate[reduce-overhead-None]	1.8751ms	1.6962ms	589.5517 Ops/s	565.7526 Ops/s	$\color{#35bf28}+4.21\%$
test_consolidate_njt[False-None]	6.7364ms	6.3987ms	156.2816 Ops/s	151.7089 Ops/s	$\color{#35bf28}+3.01\%$
test_to[False-False-None]	2.1711ms	1.7739ms	563.7423 Ops/s	568.2313 Ops/s	$\color{#d91a1a}-0.79\%$
test_to[True-False-None]	1.7241ms	1.3233ms	755.6777 Ops/s	740.3055 Ops/s	$\color{#35bf28}+2.08\%$
test_to[within-False-None]	4.4493ms	4.0606ms	246.2703 Ops/s	240.6818 Ops/s	$\color{#35bf28}+2.32\%$
test_to[True-default-None]	5.6240ms	5.2266ms	191.3295 Ops/s	186.0363 Ops/s	$\color{#35bf28}+2.85\%$
test_to_njt[False-False-None]	7.3805ms	6.9233ms	144.4397 Ops/s	144.6801 Ops/s	$\color{#d91a1a}-0.17\%$
test_to_njt[True-False-None]	5.7382ms	5.4026ms	185.0977 Ops/s	183.1095 Ops/s	$\color{#35bf28}+1.09\%$
test_to_njt[within-False-None]	12.4462ms	12.0345ms	83.0944 Ops/s	83.8068 Ops/s	$\color{#d91a1a}-0.85\%$
test_creation[device0]	0.6362ms	81.5402μs	12.2639 KOps/s	12.2686 KOps/s	$\color{#d91a1a}-0.04\%$
test_creation_from_tensor	0.5068ms	84.6081μs	11.8192 KOps/s	11.7170 KOps/s	$\color{#35bf28}+0.87\%$
test_add_one[memmap_tensor0]	0.3273ms	6.9208μs	144.4910 KOps/s	141.3887 KOps/s	$\color{#35bf28}+2.19\%$
test_contiguous[memmap_tensor0]	2.4336μs	0.4211μs	2.3749 MOps/s	2.3851 MOps/s	$\color{#d91a1a}-0.43\%$
test_stack[memmap_tensor0]	23.0000μs	4.3580μs	229.4615 KOps/s	215.2638 KOps/s	$\textbf{\color{#35bf28}+6.60\%}$
test_memmaptd_index	2.0759ms	0.2540ms	3.9370 KOps/s	3.8290 KOps/s	$\color{#35bf28}+2.82\%$
test_memmaptd_index_astensor	0.8913ms	0.3170ms	3.1543 KOps/s	3.1137 KOps/s	$\color{#35bf28}+1.31\%$
test_memmaptd_index_op	1.0150ms	0.5731ms	1.7448 KOps/s	1.5716 KOps/s	$\textbf{\color{#35bf28}+11.02\%}$
test_serialize_model	0.1313s	0.1304s	7.6711 Ops/s	7.6455 Ops/s	$\color{#35bf28}+0.34\%$
test_serialize_model_pickle	1.3504s	1.2139s	0.8238 Ops/s	0.8247 Ops/s	$\color{#d91a1a}-0.10\%$
test_serialize_weights	0.1304s	0.1294s	7.7271 Ops/s	7.6886 Ops/s	$\color{#35bf28}+0.50\%$
test_serialize_weights_returnearly	0.3369s	63.3585ms	15.7832 Ops/s	14.1798 Ops/s	$\textbf{\color{#35bf28}+11.31\%}$
test_serialize_weights_pickle	1.3775s	1.2182s	0.8209 Ops/s	0.8220 Ops/s	$\color{#d91a1a}-0.13\%$
test_reshape_pytree	72.1420μs	21.9548μs	45.5480 KOps/s	43.4350 KOps/s	$\color{#35bf28}+4.86\%$
test_reshape_td	83.7610μs	26.4380μs	37.8244 KOps/s	34.7809 KOps/s	$\textbf{\color{#35bf28}+8.75\%}$
test_view_pytree	86.9020μs	21.3936μs	46.7429 KOps/s	42.8736 KOps/s	$\textbf{\color{#35bf28}+9.02\%}$
test_view_td	65.4210μs	29.6624μs	33.7127 KOps/s	29.2103 KOps/s	$\textbf{\color{#35bf28}+15.41\%}$
