[Feature] TensorClass

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] TensorClass #1067
• [Feature] inplace to method #1066

[github-actions]

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

Total Benchmarks: 216. Improved: $\large\color{#35bf28}36$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	47.0980μs	21.6691μs	46.1487 KOps/s	44.6897 KOps/s	$\color{#35bf28}+3.26\%$
test_plain_set_stack_nested	59.2620μs	21.8120μs	45.8463 KOps/s	44.7200 KOps/s	$\color{#35bf28}+2.52\%$
test_plain_set_nested_inplace	83.3270μs	24.6961μs	40.4922 KOps/s	41.2914 KOps/s	$\color{#d91a1a}-1.94\%$
test_plain_set_stack_nested_inplace	75.4920μs	24.0003μs	41.6662 KOps/s	41.1298 KOps/s	$\color{#35bf28}+1.30\%$
test_items	38.8330μs	4.2389μs	235.9091 KOps/s	241.0812 KOps/s	$\color{#d91a1a}-2.15\%$
test_items_nested	0.4003ms	0.3386ms	2.9536 KOps/s	2.9678 KOps/s	$\color{#d91a1a}-0.48\%$
test_items_nested_locked	0.4082ms	0.3368ms	2.9688 KOps/s	2.9534 KOps/s	$\color{#35bf28}+0.52\%$
test_items_nested_leaf	0.1335ms	71.5137μs	13.9833 KOps/s	14.0383 KOps/s	$\color{#d91a1a}-0.39\%$
test_items_stack_nested	0.4084ms	0.3428ms	2.9175 KOps/s	2.9527 KOps/s	$\color{#d91a1a}-1.19\%$
test_items_stack_nested_leaf	0.1427ms	74.8301μs	13.3636 KOps/s	13.2319 KOps/s	$\color{#35bf28}+1.00\%$
test_items_stack_nested_locked	0.4167ms	0.3413ms	2.9300 KOps/s	2.9403 KOps/s	$\color{#d91a1a}-0.35\%$
test_keys	27.0600μs	3.4976μs	285.9111 KOps/s	263.1569 KOps/s	$\textbf{\color{#35bf28}+8.65\%}$
test_keys_nested	0.1989ms	0.1370ms	7.2970 KOps/s	7.1736 KOps/s	$\color{#35bf28}+1.72\%$
test_keys_nested_locked	1.6835ms	0.1432ms	6.9816 KOps/s	6.9827 KOps/s	$\color{#d91a1a}-0.02\%$
test_keys_nested_leaf	0.1749ms	0.1183ms	8.4538 KOps/s	8.5425 KOps/s	$\color{#d91a1a}-1.04\%$
test_keys_stack_nested	0.2069ms	0.1359ms	7.3583 KOps/s	7.2589 KOps/s	$\color{#35bf28}+1.37\%$
test_keys_stack_nested_leaf	0.1894ms	0.1158ms	8.6385 KOps/s	8.5759 KOps/s	$\color{#35bf28}+0.73\%$
test_keys_stack_nested_locked	0.2083ms	0.1401ms	7.1387 KOps/s	7.0335 KOps/s	$\color{#35bf28}+1.50\%$
test_values	25.0650μs	1.0334μs	967.7138 KOps/s	959.9273 KOps/s	$\color{#35bf28}+0.81\%$
test_values_nested	0.1230ms	55.5870μs	17.9898 KOps/s	18.1269 KOps/s	$\color{#d91a1a}-0.76\%$
test_values_nested_locked	0.1145ms	55.5089μs	18.0151 KOps/s	18.0557 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_nested_leaf	0.1177ms	60.1662μs	16.6206 KOps/s	16.1619 KOps/s	$\color{#35bf28}+2.84\%$
test_values_stack_nested	0.3827ms	59.1165μs	16.9157 KOps/s	17.6057 KOps/s	$\color{#d91a1a}-3.92\%$
test_values_stack_nested_leaf	0.1178ms	60.9096μs	16.4178 KOps/s	16.4407 KOps/s	$\color{#d91a1a}-0.14\%$
test_values_stack_nested_locked	0.1027ms	57.1552μs	17.4962 KOps/s	17.5490 KOps/s	$\color{#d91a1a}-0.30\%$
test_membership	18.7650μs	0.8738μs	1.1445 MOps/s	1.0969 MOps/s	$\color{#35bf28}+4.34\%$
test_membership_nested	25.9890μs	2.7517μs	363.4104 KOps/s	362.1223 KOps/s	$\color{#35bf28}+0.36\%$
test_membership_nested_leaf	30.6770μs	2.7755μs	360.2902 KOps/s	369.2515 KOps/s	$\color{#d91a1a}-2.43\%$
test_membership_stacked_nested	19.4960μs	2.7460μs	364.1634 KOps/s	369.4181 KOps/s	$\color{#d91a1a}-1.42\%$
test_membership_stacked_nested_leaf	23.6250μs	2.7787μs	359.8838 KOps/s	366.9347 KOps/s	$\color{#d91a1a}-1.92\%$
test_membership_nested_last	32.4110μs	4.0672μs	245.8689 KOps/s	244.6870 KOps/s	$\color{#35bf28}+0.48\%$
test_membership_nested_leaf_last	41.4070μs	4.1001μs	243.8974 KOps/s	232.3761 KOps/s	$\color{#35bf28}+4.96\%$
test_membership_stacked_nested_last	52.4590μs	5.6872μs	175.8345 KOps/s	215.7172 KOps/s	$\textbf{\color{#d91a1a}-18.49\%}$
test_membership_stacked_nested_leaf_last	46.7170μs	5.7508μs	173.8886 KOps/s	212.5569 KOps/s	$\textbf{\color{#d91a1a}-18.19\%}$
test_nested_getleaf	62.1760μs	10.7111μs	93.3609 KOps/s	91.5709 KOps/s	$\color{#35bf28}+1.95\%$
test_nested_get	44.7840μs	10.2789μs	97.2868 KOps/s	97.3528 KOps/s	$\color{#d91a1a}-0.07\%$
test_stacked_getleaf	33.2930μs	10.7729μs	92.8258 KOps/s	93.0206 KOps/s	$\color{#d91a1a}-0.21\%$
test_stacked_get	55.7050μs	9.9044μs	100.9656 KOps/s	95.7751 KOps/s	$\textbf{\color{#35bf28}+5.42\%}$
test_nested_getitemleaf	41.0980μs	11.3605μs	88.0241 KOps/s	87.4537 KOps/s	$\color{#35bf28}+0.65\%$
test_nested_getitem	47.5400μs	10.5099μs	95.1480 KOps/s	93.6600 KOps/s	$\color{#35bf28}+1.59\%$
test_stacked_getitemleaf	46.0560μs	10.9616μs	91.2272 KOps/s	87.6330 KOps/s	$\color{#35bf28}+4.10\%$
test_stacked_getitem	53.8510μs	10.2310μs	97.7422 KOps/s	93.5661 KOps/s	$\color{#35bf28}+4.46\%$
test_lock_nested	2.9294ms	0.5014ms	1.9945 KOps/s	2.0241 KOps/s	$\color{#d91a1a}-1.46\%$
test_lock_stack_nested	0.6664ms	0.4565ms	2.1908 KOps/s	2.1585 KOps/s	$\color{#35bf28}+1.49\%$
test_unlock_nested	1.2489ms	0.4238ms	2.3596 KOps/s	2.4050 KOps/s	$\color{#d91a1a}-1.88\%$
test_unlock_stack_nested	0.7803ms	0.3748ms	2.6683 KOps/s	2.6367 KOps/s	$\color{#35bf28}+1.20\%$
test_flatten_speed	0.2066ms	92.6166μs	10.7972 KOps/s	10.8828 KOps/s	$\color{#d91a1a}-0.79\%$
test_unflatten_speed	0.7188ms	0.4828ms	2.0711 KOps/s	2.0606 KOps/s	$\color{#35bf28}+0.51\%$
test_common_ops	4.4966ms	1.1538ms	866.7299 Ops/s	844.0529 Ops/s	$\color{#35bf28}+2.69\%$
