[BugFix] Consistent behavior for pad_sequence with one and many non-tensors

[vmoens]

Stack from ghstack (oldest at bottom):

• (to be filled)

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	38.1210μs	20.9626μs	47.7039 KOps/s	49.6185 KOps/s	$\color{#d91a1a}-3.86\%$
test_plain_set_stack_nested	46.3470μs	21.1829μs	47.2079 KOps/s	49.1317 KOps/s	$\color{#d91a1a}-3.92\%$
test_plain_set_nested_inplace	69.1500μs	22.6466μs	44.1568 KOps/s	45.1603 KOps/s	$\color{#d91a1a}-2.22\%$
test_plain_set_stack_nested_inplace	58.5200μs	22.7535μs	43.9492 KOps/s	45.3280 KOps/s	$\color{#d91a1a}-3.04\%$
test_items	51.5360μs	4.1376μs	241.6852 KOps/s	232.9455 KOps/s	$\color{#35bf28}+3.75\%$
test_items_nested	0.6914ms	0.3909ms	2.5583 KOps/s	2.4665 KOps/s	$\color{#35bf28}+3.72\%$
test_items_nested_locked	0.6095ms	0.3952ms	2.5301 KOps/s	2.4235 KOps/s	$\color{#35bf28}+4.40\%$
test_items_nested_leaf	0.1506ms	76.4775μs	13.0757 KOps/s	12.9043 KOps/s	$\color{#35bf28}+1.33\%$
test_items_stack_nested	0.6270ms	0.3960ms	2.5254 KOps/s	2.4502 KOps/s	$\color{#35bf28}+3.07\%$
test_items_stack_nested_leaf	0.1484ms	79.5140μs	12.5764 KOps/s	12.9864 KOps/s	$\color{#d91a1a}-3.16\%$
test_items_stack_nested_locked	0.5950ms	0.4006ms	2.4963 KOps/s	2.4467 KOps/s	$\color{#35bf28}+2.03\%$
test_keys	26.8900μs	3.4277μs	291.7426 KOps/s	278.6066 KOps/s	$\color{#35bf28}+4.71\%$
test_keys_nested	0.3249ms	0.1608ms	6.2193 KOps/s	6.0527 KOps/s	$\color{#35bf28}+2.75\%$
test_keys_nested_locked	0.6596ms	0.1676ms	5.9681 KOps/s	5.8760 KOps/s	$\color{#35bf28}+1.57\%$
test_keys_nested_leaf	0.2361ms	0.1410ms	7.0935 KOps/s	6.9838 KOps/s	$\color{#35bf28}+1.57\%$
test_keys_stack_nested	0.2732ms	0.1616ms	6.1883 KOps/s	6.0606 KOps/s	$\color{#35bf28}+2.11\%$
test_keys_stack_nested_leaf	0.2508ms	0.1393ms	7.1806 KOps/s	6.9654 KOps/s	$\color{#35bf28}+3.09\%$
test_keys_stack_nested_locked	0.2604ms	0.1665ms	6.0058 KOps/s	5.8097 KOps/s	$\color{#35bf28}+3.38\%$
test_values	9.7622μs	1.0294μs	971.4218 KOps/s	970.7565 KOps/s	$\color{#35bf28}+0.07\%$
test_values_nested	0.1070ms	60.9026μs	16.4197 KOps/s	16.1191 KOps/s	$\color{#35bf28}+1.86\%$
test_values_nested_locked	0.1070ms	61.1612μs	16.3502 KOps/s	16.2527 KOps/s	$\color{#35bf28}+0.60\%$
test_values_nested_leaf	0.1312ms	70.7451μs	14.1353 KOps/s	14.0230 KOps/s	$\color{#35bf28}+0.80\%$
test_values_stack_nested	0.1187ms	61.9603μs	16.1394 KOps/s	15.4264 KOps/s	$\color{#35bf28}+4.62\%$
test_values_stack_nested_leaf	0.1319ms	70.9113μs	14.1021 KOps/s	13.6991 KOps/s	$\color{#35bf28}+2.94\%$
test_values_stack_nested_locked	0.1158ms	61.7060μs	16.2059 KOps/s	16.0582 KOps/s	$\color{#35bf28}+0.92\%$
test_membership	3.7956μs	0.6984μs	1.4319 MOps/s	1.1668 MOps/s	$\textbf{\color{#35bf28}+22.72\%}$
test_membership_nested	46.3360μs	2.9014μs	344.6585 KOps/s	348.4337 KOps/s	$\color{#d91a1a}-1.08\%$
test_membership_nested_leaf	20.8890μs	2.8712μs	348.2841 KOps/s	348.0872 KOps/s	$\color{#35bf28}+0.06\%$
test_membership_stacked_nested	44.1630μs	2.8494μs	350.9486 KOps/s	347.4753 KOps/s	$\color{#35bf28}+1.00\%$
test_membership_stacked_nested_leaf	22.3020μs	2.8775μs	347.5248 KOps/s	350.9461 KOps/s	$\color{#d91a1a}-0.97\%$
test_membership_nested_last	49.8330μs	4.2679μs	234.3067 KOps/s	232.4561 KOps/s	$\color{#35bf28}+0.80\%$
test_membership_nested_leaf_last	31.2580μs	4.2864μs	233.2974 KOps/s	228.9295 KOps/s	$\color{#35bf28}+1.91\%$
test_membership_stacked_nested_last	50.8660μs	5.0418μs	198.3426 KOps/s	222.8236 KOps/s	$\textbf{\color{#d91a1a}-10.99\%}$
test_membership_stacked_nested_leaf_last	34.2440μs	5.0815μs	196.7929 KOps/s	227.6950 KOps/s	$\textbf{\color{#d91a1a}-13.57\%}$
test_nested_getleaf	0.1242ms	10.9803μs	91.0723 KOps/s	96.0736 KOps/s	$\textbf{\color{#d91a1a}-5.21\%}$
test_nested_get	60.3520μs	10.0852μs	99.1547 KOps/s	101.1642 KOps/s	$\color{#d91a1a}-1.99\%$
test_stacked_getleaf	34.6350μs	10.5677μs	94.6281 KOps/s	94.7768 KOps/s	$\color{#d91a1a}-0.16\%$
test_stacked_get	54.0810μs	10.1090μs	98.9220 KOps/s	98.7545 KOps/s	$\color{#35bf28}+0.17\%$
test_nested_getitemleaf	35.6170μs	10.7956μs	92.6301 KOps/s	90.5357 KOps/s	$\color{#35bf28}+2.31\%$
test_nested_getitem	56.3350μs	10.2692μs	97.3786 KOps/s	96.9663 KOps/s	$\color{#35bf28}+0.43\%$
test_stacked_getitemleaf	53.1690μs	10.9746μs	91.1194 KOps/s	90.7705 KOps/s	$\color{#35bf28}+0.38\%$
test_stacked_getitem	32.2900μs	10.2988μs	97.0989 KOps/s	97.7674 KOps/s	$\color{#d91a1a}-0.68\%$
test_lock_nested	1.2354ms	0.4526ms	2.2095 KOps/s	2.1171 KOps/s	$\color{#35bf28}+4.36\%$
test_lock_stack_nested	0.7839ms	0.4223ms	2.3680 KOps/s	2.2985 KOps/s	$\color{#35bf28}+3.03\%$
test_unlock_nested	0.8633ms	0.3731ms	2.6802 KOps/s	2.6390 KOps/s	$\color{#35bf28}+1.56\%$
test_unlock_stack_nested	0.7340ms	0.3465ms	2.8864 KOps/s	2.8504 KOps/s	$\color{#35bf28}+1.26\%$
test_flatten_speed	0.1730ms	99.3388μs	10.0666 KOps/s	10.1349 KOps/s	$\color{#d91a1a}-0.67\%$
test_unflatten_speed	0.7012ms	0.5167ms	1.9354 KOps/s	1.9386 KOps/s	$\color{#d91a1a}-0.16\%$
test_common_ops	5.4275ms	0.8008ms	1.2488 KOps/s	1.2488 KOps/s	$-0.00\%$
test_creation	41.3560μs	2.5118μs	398.1226 KOps/s	395.3748 KOps/s	$\color{#35bf28}+0.69\%$
test_creation_empty	40.1150μs	12.6219μs	79.2274 KOps/s	88.7580 KOps/s	$\textbf{\color{#d91a1a}-10.74\%}$
test_creation_nested_1	54.2210μs	15.7142μs	63.6368 KOps/s	70.3609 KOps/s	$\textbf{\color{#d91a1a}-9.56\%}$
test_creation_nested_2	0.2792ms	20.1907μs	49.5279 KOps/s	52.3869 KOps/s	$\textbf{\color{#d91a1a}-5.46\%}$
test_clone	1.2899ms	14.4663μs	69.1264 KOps/s	71.3747 KOps/s	$\color{#d91a1a}-3.15\%$
test_getitem[int]	0.9598ms	13.6630μs	73.1902 KOps/s	73.7698 KOps/s	$\color{#d91a1a}-0.79\%$
test_getitem[slice_int]	0.1420ms	24.5921μs	40.6635 KOps/s	39.0788 KOps/s	$\color{#35bf28}+4.06\%$
test_getitem[range]	0.2761ms	46.8321μs	21.3529 KOps/s	20.3885 KOps/s	$\color{#35bf28}+4.73\%$