test_unbind_pytree	59.6010μs	28.0015μs	35.7124 KOps/s	33.7267 KOps/s	$\textbf{\color{#35bf28}+5.89\%}$
test_unbind_td	0.7209ms	35.7980μs	27.9346 KOps/s	25.6658 KOps/s	$\textbf{\color{#35bf28}+8.84\%}$
test_split_pytree	61.1410μs	29.0936μs	34.3718 KOps/s	31.0306 KOps/s	$\textbf{\color{#35bf28}+10.77\%}$
test_split_td	0.9203ms	37.7995μs	26.4553 KOps/s	24.1782 KOps/s	$\textbf{\color{#35bf28}+9.42\%}$
test_add_pytree	77.6420μs	34.4303μs	29.0442 KOps/s	27.3777 KOps/s	$\textbf{\color{#35bf28}+6.09\%}$
test_add_td	98.9720μs	47.1307μs	21.2176 KOps/s	17.5366 KOps/s	$\textbf{\color{#35bf28}+20.99\%}$
test_compile_add_one_nested[tensordict-compile]	0.1827ms	0.1179ms	8.4834 KOps/s	8.0563 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_compile_add_one_nested[tensordict-eager]	0.2632ms	0.1281ms	7.8053 KOps/s	7.5159 KOps/s	$\color{#35bf28}+3.85\%$
test_compile_add_one_nested[pytree-compile]	0.1462ms	94.3539μs	10.5984 KOps/s	10.0341 KOps/s	$\textbf{\color{#35bf28}+5.62\%}$
test_compile_add_one_nested[pytree-eager]	0.3258ms	0.1507ms	6.6335 KOps/s	6.4517 KOps/s	$\color{#35bf28}+2.82\%$
test_compile_copy_nested[tensordict-compile]	93.3120μs	22.2595μs	44.9246 KOps/s	42.5976 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_compile_copy_nested[tensordict-eager]	68.6910μs	29.3211μs	34.1052 KOps/s	33.4911 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_copy_nested[pytree-compile]	0.4743ms	64.7040μs	15.4550 KOps/s	15.2251 KOps/s	$\color{#35bf28}+1.51\%$
test_compile_copy_nested[pytree-eager]	90.8820μs	49.1614μs	20.3412 KOps/s	20.1244 KOps/s	$\color{#35bf28}+1.08\%$
test_compile_add_one_flat[tensordict-compile]	0.1893ms	0.1382ms	7.2374 KOps/s	7.0009 KOps/s	$\color{#35bf28}+3.38\%$
test_compile_add_one_flat[tensordict-eager]	0.3311ms	0.2158ms	4.6339 KOps/s	4.6333 KOps/s	$\color{#35bf28}+0.01\%$
test_compile_add_one_flat[tensorclass-compile]	0.1492ms	95.5356μs	10.4673 KOps/s	10.1196 KOps/s	$\color{#35bf28}+3.44\%$
test_compile_add_one_flat[tensorclass-eager]	0.1676ms	54.2442μs	18.4351 KOps/s	18.2232 KOps/s	$\color{#35bf28}+1.16\%$
test_compile_add_one_flat[pytree-compile]	0.1714ms	0.1314ms	7.6122 KOps/s	7.3157 KOps/s	$\color{#35bf28}+4.05\%$
test_compile_add_one_flat[pytree-eager]	0.5398ms	0.4963ms	2.0148 KOps/s	1.8766 KOps/s	$\textbf{\color{#35bf28}+7.36\%}$
test_compile_add_self_flat[tensordict-eager]	0.3731ms	0.2568ms	3.8945 KOps/s	3.8445 KOps/s	$\color{#35bf28}+1.30\%$
test_compile_add_self_flat[tensordict-compile]	0.1798ms	0.1388ms	7.2062 KOps/s	7.0580 KOps/s	$\color{#35bf28}+2.10\%$