test_creation	63.3720μs	2.1429μs	466.6559 KOps/s	455.2327 KOps/s	$\color{#35bf28}+2.51\%$
test_creation_empty	51.5870μs	18.5474μs	53.9159 KOps/s	49.8842 KOps/s	$\textbf{\color{#35bf28}+8.08\%}$
test_creation_nested_1	72.9370μs	21.9627μs	45.5316 KOps/s	43.8653 KOps/s	$\color{#35bf28}+3.80\%$
test_creation_nested_2	71.8150μs	25.9712μs	38.5042 KOps/s	36.6004 KOps/s	$\textbf{\color{#35bf28}+5.20\%}$
test_clone	79.4290μs	17.1105μs	58.4438 KOps/s	57.5163 KOps/s	$\color{#35bf28}+1.61\%$
test_getitem[int]	1.0299ms	16.6165μs	60.1810 KOps/s	58.3663 KOps/s	$\color{#35bf28}+3.11\%$
test_getitem[slice_int]	0.1382ms	31.5114μs	31.7345 KOps/s	30.8247 KOps/s	$\color{#35bf28}+2.95\%$
test_getitem[range]	0.3014ms	57.6586μs	17.3435 KOps/s	16.8690 KOps/s	$\color{#35bf28}+2.81\%$
test_getitem[tuple]	0.1332ms	25.6508μs	38.9851 KOps/s	38.0368 KOps/s	$\color{#35bf28}+2.49\%$
test_getitem[list]	0.3745ms	53.1582μs	18.8118 KOps/s	18.1290 KOps/s	$\color{#35bf28}+3.77\%$
test_setitem_dim[int]	66.4050μs	34.9795μs	28.5881 KOps/s	29.5087 KOps/s	$\color{#d91a1a}-3.12\%$
test_setitem_dim[slice_int]	0.1066ms	63.6999μs	15.6986 KOps/s	15.6007 KOps/s	$\color{#35bf28}+0.63\%$
test_setitem_dim[range]	0.1454ms	84.4410μs	11.8426 KOps/s	11.4466 KOps/s	$\color{#35bf28}+3.46\%$
test_setitem_dim[tuple]	0.1107ms	51.9453μs	19.2510 KOps/s	19.4319 KOps/s	$\color{#d91a1a}-0.93\%$
test_setitem	0.2069ms	31.1858μs	32.0658 KOps/s	32.2095 KOps/s	$\color{#d91a1a}-0.45\%$
test_set	0.2503ms	30.1797μs	33.1349 KOps/s	32.3821 KOps/s	$\color{#35bf28}+2.32\%$
test_set_shared	3.4797ms	0.2213ms	4.5189 KOps/s	4.5435 KOps/s	$\color{#d91a1a}-0.54\%$
test_update	0.1254ms	37.7684μs	26.4772 KOps/s	25.5252 KOps/s	$\color{#35bf28}+3.73\%$
test_update_nested	0.2220ms	49.0160μs	20.4015 KOps/s	19.7331 KOps/s	$\color{#35bf28}+3.39\%$
test_update__nested	0.8112ms	41.1307μs	24.3128 KOps/s	24.1470 KOps/s	$\color{#35bf28}+0.69\%$
test_set_nested	0.1757ms	32.9368μs	30.3612 KOps/s	30.1722 KOps/s	$\color{#35bf28}+0.63\%$
test_set_nested_new	0.1931ms	38.1623μs	26.2038 KOps/s	26.2503 KOps/s	$\color{#d91a1a}-0.18\%$
test_select	0.2491ms	57.1633μs	17.4938 KOps/s	17.5913 KOps/s	$\color{#d91a1a}-0.55\%$
test_select_nested	0.1165ms	60.4606μs	16.5397 KOps/s	16.2548 KOps/s	$\color{#35bf28}+1.75\%$
test_exclude_nested	0.1613ms	76.2382μs	13.1168 KOps/s	13.2684 KOps/s	$\color{#d91a1a}-1.14\%$
test_empty[True]	0.7308ms	0.3535ms	2.8287 KOps/s	2.8597 KOps/s	$\color{#d91a1a}-1.08\%$
test_empty[False]	8.5160μs	1.2178μs	821.1436 KOps/s	791.7975 KOps/s	$\color{#35bf28}+3.71\%$
test_unbind_speed	0.3893ms	0.3044ms	3.2853 KOps/s	3.2783 KOps/s	$\color{#35bf28}+0.21\%$
test_unbind_speed_stack0	0.5268ms	0.2947ms	3.3931 KOps/s	3.3453 KOps/s	$\color{#35bf28}+1.43\%$
test_unbind_speed_stack1	0.1027s	0.7328ms	1.3646 KOps/s	1.3414 KOps/s	$\color{#35bf28}+1.73\%$
test_split	0.1017s	2.1963ms	455.3178 Ops/s	493.7733 Ops/s	$\textbf{\color{#d91a1a}-7.79\%}$
test_chunk	0.1109s	2.2552ms	443.4129 Ops/s	403.2814 Ops/s	$\textbf{\color{#35bf28}+9.95\%}$
test_creation[device0]	3.8830ms	0.1190ms	8.4033 KOps/s	8.2148 KOps/s	$\color{#35bf28}+2.29\%$
test_creation_from_tensor	0.2341ms	0.1161ms	8.6117 KOps/s	8.5142 KOps/s	$\color{#35bf28}+1.15\%$
test_add_one[memmap_tensor0]	0.2274ms	6.7967μs	147.1310 KOps/s	133.6397 KOps/s	$\textbf{\color{#35bf28}+10.10\%}$
test_contiguous[memmap_tensor0]	28.8140μs	1.8791μs	532.1731 KOps/s	521.2989 KOps/s	$\color{#35bf28}+2.09\%$
test_stack[memmap_tensor0]	55.8350μs	5.3374μs	187.3572 KOps/s	171.6330 KOps/s	$\textbf{\color{#35bf28}+9.16\%}$
test_memmaptd_index	1.0627ms	0.4079ms	2.4518 KOps/s	2.4941 KOps/s	$\color{#d91a1a}-1.69\%$
test_memmaptd_index_astensor	0.7498ms	0.4864ms	2.0560 KOps/s	2.0459 KOps/s	$\color{#35bf28}+0.49\%$
test_memmaptd_index_op	1.6655ms	1.0186ms	981.7624 Ops/s	943.0306 Ops/s	$\color{#35bf28}+4.11\%$
test_serialize_model	0.1259s	0.1193s	8.3827 Ops/s	8.1005 Ops/s	$\color{#35bf28}+3.48\%$
test_serialize_model_pickle	0.5000s	0.3993s	2.5043 Ops/s	2.4456 Ops/s	$\color{#35bf28}+2.40\%$
test_serialize_weights	0.1239s	0.1170s	8.5503 Ops/s	7.4303 Ops/s	$\textbf{\color{#35bf28}+15.07\%}$
test_serialize_weights_returnearly	0.1676s	0.1615s	6.1935 Ops/s	6.3381 Ops/s	$\color{#d91a1a}-2.28\%$
test_serialize_weights_pickle	0.4421s	0.3940s	2.5380 Ops/s	1.1961 Ops/s	$\textbf{\color{#35bf28}+112.19\%}$
test_serialize_weights_filesystem	0.2338s	0.1564s	6.3952 Ops/s	7.0722 Ops/s	$\textbf{\color{#d91a1a}-9.57\%}$
test_serialize_model_filesystem	0.1663s	0.1485s	6.7328 Ops/s	6.9817 Ops/s	$\color{#d91a1a}-3.56\%$
test_reshape_pytree	85.2600μs	40.2254μs	24.8599 KOps/s	25.8853 KOps/s	$\color{#d91a1a}-3.96\%$
test_reshape_td	0.1170ms	47.6988μs	20.9649 KOps/s	21.1155 KOps/s	$\color{#d91a1a}-0.71\%$
test_view_pytree	0.1332ms	41.0422μs	24.3652 KOps/s	25.7888 KOps/s	$\textbf{\color{#d91a1a}-5.52\%}$
test_view_td	0.1292ms	54.1758μs	18.4584 KOps/s	19.3936 KOps/s	$\color{#d91a1a}-4.82\%$
test_unbind_pytree	77.2550μs	36.0702μs	27.7237 KOps/s	27.7368 KOps/s	$\color{#d91a1a}-0.05\%$
test_unbind_td	0.3081ms	45.7526μs	21.8567 KOps/s	22.0838 KOps/s	$\color{#d91a1a}-1.03\%$
test_split_pytree	81.3530μs	39.2339μs	25.4881 KOps/s	26.3426 KOps/s	$\color{#d91a1a}-3.24\%$