test_getitem[tuple]	0.1312ms	20.3199μs	49.2129 KOps/s	46.5848 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_getitem[list]	0.2236ms	42.2101μs	23.6910 KOps/s	22.6408 KOps/s	$\color{#35bf28}+4.64\%$
test_setitem_dim[int]	51.3560μs	24.4585μs	40.8856 KOps/s	38.7561 KOps/s	$\textbf{\color{#35bf28}+5.49\%}$
test_setitem_dim[slice_int]	80.9720μs	51.3845μs	19.4611 KOps/s	19.2957 KOps/s	$\color{#35bf28}+0.86\%$
test_setitem_dim[range]	0.1101ms	72.1208μs	13.8656 KOps/s	13.3970 KOps/s	$\color{#35bf28}+3.50\%$
test_setitem_dim[tuple]	79.4190μs	39.9518μs	25.0301 KOps/s	24.0364 KOps/s	$\color{#35bf28}+4.13\%$
test_setitem	0.1747ms	20.7834μs	48.1152 KOps/s	48.2307 KOps/s	$\color{#d91a1a}-0.24\%$
test_set	0.1850ms	20.2282μs	49.4358 KOps/s	48.8117 KOps/s	$\color{#35bf28}+1.28\%$
test_set_shared	1.1886ms	0.1685ms	5.9331 KOps/s	5.9131 KOps/s	$\color{#35bf28}+0.34\%$
test_update	0.1498ms	23.6924μs	42.2076 KOps/s	43.9032 KOps/s	$\color{#d91a1a}-3.86\%$
test_update_nested	0.1609ms	33.6206μs	29.7436 KOps/s	30.1002 KOps/s	$\color{#d91a1a}-1.18\%$
test_update__nested	0.6867ms	34.1840μs	29.2535 KOps/s	28.8861 KOps/s	$\color{#35bf28}+1.27\%$
test_set_nested	0.1328ms	22.4229μs	44.5973 KOps/s	44.5371 KOps/s	$\color{#35bf28}+0.14\%$
test_set_nested_new	0.1062ms	26.6547μs	37.5169 KOps/s	36.9381 KOps/s	$\color{#35bf28}+1.57\%$
test_select	0.1520ms	42.9062μs	23.3067 KOps/s	22.9428 KOps/s	$\color{#35bf28}+1.59\%$
test_select_nested	0.1306ms	63.2525μs	15.8096 KOps/s	15.0854 KOps/s	$\color{#35bf28}+4.80\%$
test_exclude_nested	0.1505ms	80.9443μs	12.3542 KOps/s	11.7563 KOps/s	$\textbf{\color{#35bf28}+5.09\%}$
test_empty[True]	0.7162ms	0.4055ms	2.4663 KOps/s	2.4087 KOps/s	$\color{#35bf28}+2.39\%$
test_empty[False]	12.6213μs	1.3522μs	739.5148 KOps/s	726.2793 KOps/s	$\color{#35bf28}+1.82\%$
test_unbind_speed	0.3850ms	0.2685ms	3.7250 KOps/s	3.6282 KOps/s	$\color{#35bf28}+2.67\%$
test_unbind_speed_stack0	0.4875ms	0.2665ms	3.7520 KOps/s	3.6518 KOps/s	$\color{#35bf28}+2.75\%$
test_unbind_speed_stack1	0.1064s	0.8595ms	1.1635 KOps/s	1.3331 KOps/s	$\textbf{\color{#d91a1a}-12.72\%}$
test_split	0.1023s	1.7634ms	567.0805 Ops/s	542.5082 Ops/s	$\color{#35bf28}+4.53\%$
test_chunk	0.1061s	1.7663ms	566.1393 Ops/s	544.2492 Ops/s	$\color{#35bf28}+4.02\%$
test_consolidate_njt[False-None]	9.8956ms	8.2779ms	120.8032 Ops/s	119.6245 Ops/s	$\color{#35bf28}+0.99\%$
test_creation[device0]	0.2701ms	93.6013μs	10.6836 KOps/s	10.4891 KOps/s	$\color{#35bf28}+1.85\%$
test_creation_from_tensor	0.2940ms	95.4305μs	10.4788 KOps/s	10.5205 KOps/s	$\color{#d91a1a}-0.40\%$
test_add_one[memmap_tensor0]	0.4192ms	5.1211μs	195.2715 KOps/s	197.5410 KOps/s	$\color{#d91a1a}-1.15\%$
test_contiguous[memmap_tensor0]	13.1450μs	0.5198μs	1.9239 MOps/s	1.9242 MOps/s	$\color{#d91a1a}-0.02\%$
test_stack[memmap_tensor0]	62.8470μs	3.4867μs	286.8052 KOps/s	286.1439 KOps/s	$\color{#35bf28}+0.23\%$
test_memmaptd_index	1.0082ms	0.2379ms	4.2032 KOps/s	4.1380 KOps/s	$\color{#35bf28}+1.58\%$
test_memmaptd_index_astensor	0.6573ms	0.3234ms	3.0918 KOps/s	3.0446 KOps/s	$\color{#35bf28}+1.55\%$
test_memmaptd_index_op	0.9892ms	0.6082ms	1.6442 KOps/s	1.6512 KOps/s	$\color{#d91a1a}-0.42\%$
test_serialize_model	0.1302s	0.1179s	8.4807 Ops/s	7.4058 Ops/s	$\textbf{\color{#35bf28}+14.51\%}$
test_serialize_model_pickle	0.4979s	0.4020s	2.4875 Ops/s	2.5631 Ops/s	$\color{#d91a1a}-2.95\%$
test_serialize_weights	0.2189s	0.1327s	7.5383 Ops/s	8.6046 Ops/s	$\textbf{\color{#d91a1a}-12.39\%}$
test_serialize_weights_returnearly	0.1703s	0.1590s	6.2891 Ops/s	6.2587 Ops/s	$\color{#35bf28}+0.49\%$
test_serialize_weights_pickle	0.4881s	0.4102s	2.4380 Ops/s	2.5015 Ops/s	$\color{#d91a1a}-2.54\%$
test_serialize_weights_filesystem	0.1519s	0.1430s	6.9921 Ops/s	7.0211 Ops/s	$\color{#d91a1a}-0.41\%$
test_serialize_model_filesystem	0.1562s	0.1498s	6.6774 Ops/s	5.7911 Ops/s	$\textbf{\color{#35bf28}+15.30\%}$
test_reshape_pytree	62.5470μs	26.3698μs	37.9221 KOps/s	37.4889 KOps/s	$\color{#35bf28}+1.16\%$
test_reshape_td	0.1071ms	32.6802μs	30.5996 KOps/s	28.9401 KOps/s	$\textbf{\color{#35bf28}+5.73\%}$
test_view_pytree	73.5070μs	26.4200μs	37.8502 KOps/s	37.6216 KOps/s	$\color{#35bf28}+0.61\%$
test_view_td	83.5670μs	37.5105μs	26.6592 KOps/s	26.2406 KOps/s	$\color{#35bf28}+1.60\%$
test_unbind_pytree	67.7560μs	29.1524μs	34.3025 KOps/s	33.6545 KOps/s	$\color{#35bf28}+1.93\%$
test_unbind_td	0.3349ms	39.5778μs	25.2667 KOps/s	24.8281 KOps/s	$\color{#35bf28}+1.77\%$
test_split_pytree	75.3510μs	29.1968μs	34.2503 KOps/s	33.7738 KOps/s	$\color{#35bf28}+1.41\%$
test_split_td	0.5197ms	45.0554μs	22.1949 KOps/s	21.6823 KOps/s	$\color{#35bf28}+2.36\%$
test_add_pytree	0.1019ms	36.4590μs	27.4281 KOps/s	28.6725 KOps/s	$\color{#d91a1a}-4.34\%$
test_add_td	0.1367ms	57.2378μs	17.4710 KOps/s	18.0895 KOps/s	$\color{#d91a1a}-3.42\%$
test_compile_add_one_nested[tensordict-compile]	0.1252ms	62.0938μs	16.1047 KOps/s	15.8505 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_add_one_nested[tensordict-eager]	1.8301ms	0.1723ms	5.8036 KOps/s	5.7764 KOps/s	$\color{#35bf28}+0.47\%$
test_compile_add_one_nested[pytree-compile]	0.1060ms	45.4784μs	21.9885 KOps/s	21.9439 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_add_one_nested[pytree-eager]	0.2214ms	0.1181ms	8.4670 KOps/s	8.4829 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_copy_nested[tensordict-compile]	64.1200μs	25.8670μs	38.6593 KOps/s	38.6988 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_copy_nested[tensordict-eager]	0.2733ms	59.7092μs	16.7478 KOps/s	16.4021 KOps/s	$\color{#35bf28}+2.11\%$
test_compile_copy_nested[pytree-compile]	0.1430ms	76.4693μs	13.0771 KOps/s	12.8354 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_copy_nested[pytree-eager]	0.1249ms	65.6846μs	15.2243 KOps/s	14.9650 KOps/s	$\color{#35bf28}+1.73\%$