test_compile_add_self_flat[tensorclass-eager]	0.1680ms	64.4012μs	15.5277 KOps/s	14.6257 KOps/s	$\textbf{\color{#35bf28}+6.17\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1366ms	96.6066μs	10.3513 KOps/s	10.2451 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_add_self_flat[pytree-eager]	0.4722ms	0.4164ms	2.4016 KOps/s	2.2472 KOps/s	$\textbf{\color{#35bf28}+6.87\%}$
test_compile_add_self_flat[pytree-compile]	0.1701ms	0.1327ms	7.5360 KOps/s	7.1332 KOps/s	$\textbf{\color{#35bf28}+5.65\%}$
test_compile_copy_flat[tensordict-compile]	0.1310ms	18.1222μs	55.1808 KOps/s	56.7885 KOps/s	$\color{#d91a1a}-2.83\%$
test_compile_copy_flat[tensordict-eager]	64.1920μs	31.7606μs	31.4856 KOps/s	32.0638 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_copy_flat[pytree-compile]	0.1094ms	69.8479μs	14.3168 KOps/s	14.1754 KOps/s	$\color{#35bf28}+1.00\%$
test_compile_copy_flat[pytree-eager]	80.3320μs	51.7027μs	19.3413 KOps/s	19.4081 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_assign_and_add[tensordict-compile]	1.6595ms	0.3976ms	2.5154 KOps/s	2.2264 KOps/s	$\textbf{\color{#35bf28}+12.98\%}$
test_compile_assign_and_add[tensordict-eager]	2.9115ms	2.6535ms	376.8631 Ops/s	359.9594 Ops/s	$\color{#35bf28}+4.70\%$
test_compile_assign_and_add[pytree-compile]	1.6040ms	0.4333ms	2.3077 KOps/s	2.2004 KOps/s	$\color{#35bf28}+4.88\%$
test_compile_assign_and_add[pytree-eager]	2.7937ms	2.6971ms	370.7701 Ops/s	356.0138 Ops/s	$\color{#35bf28}+4.14\%$
test_compile_indexing[tensor-tensordict-compile]	0.6223ms	0.1156ms	8.6484 KOps/s	8.3225 KOps/s	$\color{#35bf28}+3.92\%$
test_compile_indexing[tensor-tensordict-eager]	0.5764ms	81.4616μs	12.2757 KOps/s	11.7429 KOps/s	$\color{#35bf28}+4.54\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5217ms	0.1119ms	8.9346 KOps/s	9.0681 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2003ms	70.5298μs	14.1784 KOps/s	13.8163 KOps/s	$\color{#35bf28}+2.62\%$
test_compile_indexing[tensor-pytree-compile]	0.1652ms	0.1158ms	8.6365 KOps/s	8.8227 KOps/s	$\color{#d91a1a}-2.11\%$
test_compile_indexing[tensor-pytree-eager]	0.1135ms	69.6358μs	14.3604 KOps/s	13.5475 KOps/s	$\textbf{\color{#35bf28}+6.00\%}$
test_compile_indexing[slice-tensordict-compile]	0.1594ms	0.1053ms	9.4991 KOps/s	9.8420 KOps/s	$\color{#d91a1a}-3.48\%$
test_compile_indexing[slice-tensordict-eager]	0.1432ms	16.8935μs	59.1944 KOps/s	54.6191 KOps/s	$\textbf{\color{#35bf28}+8.38\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1714ms	0.1011ms	9.8911 KOps/s	10.1704 KOps/s	$\color{#d91a1a}-2.75\%$
test_compile_indexing[slice-tensorclass-eager]	66.3810μs	15.4616μs	64.6765 KOps/s	60.3301 KOps/s	$\textbf{\color{#35bf28}+7.20\%}$
test_compile_indexing[slice-pytree-compile]	0.1521ms	0.1013ms	9.8675 KOps/s	10.1058 KOps/s	$\color{#d91a1a}-2.36\%$