test_split_td	0.4892ms	59.0510μs	16.9345 KOps/s	13.9101 KOps/s	$\textbf{\color{#35bf28}+21.74\%}$
test_add_pytree	0.1146ms	45.3325μs	22.0592 KOps/s	22.0956 KOps/s	$\color{#d91a1a}-0.16\%$
test_add_td	0.2373ms	82.6196μs	12.1037 KOps/s	11.8683 KOps/s	$\color{#35bf28}+1.98\%$
test_compile_add_one_nested[tensordict-compile]	0.1445ms	70.9069μs	14.1030 KOps/s	13.7377 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_add_one_nested[tensordict-eager]	1.4786ms	0.1913ms	5.2273 KOps/s	5.3341 KOps/s	$\color{#d91a1a}-2.00\%$
test_compile_add_one_nested[pytree-compile]	0.1706ms	54.4103μs	18.3789 KOps/s	17.7242 KOps/s	$\color{#35bf28}+3.69\%$
test_compile_add_one_nested[pytree-eager]	0.3264ms	0.1476ms	6.7772 KOps/s	6.7963 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_copy_nested[tensordict-compile]	0.1237ms	26.2615μs	38.0786 KOps/s	38.8271 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_copy_nested[tensordict-eager]	0.1743ms	72.0530μs	13.8787 KOps/s	13.9401 KOps/s	$\color{#d91a1a}-0.44\%$
test_compile_copy_nested[pytree-compile]	0.1492ms	78.5296μs	12.7341 KOps/s	12.4106 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_copy_nested[pytree-eager]	0.1245ms	67.1745μs	14.8866 KOps/s	14.5778 KOps/s	$\color{#35bf28}+2.12\%$
test_compile_add_one_flat[tensordict-compile]	0.2953ms	0.1140ms	8.7706 KOps/s	8.5440 KOps/s	$\color{#35bf28}+2.65\%$
test_compile_add_one_flat[tensordict-eager]	0.4175ms	0.2047ms	4.8841 KOps/s	4.8170 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_add_one_flat[tensorclass-compile]	0.1162ms	52.4163μs	19.0780 KOps/s	18.0405 KOps/s	$\textbf{\color{#35bf28}+5.75\%}$
test_compile_add_one_flat[tensorclass-eager]	0.4821ms	69.6046μs	14.3669 KOps/s	13.9718 KOps/s	$\color{#35bf28}+2.83\%$
test_compile_add_one_flat[pytree-compile]	0.2068ms	0.1111ms	8.9971 KOps/s	8.6797 KOps/s	$\color{#35bf28}+3.66\%$
test_compile_add_one_flat[pytree-eager]	0.4519ms	0.3037ms	3.2930 KOps/s	3.2790 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_add_self_flat[tensordict-eager]	0.3939ms	0.2204ms	4.5382 KOps/s	4.4905 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_add_self_flat[tensordict-compile]	0.1948ms	0.1136ms	8.8042 KOps/s	8.5685 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_add_self_flat[tensorclass-eager]	0.1290ms	64.0569μs	15.6111 KOps/s	15.7775 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_add_self_flat[tensorclass-compile]	0.1157ms	53.8392μs	18.5738 KOps/s	17.8605 KOps/s	$\color{#35bf28}+3.99\%$
test_compile_add_self_flat[pytree-eager]	0.7016ms	0.2455ms	4.0735 KOps/s	4.0396 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_add_self_flat[pytree-compile]	0.1852ms	0.1108ms	9.0252 KOps/s	8.8528 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_copy_flat[tensordict-compile]	61.3550μs	20.9979μs	47.6237 KOps/s	47.1007 KOps/s	$\color{#35bf28}+1.11\%$
test_compile_copy_flat[tensordict-eager]	0.2421ms	61.5593μs	16.2445 KOps/s	16.5590 KOps/s	$\color{#d91a1a}-1.90\%$
test_compile_copy_flat[pytree-compile]	0.1800ms	83.8966μs	11.9194 KOps/s	11.7378 KOps/s	$\color{#35bf28}+1.55\%$
test_compile_copy_flat[pytree-eager]	0.1443ms	70.7110μs	14.1421 KOps/s	14.5882 KOps/s	$\color{#d91a1a}-3.06\%$
test_compile_assign_and_add[tensordict-compile]	0.3120ms	0.2140ms	4.6725 KOps/s	4.5508 KOps/s	$\color{#35bf28}+2.67\%$
test_compile_assign_and_add[tensordict-eager]	3.1276ms	1.7542ms	570.0515 Ops/s	583.3555 Ops/s	$\color{#d91a1a}-2.28\%$
test_compile_assign_and_add[pytree-compile]	0.3919ms	0.2049ms	4.8813 KOps/s	4.6094 KOps/s	$\textbf{\color{#35bf28}+5.90\%}$
test_compile_assign_and_add[pytree-eager]	1.4373ms	1.1529ms	867.3451 Ops/s	845.9689 Ops/s	$\color{#35bf28}+2.53\%$
test_compile_assign_and_add_stack[compile]	0.5744ms	0.4566ms	2.1901 KOps/s	2.1643 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_assign_and_add_stack[eager]	5.1156ms	4.0862ms	244.7238 Ops/s	238.5217 Ops/s	$\color{#35bf28}+2.60\%$
test_compile_indexing[tensor-tensordict-compile]	88.9770μs	42.5907μs	23.4793 KOps/s	21.7164 KOps/s	$\textbf{\color{#35bf28}+8.12\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5092ms	52.3030μs	19.1193 KOps/s	19.6073 KOps/s	$\color{#d91a1a}-2.49\%$
test_compile_indexing[tensor-tensorclass-compile]	90.6700μs	36.4408μs	27.4418 KOps/s	25.7036 KOps/s	$\textbf{\color{#35bf28}+6.76\%}$
test_compile_indexing[tensor-tensorclass-eager]	92.9340μs	29.4205μs	33.9899 KOps/s	33.6677 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_indexing[tensor-pytree-compile]	0.1041ms	37.7086μs	26.5191 KOps/s	25.5277 KOps/s	$\color{#35bf28}+3.88\%$
test_compile_indexing[tensor-pytree-eager]	69.1590μs	29.5608μs	33.8286 KOps/s	33.9409 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_indexing[slice-tensordict-compile]	0.1504ms	77.4164μs	12.9172 KOps/s	12.5046 KOps/s	$\color{#35bf28}+3.30\%$
test_compile_indexing[slice-tensordict-eager]	0.6367ms	29.3248μs	34.1009 KOps/s	33.9223 KOps/s	$\color{#35bf28}+0.53\%$
test_compile_indexing[slice-tensorclass-compile]	0.1401ms	70.3220μs	14.2203 KOps/s	14.0305 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_indexing[slice-tensorclass-eager]	99.5170μs	24.1775μs	41.3608 KOps/s	42.7958 KOps/s	$\color{#d91a1a}-3.35\%$