test_compile_add_one_flat[tensordict-compile]	0.1820ms	0.1037ms	9.6440 KOps/s	9.3082 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_add_one_flat[tensordict-eager]	0.3213ms	0.2126ms	4.7044 KOps/s	4.5683 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_add_one_flat[tensorclass-compile]	93.9670μs	43.4677μs	23.0056 KOps/s	22.3404 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_add_one_flat[tensorclass-eager]	0.4707ms	67.4835μs	14.8184 KOps/s	15.0815 KOps/s	$\color{#d91a1a}-1.74\%$
test_compile_add_one_flat[pytree-compile]	0.2075ms	0.1024ms	9.7617 KOps/s	9.7542 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_add_one_flat[pytree-eager]	0.3542ms	0.2023ms	4.9440 KOps/s	4.9608 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_add_self_flat[tensordict-eager]	0.4298ms	0.2317ms	4.3155 KOps/s	4.1735 KOps/s	$\color{#35bf28}+3.40\%$
test_compile_add_self_flat[tensordict-compile]	0.1764ms	0.1054ms	9.4860 KOps/s	9.3232 KOps/s	$\color{#35bf28}+1.75\%$
test_compile_add_self_flat[tensorclass-eager]	1.4083ms	66.1190μs	15.1243 KOps/s	15.4538 KOps/s	$\color{#d91a1a}-2.13\%$
test_compile_add_self_flat[tensorclass-compile]	0.1196ms	48.4594μs	20.6358 KOps/s	21.2980 KOps/s	$\color{#d91a1a}-3.11\%$
test_compile_add_self_flat[pytree-eager]	0.2921ms	0.1555ms	6.4292 KOps/s	6.2536 KOps/s	$\color{#35bf28}+2.81\%$
test_compile_add_self_flat[pytree-compile]	0.1774ms	0.1028ms	9.7256 KOps/s	9.2273 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_compile_copy_flat[tensordict-compile]	53.4100μs	21.1697μs	47.2373 KOps/s	47.2357 KOps/s	$+0.00\%$
test_compile_copy_flat[tensordict-eager]	0.1333ms	65.9384μs	15.1657 KOps/s	15.0277 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_copy_flat[pytree-compile]	0.1505ms	79.2426μs	12.6195 KOps/s	12.6831 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_copy_flat[pytree-eager]	0.1418ms	66.8308μs	14.9632 KOps/s	14.7606 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_assign_and_add[tensordict-compile]	0.2911ms	0.2140ms	4.6735 KOps/s	4.5143 KOps/s	$\color{#35bf28}+3.53\%$
test_compile_assign_and_add[tensordict-eager]	1.4649ms	1.2995ms	769.5146 Ops/s	749.4317 Ops/s	$\color{#35bf28}+2.68\%$
test_compile_assign_and_add[pytree-compile]	0.3058ms	0.2078ms	4.8127 KOps/s	4.8068 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_assign_and_add[pytree-eager]	1.3237ms	0.7796ms	1.2826 KOps/s	1.2718 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_assign_and_add_stack[compile]	0.5409ms	0.4601ms	2.1737 KOps/s	2.1511 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_assign_and_add_stack[eager]	3.0332ms	2.7732ms	360.5891 Ops/s	369.9984 Ops/s	$\color{#d91a1a}-2.54\%$
test_compile_indexing[tensor-tensordict-compile]	74.8000μs	34.7438μs	28.7821 KOps/s	27.5701 KOps/s	$\color{#35bf28}+4.40\%$
test_compile_indexing[tensor-tensordict-eager]	0.4983ms	33.9330μs	29.4698 KOps/s	29.2800 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_indexing[tensor-tensorclass-compile]	83.5560μs	28.3233μs	35.3066 KOps/s	34.6495 KOps/s	$\color{#35bf28}+1.90\%$
test_compile_indexing[tensor-tensorclass-eager]	61.7150μs	22.8357μs	43.7911 KOps/s	43.1429 KOps/s	$\color{#35bf28}+1.50\%$
test_compile_indexing[tensor-pytree-compile]	78.8580μs	29.4467μs	33.9597 KOps/s	32.9110 KOps/s	$\color{#35bf28}+3.19\%$
test_compile_indexing[tensor-pytree-eager]	0.2019ms	24.1509μs	41.4064 KOps/s	43.0426 KOps/s	$\color{#d91a1a}-3.80\%$
test_compile_indexing[slice-tensordict-compile]	0.1163ms	50.4325μs	19.8285 KOps/s	19.3190 KOps/s	$\color{#35bf28}+2.64\%$
test_compile_indexing[slice-tensordict-eager]	0.5996ms	20.3988μs	49.0224 KOps/s	46.9348 KOps/s	$\color{#35bf28}+4.45\%$
test_compile_indexing[slice-tensorclass-compile]	0.1084ms	44.6403μs	22.4013 KOps/s	21.3896 KOps/s	$\color{#35bf28}+4.73\%$
test_compile_indexing[slice-tensorclass-eager]	0.1740ms	18.8936μs	52.9280 KOps/s	53.6511 KOps/s	$\color{#d91a1a}-1.35\%$
test_compile_indexing[slice-pytree-compile]	0.1157ms	45.0036μs	22.2204 KOps/s	21.9510 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_indexing[slice-pytree-eager]	53.9810μs	18.5522μs	53.9020 KOps/s	53.8193 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_indexing[int-tensordict-compile]	0.1328ms	51.4466μs	19.4376 KOps/s	18.4643 KOps/s	$\textbf{\color{#35bf28}+5.27\%}$
test_compile_indexing[int-tensordict-eager]	0.8641ms	20.7974μs	48.0829 KOps/s	46.5476 KOps/s	$\color{#35bf28}+3.30\%$
test_compile_indexing[int-tensorclass-compile]	90.3390μs	44.7902μs	22.3263 KOps/s	22.1513 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_indexing[int-tensorclass-eager]	0.1756ms	18.4896μs	54.0846 KOps/s	53.9203 KOps/s	$\color{#35bf28}+0.30\%$
test_compile_indexing[int-pytree-compile]	0.2002ms	44.5008μs	22.4715 KOps/s	22.0169 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_indexing[int-pytree-eager]	0.4251ms	18.8690μs	52.9971 KOps/s	54.0017 KOps/s	$\color{#d91a1a}-1.86\%$
test_mod_add[eager]	80.5500μs	34.5282μs	28.9618 KOps/s	28.9210 KOps/s	$\color{#35bf28}+0.14\%$
test_mod_add[compile]	0.1221ms	47.7362μs	20.9485 KOps/s	20.6425 KOps/s	$\color{#35bf28}+1.48\%$
test_mod_add[compile-overhead]	0.1246ms	47.9515μs	20.8544 KOps/s	20.3296 KOps/s	$\color{#35bf28}+2.58\%$
test_mod_wrap[eager]	0.4089ms	0.2223ms	4.4993 KOps/s	4.5092 KOps/s	$\color{#d91a1a}-0.22\%$
test_mod_wrap[compile]	0.3840ms	0.2083ms	4.8005 KOps/s	4.8006 KOps/s	$-0.00\%$
test_mod_wrap[compile-overhead]	0.4259ms	0.2077ms	4.8141 KOps/s	4.7928 KOps/s	$\color{#35bf28}+0.44\%$
test_mod_wrap_and_backward[eager]	14.7113ms	11.2022ms	89.2678 Ops/s	81.2337 Ops/s	$\textbf{\color{#35bf28}+9.89\%}$
test_mod_wrap_and_backward[compile]	12.2978ms	10.8323ms	92.3168 Ops/s	68.7919 Ops/s	$\textbf{\color{#35bf28}+34.20\%}$
test_mod_wrap_and_backward[compile-overhead]	12.3400ms	10.8205ms	92.4172 Ops/s	70.7060 Ops/s	$\textbf{\color{#35bf28}+30.71\%}$
test_seq_add[eager]	0.2060ms	0.1163ms	8.6019 KOps/s	8.2677 KOps/s	$\color{#35bf28}+4.04\%$
test_seq_add[compile]	0.1262ms	62.7326μs	15.9407 KOps/s	15.9881 KOps/s	$\color{#d91a1a}-0.30\%$