test_compile_indexing[slice-pytree-eager]	79.7520μs	15.5561μs	64.2836 KOps/s	60.7713 KOps/s	$\textbf{\color{#35bf28}+5.78\%}$
test_compile_indexing[int-tensordict-compile]	0.1550ms	0.1069ms	9.3529 KOps/s	9.7444 KOps/s	$\color{#d91a1a}-4.02\%$
test_compile_indexing[int-tensordict-eager]	0.5754ms	16.7895μs	59.5610 KOps/s	55.8740 KOps/s	$\textbf{\color{#35bf28}+6.60\%}$
test_compile_indexing[int-tensorclass-compile]	0.1647ms	97.2282μs	10.2851 KOps/s	10.1195 KOps/s	$\color{#35bf28}+1.64\%$
test_compile_indexing[int-tensorclass-eager]	0.1163ms	15.3653μs	65.0819 KOps/s	61.4494 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_compile_indexing[int-pytree-compile]	0.2238ms	0.1014ms	9.8619 KOps/s	10.1071 KOps/s	$\color{#d91a1a}-2.43\%$
test_compile_indexing[int-pytree-eager]	0.4109ms	15.6861μs	63.7506 KOps/s	61.1754 KOps/s	$\color{#35bf28}+4.21\%$
test_mod_add[eager]	83.9910μs	37.0493μs	26.9911 KOps/s	25.3344 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_mod_add[compile]	0.2242ms	78.1661μs	12.7933 KOps/s	12.3967 KOps/s	$\color{#35bf28}+3.20\%$
test_mod_add[compile-overhead]	0.3241ms	0.1658ms	6.0331 KOps/s	5.6811 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_mod_wrap[eager]	0.3971ms	0.2529ms	3.9534 KOps/s	3.6574 KOps/s	$\textbf{\color{#35bf28}+8.09\%}$
test_mod_wrap[compile]	1.1139ms	0.2839ms	3.5221 KOps/s	3.4423 KOps/s	$\color{#35bf28}+2.32\%$
test_mod_wrap[compile-overhead]	6.5189ms	3.6029ms	277.5564 Ops/s	272.9629 Ops/s	$\color{#35bf28}+1.68\%$
test_mod_wrap_and_backward[eager]	1.6530ms	1.3804ms	724.4530 Ops/s	671.6130 Ops/s	$\textbf{\color{#35bf28}+7.87\%}$
test_mod_wrap_and_backward[compile]	1.3820ms	1.2757ms	783.8707 Ops/s	713.1238 Ops/s	$\textbf{\color{#35bf28}+9.92\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3724ms	0.9285ms	1.0770 KOps/s	963.1642 Ops/s	$\textbf{\color{#35bf28}+11.82\%}$
test_seq_add[eager]	0.2070ms	0.1127ms	8.8698 KOps/s	8.0184 KOps/s	$\textbf{\color{#35bf28}+10.62\%}$
test_seq_add[compile]	0.1471ms	90.2877μs	11.0757 KOps/s	10.5502 KOps/s	$\color{#35bf28}+4.98\%$
test_seq_add[compile-overhead]	0.1730ms	0.1287ms	7.7685 KOps/s	7.3468 KOps/s	$\textbf{\color{#35bf28}+5.74\%}$
test_seq_wrap[eager]	0.4913ms	0.4177ms	2.3939 KOps/s	2.1773 KOps/s	$\textbf{\color{#35bf28}+9.95\%}$
test_seq_wrap[compile]	0.3707ms	0.2995ms	3.3385 KOps/s	3.2447 KOps/s	$\color{#35bf28}+2.89\%$
test_seq_wrap[compile-overhead]	0.2709ms	0.2226ms	4.4916 KOps/s	4.3729 KOps/s	$\color{#35bf28}+2.71\%$
test_func_call_runtime[False-eager]	0.8818ms	0.7991ms	1.2515 KOps/s	1.3166 KOps/s	$\color{#d91a1a}-4.95\%$