test_compile_indexing[slice-pytree-compile]	0.1466ms	71.2391μs	14.0372 KOps/s	13.8599 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_indexing[slice-pytree-eager]	89.8090μs	24.1519μs	41.4046 KOps/s	42.1938 KOps/s	$\color{#d91a1a}-1.87\%$
test_compile_indexing[int-tensordict-compile]	0.1740ms	77.3837μs	12.9226 KOps/s	12.3786 KOps/s	$\color{#35bf28}+4.39\%$
test_compile_indexing[int-tensordict-eager]	0.8116ms	28.9180μs	34.5805 KOps/s	34.0035 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_indexing[int-tensorclass-compile]	0.1461ms	70.6763μs	14.1490 KOps/s	14.0708 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[int-tensorclass-eager]	77.8560μs	24.0560μs	41.5696 KOps/s	42.3332 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_indexing[int-pytree-compile]	0.1788ms	70.3922μs	14.2061 KOps/s	13.8921 KOps/s	$\color{#35bf28}+2.26\%$
test_compile_indexing[int-pytree-eager]	0.1026ms	24.1781μs	41.3597 KOps/s	42.3575 KOps/s	$\color{#d91a1a}-2.36\%$
test_mod_add[eager]	67.7770μs	26.5015μs	37.7337 KOps/s	35.0205 KOps/s	$\textbf{\color{#35bf28}+7.75\%}$
test_mod_add[compile]	94.3670μs	43.0500μs	23.2288 KOps/s	21.5786 KOps/s	$\textbf{\color{#35bf28}+7.65\%}$
test_mod_add[compile-overhead]	0.1225ms	43.4770μs	23.0007 KOps/s	21.4619 KOps/s	$\textbf{\color{#35bf28}+7.17\%}$
test_mod_wrap[eager]	0.3179ms	0.2098ms	4.7673 KOps/s	4.4622 KOps/s	$\textbf{\color{#35bf28}+6.84\%}$
test_mod_wrap[compile]	1.7863ms	0.2000ms	4.9991 KOps/s	4.6991 KOps/s	$\textbf{\color{#35bf28}+6.39\%}$
test_mod_wrap[compile-overhead]	1.9744ms	0.2032ms	4.9210 KOps/s	4.6991 KOps/s	$\color{#35bf28}+4.72\%$
test_mod_wrap_and_backward[eager]	12.0829ms	10.8927ms	91.8042 Ops/s	82.0790 Ops/s	$\textbf{\color{#35bf28}+11.85\%}$
test_mod_wrap_and_backward[compile]	12.4081ms	10.8510ms	92.1570 Ops/s	78.0107 Ops/s	$\textbf{\color{#35bf28}+18.13\%}$
test_mod_wrap_and_backward[compile-overhead]	12.5471ms	10.8282ms	92.3514 Ops/s	78.1344 Ops/s	$\textbf{\color{#35bf28}+18.20\%}$
test_seq_add[eager]	0.2110ms	91.2690μs	10.9566 KOps/s	10.6498 KOps/s	$\color{#35bf28}+2.88\%$
test_seq_add[compile]	0.1411ms	57.7571μs	17.3139 KOps/s	15.6952 KOps/s	$\textbf{\color{#35bf28}+10.31\%}$
test_seq_add[compile-overhead]	0.1424ms	57.8167μs	17.2960 KOps/s	16.1282 KOps/s	$\textbf{\color{#35bf28}+7.24\%}$
test_seq_wrap[eager]	0.6304ms	0.3867ms	2.5860 KOps/s	2.4010 KOps/s	$\textbf{\color{#35bf28}+7.71\%}$
test_seq_wrap[compile]	0.4212ms	0.2255ms	4.4344 KOps/s	4.2657 KOps/s	$\color{#35bf28}+3.95\%$
test_seq_wrap[compile-overhead]	0.3465ms	0.2266ms	4.4130 KOps/s	4.2821 KOps/s	$\color{#35bf28}+3.05\%$
test_func_call_runtime[False-eager]	0.9462ms	0.5472ms	1.8274 KOps/s	1.7133 KOps/s	$\textbf{\color{#35bf28}+6.66\%}$
test_func_call_runtime[False-compile]	0.6230ms	0.4271ms	2.3413 KOps/s	2.3083 KOps/s	$\color{#35bf28}+1.43\%$
test_func_call_runtime[False-compile-overhead]	0.5292ms	0.4222ms	2.3685 KOps/s	2.2977 KOps/s	$\color{#35bf28}+3.08\%$
test_func_call_runtime[True-eager]	1.2727ms	0.7518ms	1.3302 KOps/s	1.2527 KOps/s	$\textbf{\color{#35bf28}+6.18\%}$
test_func_call_runtime[True-compile]	0.9114ms	0.4713ms	2.1218 KOps/s	2.1034 KOps/s	$\color{#35bf28}+0.87\%$
test_func_call_runtime[True-compile-overhead]	0.8931ms	0.4691ms	2.1317 KOps/s	2.1025 KOps/s	$\color{#35bf28}+1.39\%$
test_func_call_cm_runtime[False-eager]	0.6819ms	0.5444ms	1.8369 KOps/s	1.7566 KOps/s	$\color{#35bf28}+4.57\%$
test_func_call_cm_runtime[False-compile]	0.5748ms	0.4217ms	2.3712 KOps/s	2.3230 KOps/s	$\color{#35bf28}+2.08\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6133ms	0.4198ms	2.3819 KOps/s	2.3377 KOps/s	$\color{#35bf28}+1.89\%$
test_func_call_cm_runtime[True-eager]	1.5166ms	0.8949ms	1.1175 KOps/s	1.0856 KOps/s	$\color{#35bf28}+2.94\%$
test_func_call_cm_runtime[True-compile]	0.6224ms	0.4871ms	2.0529 KOps/s	1.9854 KOps/s	$\color{#35bf28}+3.40\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6125ms	0.4874ms	2.0515 KOps/s	1.9846 KOps/s	$\color{#35bf28}+3.37\%$
test_vmap_func_call_cm_runtime[eager]	2.4716ms	1.9021ms	525.7209 Ops/s	513.6490 Ops/s	$\color{#35bf28}+2.35\%$
test_vmap_func_call_cm_runtime[compile]	0.8892ms	0.5199ms	1.9235 KOps/s	1.8955 KOps/s	$\color{#35bf28}+1.48\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.0392ms	0.5139ms	1.9461 KOps/s	1.8735 KOps/s	$\color{#35bf28}+3.87\%$
test_distributed	0.2623ms	0.1262ms	7.9265 KOps/s	7.7573 KOps/s	$\color{#35bf28}+2.18\%$
test_tdmodule	72.2050μs	18.8921μs	52.9321 KOps/s	48.6130 KOps/s	$\textbf{\color{#35bf28}+8.88\%}$
test_tdmodule_dispatch	58.5100μs	36.6185μs	27.3086 KOps/s	25.2531 KOps/s	$\textbf{\color{#35bf28}+8.14\%}$
test_tdseq	40.4560μs	21.5083μs	46.4937 KOps/s	42.6865 KOps/s	$\textbf{\color{#35bf28}+8.92\%}$
test_tdseq_dispatch	71.9950μs	41.9706μs	23.8262 KOps/s	19.7498 KOps/s	$\textbf{\color{#35bf28}+20.64\%}$
test_instantiation_functorch	2.2690ms	1.5681ms	637.7052 Ops/s	643.7547 Ops/s	$\color{#d91a1a}-0.94\%$
test_exec_functorch	0.3106ms	0.1803ms	5.5450 KOps/s	5.3079 KOps/s	$\color{#35bf28}+4.47\%$
test_exec_functional_call	0.2692ms	0.1745ms	5.7300 KOps/s	5.2897 KOps/s	$\textbf{\color{#35bf28}+8.32\%}$