test_seq_add[compile-overhead]	0.1122ms	59.8835μs	16.6991 KOps/s	16.1562 KOps/s	$\color{#35bf28}+3.36\%$
test_seq_wrap[eager]	0.7247ms	0.4464ms	2.2401 KOps/s	2.2465 KOps/s	$\color{#d91a1a}-0.29\%$
test_seq_wrap[compile]	0.3688ms	0.2276ms	4.3938 KOps/s	4.3237 KOps/s	$\color{#35bf28}+1.62\%$
test_seq_wrap[compile-overhead]	0.3046ms	0.2255ms	4.4338 KOps/s	4.3547 KOps/s	$\color{#35bf28}+1.82\%$
test_func_call_runtime[False-eager]	0.6802ms	0.5415ms	1.8469 KOps/s	1.8264 KOps/s	$\color{#35bf28}+1.12\%$
test_func_call_runtime[False-compile]	0.5307ms	0.4236ms	2.3608 KOps/s	2.3220 KOps/s	$\color{#35bf28}+1.67\%$
test_func_call_runtime[False-compile-overhead]	0.5189ms	0.4248ms	2.3540 KOps/s	2.3260 KOps/s	$\color{#35bf28}+1.20\%$
test_func_call_runtime[True-eager]	0.8587ms	0.7551ms	1.3244 KOps/s	1.2852 KOps/s	$\color{#35bf28}+3.05\%$
test_func_call_runtime[True-compile]	0.5537ms	0.4657ms	2.1472 KOps/s	2.1055 KOps/s	$\color{#35bf28}+1.98\%$
test_func_call_runtime[True-compile-overhead]	0.5874ms	0.4672ms	2.1404 KOps/s	2.1043 KOps/s	$\color{#35bf28}+1.72\%$
test_func_call_cm_runtime[False-eager]	1.1383ms	0.5368ms	1.8629 KOps/s	1.8090 KOps/s	$\color{#35bf28}+2.98\%$
test_func_call_cm_runtime[False-compile]	0.5072ms	0.4266ms	2.3439 KOps/s	2.2957 KOps/s	$\color{#35bf28}+2.10\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5386ms	0.4255ms	2.3503 KOps/s	2.3178 KOps/s	$\color{#35bf28}+1.40\%$
test_func_call_cm_runtime[True-eager]	1.1389ms	0.8873ms	1.1271 KOps/s	1.0901 KOps/s	$\color{#35bf28}+3.39\%$
test_func_call_cm_runtime[True-compile]	0.5824ms	0.4911ms	2.0362 KOps/s	2.0220 KOps/s	$\color{#35bf28}+0.70\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9991ms	0.4979ms	2.0084 KOps/s	2.0221 KOps/s	$\color{#d91a1a}-0.68\%$
test_vmap_func_call_cm_runtime[eager]	2.3070ms	1.9022ms	525.6950 Ops/s	523.9364 Ops/s	$\color{#35bf28}+0.34\%$
test_vmap_func_call_cm_runtime[compile]	0.7353ms	0.5150ms	1.9416 KOps/s	1.8786 KOps/s	$\color{#35bf28}+3.35\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9028ms	0.5182ms	1.9298 KOps/s	1.8994 KOps/s	$\color{#35bf28}+1.60\%$
test_distributed	0.3025ms	0.1262ms	7.9212 KOps/s	7.6372 KOps/s	$\color{#35bf28}+3.72\%$
test_tdmodule	56.8860μs	27.3318μs	36.5875 KOps/s	37.6989 KOps/s	$\color{#d91a1a}-2.95\%$
test_tdmodule_dispatch	91.0400μs	51.0362μs	19.5939 KOps/s	20.7248 KOps/s	$\textbf{\color{#d91a1a}-5.46\%}$
test_tdseq	63.0180μs	30.7365μs	32.5346 KOps/s	33.4901 KOps/s	$\color{#d91a1a}-2.85\%$
test_tdseq_dispatch	88.4150μs	56.3849μs	17.7353 KOps/s	18.0402 KOps/s	$\color{#d91a1a}-1.69\%$
test_instantiation_functorch	2.3985ms	1.5346ms	651.6254 Ops/s	641.4194 Ops/s	$\color{#35bf28}+1.59\%$
test_exec_functorch	0.3591ms	0.1856ms	5.3883 KOps/s	5.3751 KOps/s	$\color{#35bf28}+0.25\%$
test_exec_functional_call	0.3323ms	0.1768ms	5.6547 KOps/s	5.6993 KOps/s	$\color{#d91a1a}-0.78\%$
test_exec_td_decorator	0.4445ms	0.2361ms	4.2351 KOps/s	4.2647 KOps/s	$\color{#d91a1a}-0.69\%$
test_vmap_mlp_speed_decorator[True-True]	1.0108ms	0.6572ms	1.5216 KOps/s	1.4870 KOps/s	$\color{#35bf28}+2.33\%$
test_vmap_mlp_speed_decorator[True-False]	1.1954ms	0.6591ms	1.5173 KOps/s	1.5047 KOps/s	$\color{#35bf28}+0.83\%$
test_vmap_mlp_speed_decorator[False-True]	0.7404ms	0.5276ms	1.8952 KOps/s	1.8774 KOps/s	$\color{#35bf28}+0.95\%$
test_vmap_mlp_speed_decorator[False-False]	0.8694ms	0.5285ms	1.8921 KOps/s	1.8726 KOps/s	$\color{#35bf28}+1.04\%$
test_to_module_speed[True]	1.4384ms	1.3343ms	749.4343 Ops/s	735.0307 Ops/s	$\color{#35bf28}+1.96\%$
test_to_module_speed[False]	1.7425ms	1.3136ms	761.2451 Ops/s	760.7872 Ops/s	$\color{#35bf28}+0.06\%$
test_tc_init	83.7960μs	47.0731μs	21.2435 KOps/s	21.6219 KOps/s	$\color{#d91a1a}-1.75\%$
test_tc_init_nested	0.2112ms	95.1990μs	10.5043 KOps/s	11.0654 KOps/s	$\textbf{\color{#d91a1a}-5.07\%}$
test_tc_first_layer_tensor	23.7950μs	1.5274μs	654.6976 KOps/s	660.9351 KOps/s	$\color{#d91a1a}-0.94\%$
test_tc_first_layer_nontensor	26.2590μs	4.6713μs	214.0726 KOps/s	217.2525 KOps/s	$\color{#d91a1a}-1.46\%$
test_tc_second_layer_tensor	23.5840μs	2.8337μs	352.8913 KOps/s	356.7474 KOps/s	$\color{#d91a1a}-1.08\%$
test_tc_second_layer_nontensor	38.2020μs	5.9486μs	168.1076 KOps/s	168.4291 KOps/s	$\color{#d91a1a}-0.19\%$
test_unbind	0.2218s	13.2458ms	75.4958 Ops/s	75.2368 Ops/s	$\color{#35bf28}+0.34\%$
test_full_like	18.6816ms	12.3507ms	80.9671 Ops/s	71.4832 Ops/s	$\textbf{\color{#35bf28}+13.27\%}$
test_zeros_like	11.7511ms	7.3481ms	136.0896 Ops/s	138.8744 Ops/s	$\color{#d91a1a}-2.01\%$
test_ones_like	12.8930ms	7.3971ms	135.1878 Ops/s	121.1793 Ops/s	$\textbf{\color{#35bf28}+11.56\%}$
test_clone	16.0395ms	8.9839ms	111.3107 Ops/s	102.8680 Ops/s	$\textbf{\color{#35bf28}+8.21\%}$
test_squeeze	60.4030μs	12.0696μs	82.8525 KOps/s	83.5817 KOps/s	$\color{#d91a1a}-0.87\%$
test_unsqueeze	0.1555ms	89.7169μs	11.1462 KOps/s	10.9156 KOps/s	$\color{#35bf28}+2.11\%$
test_split	0.4330ms	0.1989ms	5.0276 KOps/s	4.9864 KOps/s	$\color{#35bf28}+0.83\%$
test_permute	0.3073ms	0.2022ms	4.9450 KOps/s	4.7341 KOps/s	$\color{#35bf28}+4.46\%$
test_stack	28.0803ms	24.1750ms	41.3650 Ops/s	40.1075 Ops/s	$\color{#35bf28}+3.14\%$
test_cat	28.7554ms	24.0413ms	41.5951 Ops/s	39.9726 Ops/s	$\color{#35bf28}+4.06\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}58$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	31.3900μs	11.4480μs	87.3518 KOps/s	76.6250 KOps/s	$\textbf{\color{#35bf28}+14.00\%}$
test_plain_set_stack_nested	35.8710μs	11.6249μs	86.0223 KOps/s	74.9795 KOps/s	$\textbf{\color{#35bf28}+14.73\%}$
test_plain_set_nested_inplace	44.1510μs	12.4988μs	80.0079 KOps/s	70.1994 KOps/s	$\textbf{\color{#35bf28}+13.97\%}$
test_plain_set_stack_nested_inplace	40.5910μs	12.5076μs	79.9514 KOps/s	69.9936 KOps/s	$\textbf{\color{#35bf28}+14.23\%}$
test_items	22.6000μs	2.9202μs	342.4464 KOps/s	330.9792 KOps/s	$\color{#35bf28}+3.46\%$
test_items_nested	0.5239ms	0.3618ms	2.7640 KOps/s	2.7680 KOps/s	$\color{#d91a1a}-0.14\%$