test_func_call_runtime[False-compile]	0.8570ms	0.7512ms	1.3313 KOps/s	1.2662 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_func_call_runtime[False-compile-overhead]	0.4131ms	0.3622ms	2.7608 KOps/s	2.7226 KOps/s	$\color{#35bf28}+1.41\%$
test_func_call_runtime[True-eager]	0.9862ms	0.9164ms	1.0913 KOps/s	1.0060 KOps/s	$\textbf{\color{#35bf28}+8.47\%}$
test_func_call_runtime[True-compile]	0.8698ms	0.7765ms	1.2878 KOps/s	1.2714 KOps/s	$\color{#35bf28}+1.29\%$
test_func_call_runtime[True-compile-overhead]	0.4420ms	0.3859ms	2.5911 KOps/s	2.5843 KOps/s	$\color{#35bf28}+0.26\%$
test_func_call_cm_runtime[False-eager]	0.8417ms	0.7775ms	1.2862 KOps/s	1.2753 KOps/s	$\color{#35bf28}+0.86\%$
test_func_call_cm_runtime[False-compile]	0.8289ms	0.7483ms	1.3364 KOps/s	1.3070 KOps/s	$\color{#35bf28}+2.25\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4114ms	0.3654ms	2.7369 KOps/s	2.6994 KOps/s	$\color{#35bf28}+1.39\%$
test_func_call_cm_runtime[True-eager]	1.1112ms	1.0153ms	984.9118 Ops/s	978.6642 Ops/s	$\color{#35bf28}+0.64\%$
test_func_call_cm_runtime[True-compile]	0.8816ms	0.7969ms	1.2549 KOps/s	1.2429 KOps/s	$\color{#35bf28}+0.96\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4637ms	0.4112ms	2.4320 KOps/s	2.4019 KOps/s	$\color{#35bf28}+1.25\%$
test_vmap_func_call_cm_runtime[eager]	2.6004ms	2.1072ms	474.5730 Ops/s	469.3354 Ops/s	$\color{#35bf28}+1.12\%$
test_vmap_func_call_cm_runtime[compile]	0.9176ms	0.8238ms	1.2139 KOps/s	1.2058 KOps/s	$\color{#35bf28}+0.67\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4872ms	0.4136ms	2.4180 KOps/s	2.3990 KOps/s	$\color{#35bf28}+0.79\%$
test_distributed	2.8214ms	0.3278ms	3.0510 KOps/s	7.7865 KOps/s	$\textbf{\color{#d91a1a}-60.82\%}$
test_tdmodule	0.2273ms	19.4313μs	51.4633 KOps/s	47.9775 KOps/s	$\textbf{\color{#35bf28}+7.27\%}$
test_tdmodule_dispatch	68.8110μs	34.8563μs	28.6892 KOps/s	26.6671 KOps/s	$\textbf{\color{#35bf28}+7.58\%}$
test_tdseq	31.0110μs	19.7285μs	50.6881 KOps/s	45.7608 KOps/s	$\textbf{\color{#35bf28}+10.77\%}$
test_tdseq_dispatch	58.9820μs	36.1273μs	27.6799 KOps/s	24.2722 KOps/s	$\textbf{\color{#35bf28}+14.04\%}$
test_instantiation_functorch	1.6729ms	1.5358ms	651.1392 Ops/s	633.8500 Ops/s	$\color{#35bf28}+2.73\%$
test_exec_functorch	0.2086ms	0.1464ms	6.8299 KOps/s	6.3499 KOps/s	$\textbf{\color{#35bf28}+7.56\%}$
test_exec_functional_call	0.1867ms	0.1401ms	7.1364 KOps/s	6.7198 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_exec_td_decorator	0.3824ms	0.1880ms	5.3201 KOps/s	5.2827 KOps/s	$\color{#35bf28}+0.71\%$
test_vmap_mlp_speed_decorator[True-True]	0.7773ms	0.6906ms	1.4480 KOps/s	1.4093 KOps/s	$\color{#35bf28}+2.75\%$