test_exec_td_decorator	0.5560ms	0.2313ms	4.3240 KOps/s	4.1776 KOps/s	$\color{#35bf28}+3.50\%$
test_vmap_mlp_speed_decorator[True-True]	0.8809ms	0.6358ms	1.5729 KOps/s	1.5391 KOps/s	$\color{#35bf28}+2.20\%$
test_vmap_mlp_speed_decorator[True-False]	0.9802ms	0.6371ms	1.5696 KOps/s	1.5464 KOps/s	$\color{#35bf28}+1.49\%$
test_vmap_mlp_speed_decorator[False-True]	0.7049ms	0.5220ms	1.9156 KOps/s	1.8783 KOps/s	$\color{#35bf28}+1.99\%$
test_vmap_mlp_speed_decorator[False-False]	0.9257ms	0.5232ms	1.9115 KOps/s	1.8636 KOps/s	$\color{#35bf28}+2.57\%$
test_to_module_speed[True]	2.1072ms	1.3205ms	757.3168 Ops/s	767.0107 Ops/s	$\color{#d91a1a}-1.26\%$
test_to_module_speed[False]	2.5741ms	1.2722ms	786.0264 Ops/s	781.2388 Ops/s	$\color{#35bf28}+0.61\%$
test_tc_init	90.3300μs	45.0825μs	22.1816 KOps/s	21.4300 KOps/s	$\color{#35bf28}+3.51\%$
test_tc_init_nested	0.2231ms	90.3396μs	11.0693 KOps/s	10.9182 KOps/s	$\color{#35bf28}+1.38\%$
test_tc_first_layer_tensor	14.6970μs	1.5254μs	655.5606 KOps/s	617.6087 KOps/s	$\textbf{\color{#35bf28}+6.14\%}$
test_tc_first_layer_nontensor	48.9720μs	4.6139μs	216.7349 KOps/s	202.0358 KOps/s	$\textbf{\color{#35bf28}+7.28\%}$
test_tc_second_layer_tensor	26.4090μs	2.8592μs	349.7487 KOps/s	338.1270 KOps/s	$\color{#35bf28}+3.44\%$
test_tc_second_layer_nontensor	44.6640μs	5.9581μs	167.8393 KOps/s	158.1007 KOps/s	$\textbf{\color{#35bf28}+6.16\%}$
test_unbind	0.2309s	14.0331ms	71.2600 Ops/s	69.4205 Ops/s	$\color{#35bf28}+2.65\%$
test_full_like	9.1360ms	7.7330ms	129.3167 Ops/s	122.4797 Ops/s	$\textbf{\color{#35bf28}+5.58\%}$
test_zeros_like	3.9307ms	3.0308ms	329.9437 Ops/s	319.4744 Ops/s	$\color{#35bf28}+3.28\%$
test_ones_like	11.2754ms	6.4297ms	155.5281 Ops/s	269.3635 Ops/s	$\textbf{\color{#d91a1a}-42.26\%}$
test_clone	11.4835ms	8.2019ms	121.9226 Ops/s	176.3655 Ops/s	$\textbf{\color{#d91a1a}-30.87\%}$
test_squeeze	61.8260μs	11.7489μs	85.1142 KOps/s	81.7140 KOps/s	$\color{#35bf28}+4.16\%$
test_unsqueeze	0.1932ms	92.0852μs	10.8595 KOps/s	11.1197 KOps/s	$\color{#d91a1a}-2.34\%$
test_split	0.5019ms	0.1902ms	5.2576 KOps/s	5.0343 KOps/s	$\color{#35bf28}+4.43\%$
test_permute	0.4695ms	0.2208ms	4.5288 KOps/s	4.4760 KOps/s	$\color{#35bf28}+1.18\%$
test_stack	30.9539ms	24.9689ms	40.0498 Ops/s	38.4478 Ops/s	$\color{#35bf28}+4.17\%$
test_cat	29.9940ms	24.6875ms	40.5064 Ops/s	39.5534 Ops/s	$\color{#35bf28}+2.41\%$

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}25$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	40.9310μs	14.7692μs	67.7086 KOps/s	64.1804 KOps/s	$\textbf{\color{#35bf28}+5.50\%}$
test_plain_set_stack_nested	40.3310μs	14.9345μs	66.9592 KOps/s	63.9678 KOps/s	$\color{#35bf28}+4.68\%$
test_plain_set_nested_inplace	79.9020μs	15.6030μs	64.0903 KOps/s	60.8117 KOps/s	$\textbf{\color{#35bf28}+5.39\%}$
test_plain_set_stack_nested_inplace	45.0010μs	15.7440μs	63.5164 KOps/s	60.8583 KOps/s	$\color{#35bf28}+4.37\%$
test_items	23.1710μs	2.8863μs	346.4658 KOps/s	337.7092 KOps/s	$\color{#35bf28}+2.59\%$
test_items_nested	0.3736ms	0.3168ms	3.1567 KOps/s	3.1286 KOps/s	$\color{#35bf28}+0.90\%$
test_items_nested_locked	0.5011ms	0.3200ms	3.1251 KOps/s	3.1000 KOps/s	$\color{#35bf28}+0.81\%$
test_items_nested_leaf	0.1350ms	58.2768μs	17.1595 KOps/s	17.2829 KOps/s	$\color{#d91a1a}-0.71\%$
test_items_stack_nested	0.3946ms	0.3198ms	3.1269 KOps/s	3.1207 KOps/s	$\color{#35bf28}+0.20\%$
test_items_stack_nested_leaf	84.9910μs	58.5714μs	17.0732 KOps/s	17.3075 KOps/s	$\color{#d91a1a}-1.35\%$
test_items_stack_nested_locked	0.3545ms	0.3215ms	3.1107 KOps/s	3.0958 KOps/s	$\color{#35bf28}+0.48\%$
test_keys	22.5000μs	3.4618μs	288.8645 KOps/s	290.7143 KOps/s	$\color{#d91a1a}-0.64\%$
test_keys_nested	95.4020μs	69.4909μs	14.3904 KOps/s	14.4299 KOps/s	$\color{#d91a1a}-0.27\%$
test_keys_nested_locked	2.1936ms	75.8677μs	13.1808 KOps/s	13.1824 KOps/s	$\color{#d91a1a}-0.01\%$
test_keys_nested_leaf	83.8120μs	61.1417μs	16.3554 KOps/s	16.3772 KOps/s	$\color{#d91a1a}-0.13\%$
test_keys_stack_nested	0.1277ms	70.2464μs	14.2356 KOps/s	14.1882 KOps/s	$\color{#35bf28}+0.33\%$
test_keys_stack_nested_leaf	94.6820μs	61.6715μs	16.2149 KOps/s	16.4477 KOps/s	$\color{#d91a1a}-1.42\%$
test_keys_stack_nested_locked	0.1070ms	75.5448μs	13.2372 KOps/s	13.2896 KOps/s	$\color{#d91a1a}-0.39\%$
test_values	7.9202μs	0.8479μs	1.1794 MOps/s	1.1846 MOps/s	$\color{#d91a1a}-0.44\%$
test_values_nested	68.0510μs	31.2451μs	32.0050 KOps/s	32.0126 KOps/s	$\color{#d91a1a}-0.02\%$
test_values_nested_locked	62.2720μs	32.8273μs	30.4625 KOps/s	30.6381 KOps/s	$\color{#d91a1a}-0.57\%$
test_values_nested_leaf	65.1410μs	33.9329μs	29.4699 KOps/s	29.7464 KOps/s	$\color{#d91a1a}-0.93\%$
test_values_stack_nested	0.1010ms	31.5325μs	31.7133 KOps/s	31.8213 KOps/s	$\color{#d91a1a}-0.34\%$
test_values_stack_nested_leaf	69.5510μs	34.0712μs	29.3503 KOps/s	29.6004 KOps/s	$\color{#d91a1a}-0.84\%$
test_values_stack_nested_locked	62.2920μs	33.1042μs	30.2076 KOps/s	30.6206 KOps/s	$\color{#d91a1a}-1.35\%$
test_membership	1.9476μs	0.5254μs	1.9033 MOps/s	1.8704 MOps/s	$\color{#35bf28}+1.76\%$
test_membership_nested	28.2800μs	2.0006μs	499.8519 KOps/s	512.5612 KOps/s	$\color{#d91a1a}-2.48\%$