test_items_nested_locked	0.4667ms	0.3622ms	2.7606 KOps/s	2.7692 KOps/s	$\color{#d91a1a}-0.31\%$
test_items_nested_leaf	95.3920μs	57.8832μs	17.2762 KOps/s	17.1733 KOps/s	$\color{#35bf28}+0.60\%$
test_items_stack_nested	0.5436ms	0.3658ms	2.7340 KOps/s	2.7656 KOps/s	$\color{#d91a1a}-1.14\%$
test_items_stack_nested_leaf	96.6230μs	58.6170μs	17.0599 KOps/s	16.6503 KOps/s	$\color{#35bf28}+2.46\%$
test_items_stack_nested_locked	0.4871ms	0.3630ms	2.7548 KOps/s	2.7767 KOps/s	$\color{#d91a1a}-0.79\%$
test_keys	21.9400μs	3.5067μs	285.1711 KOps/s	288.2709 KOps/s	$\color{#d91a1a}-1.08\%$
test_keys_nested	0.1287ms	81.5461μs	12.2630 KOps/s	12.2768 KOps/s	$\color{#d91a1a}-0.11\%$
test_keys_nested_locked	0.7010ms	87.8060μs	11.3887 KOps/s	11.3972 KOps/s	$\color{#d91a1a}-0.07\%$
test_keys_nested_leaf	0.1146ms	71.8580μs	13.9163 KOps/s	13.8276 KOps/s	$\color{#35bf28}+0.64\%$
test_keys_stack_nested	0.1325ms	81.7234μs	12.2364 KOps/s	12.1389 KOps/s	$\color{#35bf28}+0.80\%$
test_keys_stack_nested_leaf	0.1194ms	72.1048μs	13.8687 KOps/s	13.6278 KOps/s	$\color{#35bf28}+1.77\%$
test_keys_stack_nested_locked	0.1328ms	88.4840μs	11.3015 KOps/s	11.3654 KOps/s	$\color{#d91a1a}-0.56\%$
test_values	4.5652μs	0.8489μs	1.1779 MOps/s	1.1725 MOps/s	$\color{#35bf28}+0.46\%$
test_values_nested	62.7210μs	34.5636μs	28.9322 KOps/s	29.0836 KOps/s	$\color{#d91a1a}-0.52\%$
test_values_nested_locked	70.8910μs	36.5764μs	27.3400 KOps/s	27.6876 KOps/s	$\color{#d91a1a}-1.26\%$
test_values_nested_leaf	70.5020μs	39.2923μs	25.4503 KOps/s	25.7859 KOps/s	$\color{#d91a1a}-1.30\%$
test_values_stack_nested	66.8610μs	34.8911μs	28.6606 KOps/s	28.5132 KOps/s	$\color{#35bf28}+0.52\%$
test_values_stack_nested_leaf	72.5720μs	39.4978μs	25.3179 KOps/s	25.4751 KOps/s	$\color{#d91a1a}-0.62\%$
test_values_stack_nested_locked	72.0220μs	36.5880μs	27.3314 KOps/s	27.4299 KOps/s	$\color{#d91a1a}-0.36\%$
test_membership	1.9025μs	0.5103μs	1.9598 MOps/s	1.9842 MOps/s	$\color{#d91a1a}-1.23\%$
test_membership_nested	32.1900μs	2.1086μs	474.2515 KOps/s	501.7458 KOps/s	$\textbf{\color{#d91a1a}-5.48\%}$
test_membership_nested_leaf	28.5955μs	1.9895μs	502.6290 KOps/s	498.1447 KOps/s	$\color{#35bf28}+0.90\%$
test_membership_stacked_nested	24.3710μs	2.0921μs	477.9869 KOps/s	475.6924 KOps/s	$\color{#35bf28}+0.48\%$
test_membership_stacked_nested_leaf	75.6910μs	2.1064μs	474.7476 KOps/s	483.9848 KOps/s	$\color{#d91a1a}-1.91\%$
test_membership_nested_last	24.8010μs	3.1343μs	319.0457 KOps/s	321.6374 KOps/s	$\color{#d91a1a}-0.81\%$
test_membership_nested_leaf_last	28.7510μs	3.1632μs	316.1405 KOps/s	320.6498 KOps/s	$\color{#d91a1a}-1.41\%$
test_membership_stacked_nested_last	57.2420μs	3.1188μs	320.6319 KOps/s	158.0447 KOps/s	$\textbf{\color{#35bf28}+102.87\%}$
test_membership_stacked_nested_leaf_last	25.6210μs	3.1296μs	319.5316 KOps/s	160.0250 KOps/s	$\textbf{\color{#35bf28}+99.68\%}$
test_nested_getleaf	29.3500μs	6.1601μs	162.3343 KOps/s	163.4183 KOps/s	$\color{#d91a1a}-0.66\%$
test_nested_get	34.9810μs	5.8351μs	171.3752 KOps/s	171.3339 KOps/s	$\color{#35bf28}+0.02\%$
test_stacked_getleaf	42.5710μs	6.1127μs	163.5936 KOps/s	160.8713 KOps/s	$\color{#35bf28}+1.69\%$
test_stacked_get	25.4600μs	5.7928μs	172.6273 KOps/s	170.1763 KOps/s	$\color{#35bf28}+1.44\%$
test_nested_getitemleaf	32.8500μs	6.2629μs	159.6716 KOps/s	160.1816 KOps/s	$\color{#d91a1a}-0.32\%$
test_nested_getitem	34.0710μs	5.9969μs	166.7530 KOps/s	167.7704 KOps/s	$\color{#d91a1a}-0.61\%$
test_stacked_getitemleaf	39.5810μs	6.1966μs	161.3786 KOps/s	158.9716 KOps/s	$\color{#35bf28}+1.51\%$
test_stacked_getitem	29.0410μs	5.8765μs	170.1682 KOps/s	167.7206 KOps/s	$\color{#35bf28}+1.46\%$
test_lock_nested	0.7908ms	0.3750ms	2.6668 KOps/s	2.6411 KOps/s	$\color{#35bf28}+0.97\%$
test_lock_stack_nested	0.3827ms	0.3483ms	2.8708 KOps/s	2.8929 KOps/s	$\color{#d91a1a}-0.76\%$
test_unlock_nested	0.6449ms	0.3165ms	3.1591 KOps/s	3.1298 KOps/s	$\color{#35bf28}+0.94\%$
test_unlock_stack_nested	0.3266ms	0.2888ms	3.4630 KOps/s	3.5486 KOps/s	$\color{#d91a1a}-2.41\%$
test_flatten_speed	0.1222ms	74.7619μs	13.3758 KOps/s	13.3088 KOps/s	$\color{#35bf28}+0.50\%$
test_unflatten_speed	0.3752ms	0.3170ms	3.1541 KOps/s	3.1315 KOps/s	$\color{#35bf28}+0.72\%$
test_common_ops	92.4776ms	0.6665ms	1.5003 KOps/s	1.5160 KOps/s	$\color{#d91a1a}-1.04\%$
test_creation	0.1002ms	1.7516μs	570.9193 KOps/s	565.4705 KOps/s	$\color{#35bf28}+0.96\%$
test_creation_empty	30.3600μs	6.8892μs	145.1546 KOps/s	99.1371 KOps/s	$\textbf{\color{#35bf28}+46.42\%}$
test_creation_nested_1	41.3010μs	8.7535μs	114.2402 KOps/s	85.5429 KOps/s	$\textbf{\color{#35bf28}+33.55\%}$
test_creation_nested_2	36.8810μs	11.4081μs	87.6573 KOps/s	69.3032 KOps/s	$\textbf{\color{#35bf28}+26.48\%}$
test_clone	76.9610μs	10.6365μs	94.0161 KOps/s	87.7231 KOps/s	$\textbf{\color{#35bf28}+7.17\%}$
test_getitem[int]	1.4012ms	10.8785μs	91.9245 KOps/s	89.8118 KOps/s	$\color{#35bf28}+2.35\%$
test_getitem[slice_int]	0.1086ms	21.0408μs	47.5268 KOps/s	45.9644 KOps/s	$\color{#35bf28}+3.40\%$
test_getitem[range]	0.1301ms	37.2941μs	26.8139 KOps/s	24.5700 KOps/s	$\textbf{\color{#35bf28}+9.13\%}$
test_getitem[tuple]	0.1383ms	18.5646μs	53.8659 KOps/s	53.1919 KOps/s	$\color{#35bf28}+1.27\%$
test_getitem[list]	0.2012ms	33.2507μs	30.0745 KOps/s	27.9368 KOps/s	$\textbf{\color{#35bf28}+7.65\%}$
test_setitem_dim[int]	42.2210μs	19.0687μs	52.4420 KOps/s	45.5481 KOps/s	$\textbf{\color{#35bf28}+15.14\%}$
test_setitem_dim[slice_int]	61.2920μs	39.2502μs	25.4776 KOps/s	23.3447 KOps/s	$\textbf{\color{#35bf28}+9.14\%}$
test_setitem_dim[range]	78.9320μs	52.1139μs	19.1887 KOps/s	17.0697 KOps/s	$\textbf{\color{#35bf28}+12.41\%}$
test_setitem_dim[tuple]	57.3720μs	32.9032μs	30.3921 KOps/s	28.3474 KOps/s	$\textbf{\color{#35bf28}+7.21\%}$
test_setitem	85.3920μs	14.1868μs	70.4880 KOps/s	57.8771 KOps/s	$\textbf{\color{#35bf28}+21.79\%}$
test_set	90.7320μs	13.8429μs	72.2392 KOps/s	60.3832 KOps/s	$\textbf{\color{#35bf28}+19.63\%}$