test_vmap_mlp_speed_decorator[True-False]	0.8486ms	0.6890ms	1.4514 KOps/s	1.4026 KOps/s	$\color{#35bf28}+3.48\%$
test_vmap_mlp_speed_decorator[False-True]	0.7217ms	0.6006ms	1.6649 KOps/s	1.5910 KOps/s	$\color{#35bf28}+4.64\%$
test_vmap_mlp_speed_decorator[False-False]	0.7192ms	0.6022ms	1.6605 KOps/s	1.6483 KOps/s	$\color{#35bf28}+0.74\%$
test_vmap_transformer_speed_decorator[True-True]	19.6092ms	19.4771ms	51.3424 Ops/s	51.3120 Ops/s	$\color{#35bf28}+0.06\%$
test_vmap_transformer_speed_decorator[True-False]	20.2600ms	19.6257ms	50.9535 Ops/s	51.4737 Ops/s	$\color{#d91a1a}-1.01\%$
test_vmap_transformer_speed_decorator[False-True]	19.5435ms	19.4431ms	51.4322 Ops/s	51.8253 Ops/s	$\color{#d91a1a}-0.76\%$
test_vmap_transformer_speed_decorator[False-False]	19.4893ms	19.4349ms	51.4537 Ops/s	51.6833 Ops/s	$\color{#d91a1a}-0.44\%$
test_to_module_speed[True]	1.1115ms	0.9711ms	1.0298 KOps/s	1.0305 KOps/s	$\color{#d91a1a}-0.07\%$
test_to_module_speed[False]	1.0727ms	0.9435ms	1.0599 KOps/s	1.0448 KOps/s	$\color{#35bf28}+1.45\%$
test_tc_init	68.2320μs	34.6385μs	28.8696 KOps/s	27.0585 KOps/s	$\textbf{\color{#35bf28}+6.69\%}$
test_tc_init_nested	0.1141ms	70.7381μs	14.1367 KOps/s	13.3625 KOps/s	$\textbf{\color{#35bf28}+5.79\%}$
test_tc_first_layer_tensor	29.7000μs	0.8187μs	1.2214 MOps/s	1.2051 MOps/s	$\color{#35bf28}+1.35\%$
test_tc_first_layer_nontensor	28.4010μs	2.2884μs	436.9899 KOps/s	437.4432 KOps/s	$\color{#d91a1a}-0.10\%$
test_tc_second_layer_tensor	8.2325μs	1.4146μs	706.8946 KOps/s	689.7571 KOps/s	$\color{#35bf28}+2.48\%$
test_tc_second_layer_nontensor	25.8500μs	3.0245μs	330.6293 KOps/s	325.6972 KOps/s	$\color{#35bf28}+1.51\%$
test_unbind	7.2682ms	6.9924ms	143.0129 Ops/s	144.6816 Ops/s	$\color{#d91a1a}-1.15\%$
test_full_like	9.2198ms	9.1117ms	109.7485 Ops/s	108.5052 Ops/s	$\color{#35bf28}+1.15\%$
test_zeros_like	4.8379ms	4.3228ms	231.3333 Ops/s	230.6655 Ops/s	$\color{#35bf28}+0.29\%$
test_ones_like	5.3750ms	4.3241ms	231.2608 Ops/s	231.1621 Ops/s	$\color{#35bf28}+0.04\%$
test_clone	6.4720ms	6.3586ms	157.2664 Ops/s	156.4724 Ops/s	$\color{#35bf28}+0.51\%$
test_squeeze	82.1320μs	9.7365μs	102.7068 KOps/s	106.1526 KOps/s	$\color{#d91a1a}-3.25\%$
test_unsqueeze	0.1234ms	73.1868μs	13.6637 KOps/s	14.0405 KOps/s	$\color{#d91a1a}-2.68\%$
test_split	0.3311ms	0.1552ms	6.4449 KOps/s	6.2482 KOps/s	$\color{#35bf28}+3.15\%$
test_permute	0.3264ms	0.1796ms	5.5681 KOps/s	5.4937 KOps/s	$\color{#35bf28}+1.35\%$
test_stack	50.8467ms	50.4588ms	19.8182 Ops/s	20.0407 Ops/s	$\color{#d91a1a}-1.11\%$
test_cat	50.7124ms	50.3999ms	19.8413 Ops/s	19.9783 Ops/s	$\color{#d91a1a}-0.69\%$