test_membership_nested_leaf	17.1400μs	1.9397μs	515.5484 KOps/s	519.3820 KOps/s	$\color{#d91a1a}-0.74\%$
test_membership_stacked_nested	40.4610μs	1.9823μs	504.4743 KOps/s	500.4168 KOps/s	$\color{#35bf28}+0.81\%$
test_membership_stacked_nested_leaf	24.0310μs	2.0311μs	492.3423 KOps/s	503.2185 KOps/s	$\color{#d91a1a}-2.16\%$
test_membership_nested_last	33.1410μs	2.8921μs	345.7743 KOps/s	353.0039 KOps/s	$\color{#d91a1a}-2.05\%$
test_membership_nested_leaf_last	26.2100μs	2.9401μs	340.1283 KOps/s	348.0999 KOps/s	$\color{#d91a1a}-2.29\%$
test_membership_stacked_nested_last	33.6000μs	2.9375μs	340.4200 KOps/s	353.8171 KOps/s	$\color{#d91a1a}-3.79\%$
test_membership_stacked_nested_leaf_last	41.0010μs	2.9159μs	342.9530 KOps/s	350.7468 KOps/s	$\color{#d91a1a}-2.22\%$
test_nested_getleaf	45.4110μs	5.9830μs	167.1390 KOps/s	165.9326 KOps/s	$\color{#35bf28}+0.73\%$
test_nested_get	27.0300μs	5.6603μs	176.6701 KOps/s	175.1031 KOps/s	$\color{#35bf28}+0.89\%$
test_stacked_getleaf	26.2010μs	6.0157μs	166.2309 KOps/s	166.4124 KOps/s	$\color{#d91a1a}-0.11\%$
test_stacked_get	25.7810μs	5.7028μs	175.3539 KOps/s	175.0191 KOps/s	$\color{#35bf28}+0.19\%$
test_nested_getitemleaf	31.6910μs	6.0790μs	164.4998 KOps/s	164.0602 KOps/s	$\color{#35bf28}+0.27\%$
test_nested_getitem	26.1800μs	5.7813μs	172.9706 KOps/s	171.9362 KOps/s	$\color{#35bf28}+0.60\%$
test_stacked_getitemleaf	39.0910μs	6.0742μs	164.6306 KOps/s	164.9232 KOps/s	$\color{#d91a1a}-0.18\%$
test_stacked_getitem	25.3710μs	5.7581μs	173.6699 KOps/s	173.3259 KOps/s	$\color{#35bf28}+0.20\%$
test_lock_nested	4.4199ms	0.4230ms	2.3642 KOps/s	2.3768 KOps/s	$\color{#d91a1a}-0.53\%$
test_lock_stack_nested	0.4272ms	0.3879ms	2.5778 KOps/s	2.5728 KOps/s	$\color{#35bf28}+0.19\%$
test_unlock_nested	0.7483ms	0.3592ms	2.7838 KOps/s	2.7607 KOps/s	$\color{#35bf28}+0.84\%$
test_unlock_stack_nested	0.3613ms	0.3278ms	3.0503 KOps/s	3.0372 KOps/s	$\color{#35bf28}+0.43\%$
test_flatten_speed	0.1009ms	72.4992μs	13.7933 KOps/s	13.8976 KOps/s	$\color{#d91a1a}-0.75\%$
test_unflatten_speed	0.3323ms	0.2931ms	3.4123 KOps/s	3.4314 KOps/s	$\color{#d91a1a}-0.56\%$
test_common_ops	1.6392ms	1.3255ms	754.4100 Ops/s	764.6901 Ops/s	$\color{#d91a1a}-1.34\%$
test_creation	22.3900μs	1.4796μs	675.8368 KOps/s	675.0861 KOps/s	$\color{#35bf28}+0.11\%$
test_creation_empty	39.9010μs	15.9120μs	62.8457 KOps/s	56.8534 KOps/s	$\textbf{\color{#35bf28}+10.54\%}$
test_creation_nested_1	44.5810μs	17.5314μs	57.0406 KOps/s	51.9563 KOps/s	$\textbf{\color{#35bf28}+9.79\%}$
test_creation_nested_2	52.9110μs	20.2592μs	49.3603 KOps/s	45.3931 KOps/s	$\textbf{\color{#35bf28}+8.74\%}$
test_clone	60.9220μs	30.0622μs	33.2644 KOps/s	32.7671 KOps/s	$\color{#35bf28}+1.52\%$
test_getitem[int]	1.3715ms	17.6106μs	56.7838 KOps/s	59.5408 KOps/s	$\color{#d91a1a}-4.63\%$
test_getitem[slice_int]	0.1301ms	29.5653μs	33.8234 KOps/s	34.4278 KOps/s	$\color{#d91a1a}-1.76\%$
test_getitem[range]	0.2419ms	0.1155ms	8.6585 KOps/s	8.6457 KOps/s	$\color{#35bf28}+0.15\%$
test_getitem[tuple]	0.1315ms	25.5194μs	39.1859 KOps/s	39.8456 KOps/s	$\color{#d91a1a}-1.66\%$
test_getitem[list]	0.2482ms	0.1037ms	9.6421 KOps/s	9.5958 KOps/s	$\color{#35bf28}+0.48\%$
test_setitem_dim[int]	78.2320μs	45.4524μs	22.0010 KOps/s	22.1147 KOps/s	$\color{#d91a1a}-0.51\%$
test_setitem_dim[slice_int]	0.1706ms	68.7206μs	14.5517 KOps/s	14.4572 KOps/s	$\color{#35bf28}+0.65\%$
test_setitem_dim[range]	0.2171ms	0.1303ms	7.6756 KOps/s	7.4199 KOps/s	$\color{#35bf28}+3.45\%$
test_setitem_dim[tuple]	88.2020μs	62.3738μs	16.0324 KOps/s	15.8673 KOps/s	$\color{#35bf28}+1.04\%$
test_setitem	79.2220μs	44.9676μs	22.2382 KOps/s	22.4883 KOps/s	$\color{#d91a1a}-1.11\%$
test_set	93.6320μs	43.7061μs	22.8801 KOps/s	22.8950 KOps/s	$\color{#d91a1a}-0.06\%$
test_set_shared	0.2896ms	52.6940μs	18.9775 KOps/s	19.0805 KOps/s	$\color{#d91a1a}-0.54\%$
test_update	89.8010μs	50.2513μs	19.9000 KOps/s	18.7422 KOps/s	$\textbf{\color{#35bf28}+6.18\%}$
test_update_nested	0.1127ms	57.4805μs	17.3972 KOps/s	16.4317 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_update__nested	0.1720ms	61.2665μs	16.3221 KOps/s	16.4370 KOps/s	$\color{#d91a1a}-0.70\%$
test_set_nested	91.8920μs	47.2153μs	21.1796 KOps/s	21.5886 KOps/s	$\color{#d91a1a}-1.89\%$
test_set_nested_new	79.8520μs	49.2636μs	20.2990 KOps/s	20.3380 KOps/s	$\color{#d91a1a}-0.19\%$
test_select	0.1059ms	62.9834μs	15.8772 KOps/s	15.8963 KOps/s	$\color{#d91a1a}-0.12\%$
test_select_nested	61.4110μs	42.7964μs	23.3664 KOps/s	23.9898 KOps/s	$\color{#d91a1a}-2.60\%$
test_exclude_nested	0.1090ms	59.2926μs	16.8655 KOps/s	17.0060 KOps/s	$\color{#d91a1a}-0.83\%$
test_empty[True]	0.3123ms	0.2540ms	3.9364 KOps/s	3.8774 KOps/s	$\color{#35bf28}+1.52\%$
test_empty[False]	3.9271μs	0.7455μs	1.3414 MOps/s	1.3280 MOps/s	$\color{#35bf28}+1.01\%$
test_to	80.8220μs	51.9391μs	19.2533 KOps/s	18.7278 KOps/s	$\color{#35bf28}+2.81\%$
test_to_nonblocking	89.3720μs	53.4363μs	18.7139 KOps/s	19.4615 KOps/s	$\color{#d91a1a}-3.84\%$
test_unbind_speed	0.3253ms	0.2779ms	3.5990 KOps/s	3.5774 KOps/s	$\color{#35bf28}+0.60\%$
test_unbind_speed_stack0	0.3368ms	0.2779ms	3.5978 KOps/s	3.6157 KOps/s	$\color{#d91a1a}-0.50\%$