test_set_shared	1.5748ms	0.1520ms	6.5804 KOps/s	6.5046 KOps/s	$\color{#35bf28}+1.16\%$
test_update	1.2908ms	15.8942μs	62.9159 KOps/s	49.7631 KOps/s	$\textbf{\color{#35bf28}+26.43\%}$
test_update_nested	84.3620μs	21.8360μs	45.7959 KOps/s	38.2554 KOps/s	$\textbf{\color{#35bf28}+19.71\%}$
test_update__nested	0.2456ms	25.8105μs	38.7440 KOps/s	36.1944 KOps/s	$\textbf{\color{#35bf28}+7.04\%}$
test_set_nested	75.3510μs	15.2726μs	65.4767 KOps/s	57.0232 KOps/s	$\textbf{\color{#35bf28}+14.82\%}$
test_set_nested_new	84.7320μs	17.5017μs	57.1373 KOps/s	50.3827 KOps/s	$\textbf{\color{#35bf28}+13.41\%}$
test_select	91.4220μs	29.3234μs	34.1024 KOps/s	31.3754 KOps/s	$\textbf{\color{#35bf28}+8.69\%}$
test_select_nested	82.0620μs	43.4445μs	23.0179 KOps/s	23.1606 KOps/s	$\color{#d91a1a}-0.62\%$
test_exclude_nested	0.1023ms	62.9055μs	15.8969 KOps/s	15.8277 KOps/s	$\color{#35bf28}+0.44\%$
test_empty[True]	0.4063ms	0.2875ms	3.4784 KOps/s	3.5017 KOps/s	$\color{#d91a1a}-0.67\%$
test_empty[False]	3.4930μs	0.8240μs	1.2136 MOps/s	1.1933 MOps/s	$\color{#35bf28}+1.70\%$
test_to	85.7520μs	55.2127μs	18.1118 KOps/s	17.2968 KOps/s	$\color{#35bf28}+4.71\%$
test_to_nonblocking	82.3720μs	46.9556μs	21.2967 KOps/s	20.3008 KOps/s	$\color{#35bf28}+4.91\%$
test_unbind_speed	0.2850ms	0.2388ms	4.1882 KOps/s	4.0897 KOps/s	$\color{#35bf28}+2.41\%$
test_unbind_speed_stack0	0.2875ms	0.2370ms	4.2186 KOps/s	4.2583 KOps/s	$\color{#d91a1a}-0.93\%$
test_unbind_speed_stack1	93.2363ms	0.6799ms	1.4707 KOps/s	1.5053 KOps/s	$\color{#d91a1a}-2.30\%$
test_split	95.0608ms	1.7392ms	574.9884 Ops/s	616.9553 Ops/s	$\textbf{\color{#d91a1a}-6.80\%}$
test_chunk	1.5968ms	1.4657ms	682.2523 Ops/s	610.6830 Ops/s	$\textbf{\color{#35bf28}+11.72\%}$
test_consolidate[False-None]	98.3596ms	2.9533ms	338.6009 Ops/s	335.1898 Ops/s	$\color{#35bf28}+1.02\%$
test_consolidate[default-None]	2.1412ms	1.7092ms	585.0741 Ops/s	575.9215 Ops/s	$\color{#35bf28}+1.59\%$
test_consolidate[reduce-overhead-None]	2.0304ms	1.7446ms	573.1820 Ops/s	547.6456 Ops/s	$\color{#35bf28}+4.66\%$
test_consolidate_njt[False-None]	6.8841ms	6.4858ms	154.1830 Ops/s	152.2031 Ops/s	$\color{#35bf28}+1.30\%$
test_to[False-False-None]	2.1240ms	1.6885ms	592.2415 Ops/s	568.9732 Ops/s	$\color{#35bf28}+4.09\%$
test_to[True-False-None]	1.7437ms	1.3101ms	763.2757 Ops/s	718.5773 Ops/s	$\textbf{\color{#35bf28}+6.22\%}$
test_to[within-False-None]	4.6021ms	4.1317ms	242.0301 Ops/s	237.7726 Ops/s	$\color{#35bf28}+1.79\%$
test_to[True-default-None]	5.7256ms	5.3635ms	186.4463 Ops/s	187.5636 Ops/s	$\color{#d91a1a}-0.60\%$
test_to_njt[False-False-None]	7.3928ms	6.9346ms	144.2034 Ops/s	144.5990 Ops/s	$\color{#d91a1a}-0.27\%$
test_to_njt[True-False-None]	5.9308ms	5.4247ms	184.3417 Ops/s	184.2853 Ops/s	$\color{#35bf28}+0.03\%$
test_to_njt[within-False-None]	12.4352ms	11.9963ms	83.3589 Ops/s	81.7319 Ops/s	$\color{#35bf28}+1.99\%$
test_creation[device0]	0.4672ms	81.1676μs	12.3202 KOps/s	12.2427 KOps/s	$\color{#35bf28}+0.63\%$
test_creation_from_tensor	0.6253ms	84.8030μs	11.7920 KOps/s	11.7872 KOps/s	$\color{#35bf28}+0.04\%$
test_add_one[memmap_tensor0]	0.3443ms	6.7151μs	148.9192 KOps/s	136.3914 KOps/s	$\textbf{\color{#35bf28}+9.19\%}$
test_contiguous[memmap_tensor0]	1.8726μs	0.4159μs	2.4041 MOps/s	2.3553 MOps/s	$\color{#35bf28}+2.07\%$
test_stack[memmap_tensor0]	35.7310μs	4.2875μs	233.2383 KOps/s	211.0355 KOps/s	$\textbf{\color{#35bf28}+10.52\%}$
test_memmaptd_index	1.6060ms	0.2475ms	4.0400 KOps/s	3.8540 KOps/s	$\color{#35bf28}+4.83\%$
test_memmaptd_index_astensor	0.5956ms	0.3114ms	3.2114 KOps/s	3.1257 KOps/s	$\color{#35bf28}+2.74\%$
test_memmaptd_index_op	1.0033ms	0.5567ms	1.7964 KOps/s	1.5684 KOps/s	$\textbf{\color{#35bf28}+14.54\%}$
test_serialize_model	0.1313s	0.1303s	7.6727 Ops/s	7.6745 Ops/s	$\color{#d91a1a}-0.02\%$
test_serialize_model_pickle	1.3838s	1.2177s	0.8212 Ops/s	0.8185 Ops/s	$\color{#35bf28}+0.34\%$
test_serialize_weights	0.1310s	0.1300s	7.6907 Ops/s	7.7321 Ops/s	$\color{#d91a1a}-0.54\%$
test_serialize_weights_returnearly	0.5303s	62.3566ms	16.0368 Ops/s	15.4888 Ops/s	$\color{#35bf28}+3.54\%$
test_serialize_weights_pickle	1.3756s	1.2162s	0.8222 Ops/s	0.8366 Ops/s	$\color{#d91a1a}-1.71\%$
test_reshape_pytree	57.6710μs	21.9603μs	45.5366 KOps/s	43.8910 KOps/s	$\color{#35bf28}+3.75\%$
test_reshape_td	58.0210μs	26.8388μs	37.2595 KOps/s	35.8008 KOps/s	$\color{#35bf28}+4.07\%$
test_view_pytree	48.3710μs	22.0654μs	45.3197 KOps/s	45.0801 KOps/s	$\color{#35bf28}+0.53\%$
test_view_td	73.9820μs	31.7624μs	31.4838 KOps/s	30.9242 KOps/s	$\color{#35bf28}+1.81\%$
test_unbind_pytree	72.6320μs	28.2096μs	35.4490 KOps/s	34.9000 KOps/s	$\color{#35bf28}+1.57\%$
test_unbind_td	0.7934ms	35.7683μs	27.9577 KOps/s	26.0414 KOps/s	$\textbf{\color{#35bf28}+7.36\%}$
test_split_pytree	76.7120μs	29.9968μs	33.3369 KOps/s	33.0689 KOps/s	$\color{#35bf28}+0.81\%$
test_split_td	0.9855ms	38.3575μs	26.0705 KOps/s	25.2763 KOps/s	$\color{#35bf28}+3.14\%$
test_add_pytree	90.6520μs	34.1544μs	29.2788 KOps/s	27.2104 KOps/s	$\textbf{\color{#35bf28}+7.60\%}$
test_add_td	94.1720μs	46.1802μs	21.6543 KOps/s	18.3055 KOps/s	$\textbf{\color{#35bf28}+18.29\%}$
test_compile_add_one_nested[tensordict-compile]	0.1850ms	0.1205ms	8.2994 KOps/s	8.0904 KOps/s	$\color{#35bf28}+2.58\%$
test_compile_add_one_nested[tensordict-eager]	0.2794ms	0.1333ms	7.5034 KOps/s	7.5114 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_add_one_nested[pytree-compile]	0.1345ms	95.5907μs	10.4613 KOps/s	10.2001 KOps/s	$\color{#35bf28}+2.56\%$
test_compile_add_one_nested[pytree-eager]	0.2922ms	0.1467ms	6.8171 KOps/s	6.5260 KOps/s	$\color{#35bf28}+4.46\%$
test_compile_copy_nested[tensordict-compile]	61.6810μs	22.5008μs	44.4428 KOps/s	43.0188 KOps/s	$\color{#35bf28}+3.31\%$
test_compile_copy_nested[tensordict-eager]	0.1107ms	29.2160μs	34.2278 KOps/s	33.9755 KOps/s	$\color{#35bf28}+0.74\%$
test_compile_copy_nested[pytree-compile]	0.4663ms	65.1111μs	15.3584 KOps/s	15.4868 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_copy_nested[pytree-eager]	81.4720μs	49.3401μs	20.2675 KOps/s	20.2266 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_add_one_flat[tensordict-compile]	0.2229ms	0.1484ms	6.7404 KOps/s	6.9338 KOps/s	$\color{#d91a1a}-2.79\%$
test_compile_add_one_flat[tensordict-eager]	0.3154ms	0.2178ms	4.5910 KOps/s	4.5790 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_one_flat[tensorclass-compile]	0.2232ms	0.1025ms	9.7602 KOps/s	10.1240 KOps/s	$\color{#d91a1a}-3.59\%$
test_compile_add_one_flat[tensorclass-eager]	0.1167ms	57.8298μs	17.2921 KOps/s	18.0417 KOps/s	$\color{#d91a1a}-4.15\%$
test_compile_add_one_flat[pytree-compile]	0.3015ms	0.1405ms	7.1170 KOps/s	7.3743 KOps/s	$\color{#d91a1a}-3.49\%$
test_compile_add_one_flat[pytree-eager]	0.6288ms	0.4675ms	2.1391 KOps/s	2.0045 KOps/s	$\textbf{\color{#35bf28}+6.71\%}$
test_compile_add_self_flat[tensordict-eager]	0.3698ms	0.2615ms	3.8247 KOps/s	3.8351 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_add_self_flat[tensordict-compile]	0.2079ms	0.1452ms	6.8868 KOps/s	6.9847 KOps/s	$\color{#d91a1a}-1.40\%$
test_compile_add_self_flat[tensorclass-eager]	0.1730ms	69.2836μs	14.4334 KOps/s	14.2724 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_add_self_flat[tensorclass-compile]	0.2183ms	0.1009ms	9.9068 KOps/s	9.9217 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_add_self_flat[pytree-eager]	0.4969ms	0.4135ms	2.4184 KOps/s	2.4572 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_add_self_flat[pytree-compile]	0.2204ms	0.1407ms	7.1062 KOps/s	7.4153 KOps/s	$\color{#d91a1a}-4.17\%$
test_compile_copy_flat[tensordict-compile]	0.1422ms	19.6293μs	50.9444 KOps/s	57.0639 KOps/s	$\textbf{\color{#d91a1a}-10.72\%}$
test_compile_copy_flat[tensordict-eager]	0.1018ms	31.4812μs	31.7650 KOps/s	31.6481 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_copy_flat[pytree-compile]	0.1733ms	70.6112μs	14.1621 KOps/s	14.0505 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_copy_flat[pytree-eager]	0.1028ms	51.5393μs	19.4027 KOps/s	19.1815 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_assign_and_add[tensordict-compile]	1.6342ms	0.3916ms	2.5536 KOps/s	2.2010 KOps/s	$\textbf{\color{#35bf28}+16.02\%}$
test_compile_assign_and_add[tensordict-eager]	2.8176ms	2.5936ms	385.5665 Ops/s	380.9229 Ops/s	$\color{#35bf28}+1.22\%$
test_compile_assign_and_add[pytree-compile]	1.5819ms	0.3789ms	2.6392 KOps/s	2.2639 KOps/s	$\textbf{\color{#35bf28}+16.58\%}$
test_compile_assign_and_add[pytree-eager]	2.7231ms	2.5912ms	385.9177 Ops/s	370.7348 Ops/s	$\color{#35bf28}+4.10\%$
test_compile_indexing[tensor-tensordict-compile]	0.5608ms	0.1130ms	8.8521 KOps/s	8.6960 KOps/s	$\color{#35bf28}+1.80\%$
test_compile_indexing[tensor-tensordict-eager]	0.5487ms	76.9970μs	12.9875 KOps/s	12.3559 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.6673ms	0.1057ms	9.4651 KOps/s	9.4055 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1359ms	66.4143μs	15.0570 KOps/s	14.5283 KOps/s	$\color{#35bf28}+3.64\%$
test_compile_indexing[tensor-pytree-compile]	0.1556ms	0.1059ms	9.4454 KOps/s	9.2921 KOps/s	$\color{#35bf28}+1.65\%$
test_compile_indexing[tensor-pytree-eager]	0.2178ms	66.5060μs	15.0362 KOps/s	14.5647 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_indexing[slice-tensordict-compile]	0.1643ms	0.1011ms	9.8881 KOps/s	9.8573 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_indexing[slice-tensordict-eager]	0.1452ms	16.9531μs	58.9862 KOps/s	54.7118 KOps/s	$\textbf{\color{#35bf28}+7.81\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1590ms	95.4939μs	10.4719 KOps/s	10.2186 KOps/s	$\color{#35bf28}+2.48\%$
test_compile_indexing[slice-tensorclass-eager]	48.0410μs	15.5154μs	64.4519 KOps/s	61.2196 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_compile_indexing[slice-pytree-compile]	0.1579ms	96.7283μs	10.3382 KOps/s	10.1430 KOps/s	$\color{#35bf28}+1.92\%$
test_compile_indexing[slice-pytree-eager]	62.2410μs	15.4962μs	64.5320 KOps/s	61.8627 KOps/s	$\color{#35bf28}+4.31\%$
test_compile_indexing[int-tensordict-compile]	0.1767ms	0.1011ms	9.8933 KOps/s	9.7908 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_indexing[int-tensordict-eager]	0.5887ms	16.4854μs	60.6599 KOps/s	55.6476 KOps/s	$\textbf{\color{#35bf28}+9.01\%}$
test_compile_indexing[int-tensorclass-compile]	0.1443ms	96.4161μs	10.3717 KOps/s	10.1497 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_indexing[int-tensorclass-eager]	68.6610μs	15.7001μs	63.6939 KOps/s	61.4871 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_indexing[int-pytree-compile]	0.1589ms	96.4128μs	10.3721 KOps/s	10.2057 KOps/s	$\color{#35bf28}+1.63\%$
test_compile_indexing[int-pytree-eager]	0.1238ms	15.6448μs	63.9192 KOps/s	60.6470 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_mod_add[eager]	0.1119ms	36.9892μs	27.0349 KOps/s	24.6798 KOps/s	$\textbf{\color{#35bf28}+9.54\%}$
test_mod_add[compile]	0.4188ms	78.6637μs	12.7123 KOps/s	12.3568 KOps/s	$\color{#35bf28}+2.88\%$
test_mod_add[compile-overhead]	0.3253ms	0.1662ms	6.0152 KOps/s	5.5719 KOps/s	$\textbf{\color{#35bf28}+7.96\%}$
test_mod_wrap[eager]	0.3251ms	0.2453ms	4.0767 KOps/s	3.8522 KOps/s	$\textbf{\color{#35bf28}+5.83\%}$
test_mod_wrap[compile]	0.3473ms	0.2816ms	3.5505 KOps/s	3.4700 KOps/s	$\color{#35bf28}+2.32\%$
test_mod_wrap[compile-overhead]	7.1118ms	3.7051ms	269.8953 Ops/s	273.8929 Ops/s	$\color{#d91a1a}-1.46\%$
test_mod_wrap_and_backward[eager]	1.4734ms	1.3527ms	739.2662 Ops/s	697.5739 Ops/s	$\textbf{\color{#35bf28}+5.98\%}$
test_mod_wrap_and_backward[compile]	1.4439ms	1.2685ms	788.3328 Ops/s	722.4914 Ops/s	$\textbf{\color{#35bf28}+9.11\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3902ms	0.9424ms	1.0611 KOps/s	931.2414 Ops/s	$\textbf{\color{#35bf28}+13.95\%}$