test_unbind_speed_stack1	90.9222ms	0.7135ms	1.4016 KOps/s	1.4072 KOps/s	$\color{#d91a1a}-0.40\%$
test_split	92.4651ms	2.2716ms	440.2190 Ops/s	454.3692 Ops/s	$\color{#d91a1a}-3.11\%$
test_chunk	94.6731ms	2.2710ms	440.3359 Ops/s	454.4353 Ops/s	$\color{#d91a1a}-3.10\%$
test_to[False]	6.2807ms	6.1878ms	161.6073 Ops/s	152.0375 Ops/s	$\textbf{\color{#35bf28}+6.29\%}$
test_to[True]	4.8186ms	4.5433ms	220.1021 Ops/s	229.0148 Ops/s	$\color{#d91a1a}-3.89\%$
test_to_njt[False]	0.3496s	0.2712s	3.6871 Ops/s	3.6188 Ops/s	$\color{#35bf28}+1.89\%$
test_to_njt[True]	0.2628s	0.2606s	3.8371 Ops/s	3.5585 Ops/s	$\textbf{\color{#35bf28}+7.83\%}$
test_creation[device0]	0.3744ms	0.1289ms	7.7604 KOps/s	7.5778 KOps/s	$\color{#35bf28}+2.41\%$
test_creation_from_tensor	0.3631ms	0.1352ms	7.3991 KOps/s	7.3840 KOps/s	$\color{#35bf28}+0.20\%$
test_add_one[memmap_tensor0]	0.1346ms	9.4225μs	106.1285 KOps/s	107.1620 KOps/s	$\color{#d91a1a}-0.96\%$
test_contiguous[memmap_tensor0]	33.4810μs	2.2336μs	447.7077 KOps/s	449.5586 KOps/s	$\color{#d91a1a}-0.41\%$
test_stack[memmap_tensor0]	0.1600ms	7.2800μs	137.3634 KOps/s	142.5040 KOps/s	$\color{#d91a1a}-3.61\%$
test_memmaptd_index	0.9931ms	0.4403ms	2.2712 KOps/s	2.3508 KOps/s	$\color{#d91a1a}-3.39\%$
test_memmaptd_index_astensor	0.7515ms	0.4990ms	2.0040 KOps/s	2.0657 KOps/s	$\color{#d91a1a}-2.99\%$
test_memmaptd_index_op	1.4780ms	1.0521ms	950.4837 Ops/s	929.2200 Ops/s	$\color{#35bf28}+2.29\%$
test_serialize_model	0.2201s	0.1429s	6.9992 Ops/s	7.6734 Ops/s	$\textbf{\color{#d91a1a}-8.79\%}$
test_serialize_model_pickle	1.3475s	1.2128s	0.8246 Ops/s	0.8354 Ops/s	$\color{#d91a1a}-1.30\%$
test_serialize_weights	0.1302s	0.1288s	7.7669 Ops/s	7.7352 Ops/s	$\color{#35bf28}+0.41\%$
test_serialize_weights_returnearly	0.2058s	56.0316ms	17.8471 Ops/s	17.7817 Ops/s	$\color{#35bf28}+0.37\%$
test_serialize_weights_pickle	1.3813s	1.2229s	0.8177 Ops/s	0.8394 Ops/s	$\color{#d91a1a}-2.59\%$
test_reshape_pytree	73.6210μs	38.5658μs	25.9297 KOps/s	27.8487 KOps/s	$\textbf{\color{#d91a1a}-6.89\%}$
test_reshape_td	86.6620μs	43.0370μs	23.2358 KOps/s	24.5774 KOps/s	$\textbf{\color{#d91a1a}-5.46\%}$
test_view_pytree	72.3520μs	36.5942μs	27.3268 KOps/s	27.5929 KOps/s	$\color{#d91a1a}-0.96\%$
test_view_td	80.1310μs	47.6629μs	20.9807 KOps/s	21.7712 KOps/s	$\color{#d91a1a}-3.63\%$
test_unbind_pytree	0.2141ms	34.5106μs	28.9766 KOps/s	28.9376 KOps/s	$\color{#35bf28}+0.13\%$
test_unbind_td	0.5008ms	42.8923μs	23.3142 KOps/s	23.6254 KOps/s	$\color{#d91a1a}-1.32\%$
test_split_pytree	96.0920μs	49.9201μs	20.0320 KOps/s	20.9386 KOps/s	$\color{#d91a1a}-4.33\%$
test_split_td	94.3228ms	69.8527μs	14.3158 KOps/s	15.0623 KOps/s	$\color{#d91a1a}-4.96\%$
test_add_pytree	0.1015ms	58.3959μs	17.1245 KOps/s	16.8385 KOps/s	$\color{#35bf28}+1.70\%$
test_add_td	0.1411ms	94.4172μs	10.5913 KOps/s	10.2437 KOps/s	$\color{#35bf28}+3.39\%$
test_compile_add_one_nested[tensordict-compile]	0.2464ms	0.1667ms	5.9975 KOps/s	6.0217 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_add_one_nested[tensordict-eager]	0.2854ms	0.1540ms	6.4917 KOps/s	6.6476 KOps/s	$\color{#d91a1a}-2.34\%$
test_compile_add_one_nested[pytree-compile]	0.2130ms	0.1598ms	6.2587 KOps/s	6.2112 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_add_one_nested[pytree-eager]	0.3366ms	0.1872ms	5.3413 KOps/s	5.2594 KOps/s	$\color{#35bf28}+1.56\%$
test_compile_copy_nested[tensordict-compile]	54.5910μs	21.8630μs	45.7395 KOps/s	45.5562 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_copy_nested[tensordict-eager]	92.6220μs	44.8393μs	22.3018 KOps/s	22.5381 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_copy_nested[pytree-compile]	0.2231ms	66.4256μs	15.0544 KOps/s	15.2041 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_copy_nested[pytree-eager]	0.1309ms	49.8389μs	20.0647 KOps/s	19.9650 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_add_one_flat[tensordict-compile]	0.3717ms	0.3209ms	3.1164 KOps/s	3.1142 KOps/s	$\color{#35bf28}+0.07\%$
test_compile_add_one_flat[tensordict-eager]	0.3304ms	0.2196ms	4.5547 KOps/s	4.6927 KOps/s	$\color{#d91a1a}-2.94\%$
test_compile_add_one_flat[tensorclass-compile]	0.1756ms	0.1336ms	7.4869 KOps/s	7.4482 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_add_one_flat[tensorclass-eager]	0.2274ms	60.1450μs	16.6265 KOps/s	16.0335 KOps/s	$\color{#35bf28}+3.70\%$
test_compile_add_one_flat[pytree-compile]	0.3630ms	0.3285ms	3.0439 KOps/s	3.0331 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_add_one_flat[pytree-eager]	0.6840ms	0.6390ms	1.5650 KOps/s	1.5408 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_add_self_flat[tensordict-eager]	0.3610ms	0.2587ms	3.8658 KOps/s	3.8990 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_add_self_flat[tensordict-compile]	0.3735ms	0.3240ms	3.0867 KOps/s	3.0784 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_add_self_flat[tensorclass-eager]	0.2105ms	70.1885μs	14.2473 KOps/s	14.0808 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_add_self_flat[tensorclass-compile]	0.2817ms	0.1373ms	7.2848 KOps/s	7.3396 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_add_self_flat[pytree-eager]	0.7005ms	0.5330ms	1.8762 KOps/s	1.8404 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_self_flat[pytree-compile]	0.3850ms	0.3288ms	3.0415 KOps/s	3.0281 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_copy_flat[tensordict-compile]	67.6210μs	19.0295μs	52.5501 KOps/s	51.1601 KOps/s	$\color{#35bf28}+2.72\%$