test_seq_add[eager]	0.2349ms	0.1123ms	8.9051 KOps/s	8.2029 KOps/s	$\textbf{\color{#35bf28}+8.56\%}$
test_seq_add[compile]	0.6551ms	88.8802μs	11.2511 KOps/s	11.2609 KOps/s	$\color{#d91a1a}-0.09\%$
test_seq_add[compile-overhead]	0.1795ms	0.1282ms	7.8018 KOps/s	7.6954 KOps/s	$\color{#35bf28}+1.38\%$
test_seq_wrap[eager]	0.4830ms	0.4164ms	2.4018 KOps/s	2.3259 KOps/s	$\color{#35bf28}+3.26\%$
test_seq_wrap[compile]	0.3695ms	0.2983ms	3.3523 KOps/s	3.2648 KOps/s	$\color{#35bf28}+2.68\%$
test_seq_wrap[compile-overhead]	0.2884ms	0.2243ms	4.4582 KOps/s	4.3611 KOps/s	$\color{#35bf28}+2.23\%$
test_func_call_runtime[False-eager]	0.8983ms	0.7346ms	1.3612 KOps/s	1.3463 KOps/s	$\color{#35bf28}+1.10\%$
test_func_call_runtime[False-compile]	0.9994ms	0.7392ms	1.3528 KOps/s	1.3184 KOps/s	$\color{#35bf28}+2.61\%$
test_func_call_runtime[False-compile-overhead]	0.4142ms	0.3647ms	2.7422 KOps/s	2.6999 KOps/s	$\color{#35bf28}+1.57\%$
test_func_call_runtime[True-eager]	0.9822ms	0.8916ms	1.1216 KOps/s	1.0979 KOps/s	$\color{#35bf28}+2.16\%$
test_func_call_runtime[True-compile]	0.8249ms	0.7586ms	1.3182 KOps/s	1.2366 KOps/s	$\textbf{\color{#35bf28}+6.60\%}$
test_func_call_runtime[True-compile-overhead]	0.4330ms	0.3885ms	2.5743 KOps/s	2.5647 KOps/s	$\color{#35bf28}+0.37\%$
test_func_call_cm_runtime[False-eager]	0.8717ms	0.7282ms	1.3733 KOps/s	1.2776 KOps/s	$\textbf{\color{#35bf28}+7.49\%}$
test_func_call_cm_runtime[False-compile]	0.8325ms	0.7401ms	1.3512 KOps/s	1.3255 KOps/s	$\color{#35bf28}+1.94\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4177ms	0.3701ms	2.7020 KOps/s	2.6946 KOps/s	$\color{#35bf28}+0.28\%$
test_func_call_cm_runtime[True-eager]	1.1022ms	0.9916ms	1.0085 KOps/s	983.1563 Ops/s	$\color{#35bf28}+2.58\%$
test_func_call_cm_runtime[True-compile]	0.9024ms	0.8039ms	1.2439 KOps/s	1.2425 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4808ms	0.4145ms	2.4124 KOps/s	2.4000 KOps/s	$\color{#35bf28}+0.52\%$
test_vmap_func_call_cm_runtime[eager]	2.5823ms	2.0963ms	477.0281 Ops/s	469.2623 Ops/s	$\color{#35bf28}+1.65\%$
test_vmap_func_call_cm_runtime[compile]	0.8843ms	0.8024ms	1.2463 KOps/s	1.2181 KOps/s	$\color{#35bf28}+2.32\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4638ms	0.4166ms	2.4005 KOps/s	2.3897 KOps/s	$\color{#35bf28}+0.45\%$
test_distributed	2.3749ms	0.2907ms	3.4405 KOps/s	8.5398 KOps/s	$\textbf{\color{#d91a1a}-59.71\%}$
test_tdmodule	0.1328ms	18.9237μs	52.8438 KOps/s	47.3714 KOps/s	$\textbf{\color{#35bf28}+11.55\%}$
test_tdmodule_dispatch	57.1720μs	33.9015μs	29.4972 KOps/s	25.9284 KOps/s	$\textbf{\color{#35bf28}+13.76\%}$
test_tdseq	39.9910μs	20.0451μs	49.8874 KOps/s	45.2606 KOps/s	$\textbf{\color{#35bf28}+10.22\%}$
test_tdseq_dispatch	69.2620μs	37.0443μs	26.9947 KOps/s	24.1426 KOps/s	$\textbf{\color{#35bf28}+11.81\%}$
test_instantiation_functorch	1.6418ms	1.5643ms	639.2774 Ops/s	630.5299 Ops/s	$\color{#35bf28}+1.39\%$
test_exec_functorch	0.2884ms	0.1439ms	6.9481 KOps/s	6.6202 KOps/s	$\color{#35bf28}+4.95\%$
test_exec_functional_call	0.1868ms	0.1377ms	7.2644 KOps/s	6.9058 KOps/s	$\textbf{\color{#35bf28}+5.19\%}$
test_exec_td_decorator	0.3697ms	0.1863ms	5.3679 KOps/s	5.1756 KOps/s	$\color{#35bf28}+3.72\%$
test_vmap_mlp_speed_decorator[True-True]	0.7997ms	0.6779ms	1.4750 KOps/s	1.4307 KOps/s	$\color{#35bf28}+3.10\%$
test_vmap_mlp_speed_decorator[True-False]	0.8800ms	0.6755ms	1.4804 KOps/s	1.4384 KOps/s	$\color{#35bf28}+2.92\%$
test_vmap_mlp_speed_decorator[False-True]	0.7443ms	0.5916ms	1.6903 KOps/s	1.6530 KOps/s	$\color{#35bf28}+2.26\%$
test_vmap_mlp_speed_decorator[False-False]	0.7258ms	0.5959ms	1.6781 KOps/s	1.6622 KOps/s	$\color{#35bf28}+0.96\%$
test_vmap_transformer_speed_decorator[True-True]	19.2996ms	19.1187ms	52.3048 Ops/s	51.6694 Ops/s	$\color{#35bf28}+1.23\%$
test_vmap_transformer_speed_decorator[True-False]	19.3472ms	19.2011ms	52.0804 Ops/s	51.5810 Ops/s	$\color{#35bf28}+0.97\%$
test_vmap_transformer_speed_decorator[False-True]	19.3026ms	19.0903ms	52.3826 Ops/s	52.0455 Ops/s	$\color{#35bf28}+0.65\%$
test_vmap_transformer_speed_decorator[False-False]	19.1598ms	19.0645ms	52.4536 Ops/s	52.1080 Ops/s	$\color{#35bf28}+0.66\%$
test_to_module_speed[True]	1.0585ms	0.9693ms	1.0316 KOps/s	1.0098 KOps/s	$\color{#35bf28}+2.16\%$
test_to_module_speed[False]	1.5731ms	0.9593ms	1.0424 KOps/s	1.0413 KOps/s	$\color{#35bf28}+0.11\%$
test_tc_init	63.6510μs	34.9899μs	28.5797 KOps/s	26.1232 KOps/s	$\textbf{\color{#35bf28}+9.40\%}$
test_tc_init_nested	0.1122ms	69.1364μs	14.4642 KOps/s	13.1342 KOps/s	$\textbf{\color{#35bf28}+10.13\%}$
test_tc_first_layer_tensor	11.5131μs	0.7031μs	1.4223 MOps/s	1.2385 MOps/s	$\textbf{\color{#35bf28}+14.84\%}$
test_tc_first_layer_nontensor	0.1196ms	2.2847μs	437.6899 KOps/s	435.7543 KOps/s	$\color{#35bf28}+0.44\%$
test_tc_second_layer_tensor	13.1070μs	1.4495μs	689.8952 KOps/s	692.5165 KOps/s	$\color{#d91a1a}-0.38\%$
test_tc_second_layer_nontensor	29.5610μs	3.0191μs	331.2243 KOps/s	330.5573 KOps/s	$\color{#35bf28}+0.20\%$
test_unbind	0.2195s	10.2791ms	97.2845 Ops/s	79.5540 Ops/s	$\textbf{\color{#35bf28}+22.29\%}$
test_full_like	9.8868ms	9.3655ms	106.7754 Ops/s	105.1840 Ops/s	$\color{#35bf28}+1.51\%$
test_zeros_like	4.9667ms	4.3554ms	229.6003 Ops/s	229.8851 Ops/s	$\color{#d91a1a}-0.12\%$
test_ones_like	4.9824ms	4.3585ms	229.4355 Ops/s	231.9223 Ops/s	$\color{#d91a1a}-1.07\%$
test_clone	6.9558ms	6.5908ms	151.7277 Ops/s	152.6416 Ops/s	$\color{#d91a1a}-0.60\%$
test_squeeze	68.6310μs	9.8832μs	101.1822 KOps/s	104.2521 KOps/s	$\color{#d91a1a}-2.94\%$
test_unsqueeze	0.2155ms	73.3617μs	13.6311 KOps/s	13.4621 KOps/s	$\color{#35bf28}+1.26\%$
test_split	0.2653ms	0.1616ms	6.1879 KOps/s	5.9717 KOps/s	$\color{#35bf28}+3.62\%$
test_permute	0.2649ms	0.1869ms	5.3514 KOps/s	5.5660 KOps/s	$\color{#d91a1a}-3.86\%$
test_stack	51.9036ms	51.1818ms	19.5382 Ops/s	19.5287 Ops/s	$\color{#35bf28}+0.05\%$
test_cat	51.4121ms	50.6159ms	19.7566 Ops/s	19.5819 Ops/s	$\color{#35bf28}+0.89\%$