test_compile_copy_flat[tensordict-eager]	68.8410μs	28.9199μs	34.5783 KOps/s	35.2808 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_copy_flat[pytree-compile]	0.1011ms	69.7279μs	14.3415 KOps/s	14.3321 KOps/s	$\color{#35bf28}+0.07\%$
test_compile_copy_flat[pytree-eager]	84.3220μs	52.1968μs	19.1583 KOps/s	19.2252 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_assign_and_add[tensordict-compile]	2.3934ms	0.8273ms	1.2087 KOps/s	1.1120 KOps/s	$\textbf{\color{#35bf28}+8.69\%}$
test_compile_assign_and_add[tensordict-eager]	3.4044ms	3.3067ms	302.4163 Ops/s	296.5887 Ops/s	$\color{#35bf28}+1.96\%$
test_compile_assign_and_add[pytree-compile]	2.4068ms	0.8350ms	1.1976 KOps/s	1.1007 KOps/s	$\textbf{\color{#35bf28}+8.81\%}$
test_compile_assign_and_add[pytree-eager]	3.4474ms	3.3371ms	299.6597 Ops/s	289.2570 Ops/s	$\color{#35bf28}+3.60\%$
test_compile_indexing[tensor-tensordict-compile]	0.1702ms	0.1238ms	8.0781 KOps/s	8.0592 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_indexing[tensor-tensordict-eager]	0.1880ms	61.7563μs	16.1927 KOps/s	15.6323 KOps/s	$\color{#35bf28}+3.58\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1673ms	0.1172ms	8.5315 KOps/s	8.4905 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_indexing[tensor-tensorclass-eager]	94.1620μs	42.4042μs	23.5825 KOps/s	22.1617 KOps/s	$\textbf{\color{#35bf28}+6.41\%}$
test_compile_indexing[tensor-pytree-compile]	0.3069ms	0.1180ms	8.4768 KOps/s	8.4002 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_indexing[tensor-pytree-eager]	85.3510μs	42.1777μs	23.7092 KOps/s	22.4585 KOps/s	$\textbf{\color{#35bf28}+5.57\%}$
test_compile_indexing[slice-tensordict-compile]	0.1845ms	0.1522ms	6.5700 KOps/s	6.5686 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_indexing[slice-tensordict-eager]	0.1646ms	26.3506μs	37.9498 KOps/s	36.9454 KOps/s	$\color{#35bf28}+2.72\%$
test_compile_indexing[slice-tensorclass-compile]	0.1917ms	0.1449ms	6.9004 KOps/s	6.9352 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_indexing[slice-tensorclass-eager]	59.5020μs	21.3556μs	46.8262 KOps/s	47.6932 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_indexing[slice-pytree-compile]	0.1951ms	0.1466ms	6.8220 KOps/s	6.8646 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_indexing[slice-pytree-eager]	0.1278ms	24.6481μs	40.5711 KOps/s	47.9341 KOps/s	$\textbf{\color{#d91a1a}-15.36\%}$
test_compile_indexing[int-tensordict-compile]	0.1984ms	0.1535ms	6.5151 KOps/s	6.4999 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_indexing[int-tensordict-eager]	0.5683ms	26.1827μs	38.1932 KOps/s	32.4641 KOps/s	$\textbf{\color{#35bf28}+17.65\%}$
test_compile_indexing[int-tensorclass-compile]	0.1944ms	0.1464ms	6.8307 KOps/s	6.8438 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_indexing[int-tensorclass-eager]	54.6510μs	21.1969μs	47.1768 KOps/s	47.1297 KOps/s	$\color{#35bf28}+0.10\%$
test_compile_indexing[int-pytree-compile]	0.1945ms	0.1490ms	6.7133 KOps/s	6.8438 KOps/s	$\color{#d91a1a}-1.91\%$
test_compile_indexing[int-pytree-eager]	60.0110μs	21.4855μs	46.5430 KOps/s	47.4074 KOps/s	$\color{#d91a1a}-1.82\%$
test_mod_add[eager]	83.0420μs	33.8165μs	29.5713 KOps/s	28.7713 KOps/s	$\color{#35bf28}+2.78\%$
test_mod_add[compile]	0.2825ms	77.4850μs	12.9057 KOps/s	12.6677 KOps/s	$\color{#35bf28}+1.88\%$
test_mod_add[compile-overhead]	0.3025ms	0.1564ms	6.3925 KOps/s	5.6211 KOps/s	$\textbf{\color{#35bf28}+13.72\%}$
test_mod_wrap[eager]	0.3236ms	0.2535ms	3.9452 KOps/s	3.9015 KOps/s	$\color{#35bf28}+1.12\%$
test_mod_wrap[compile]	0.3763ms	0.2852ms	3.5057 KOps/s	3.4293 KOps/s	$\color{#35bf28}+2.23\%$
test_mod_wrap[compile-overhead]	7.8053ms	4.1170ms	242.8926 Ops/s	243.9602 Ops/s	$\color{#d91a1a}-0.44\%$
test_mod_wrap_and_backward[eager]	1.4827ms	1.3602ms	735.1863 Ops/s	673.9696 Ops/s	$\textbf{\color{#35bf28}+9.08\%}$
test_mod_wrap_and_backward[compile]	1.5265ms	1.2809ms	780.7154 Ops/s	716.3165 Ops/s	$\textbf{\color{#35bf28}+8.99\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3663ms	0.9225ms	1.0840 KOps/s	956.6337 Ops/s	$\textbf{\color{#35bf28}+13.31\%}$
test_seq_add[eager]	0.1470ms	0.1021ms	9.7914 KOps/s	9.7835 KOps/s	$\color{#35bf28}+0.08\%$
test_seq_add[compile]	0.2342ms	88.4529μs	11.3055 KOps/s	11.2840 KOps/s	$\color{#35bf28}+0.19\%$
test_seq_add[compile-overhead]	0.1667ms	0.1257ms	7.9568 KOps/s	7.8065 KOps/s	$\color{#35bf28}+1.93\%$
test_seq_wrap[eager]	0.4494ms	0.3829ms	2.6117 KOps/s	2.4752 KOps/s	$\textbf{\color{#35bf28}+5.51\%}$
test_seq_wrap[compile]	0.4894ms	0.3098ms	3.2279 KOps/s	3.1588 KOps/s	$\color{#35bf28}+2.19\%$
test_seq_wrap[compile-overhead]	0.2744ms	0.2224ms	4.4969 KOps/s	4.4192 KOps/s	$\color{#35bf28}+1.76\%$
test_func_call_runtime[False-eager]	0.8634ms	0.7425ms	1.3468 KOps/s	1.2861 KOps/s	$\color{#35bf28}+4.72\%$
test_func_call_runtime[False-compile]	1.1018ms	0.7717ms	1.2958 KOps/s	1.2975 KOps/s	$\color{#d91a1a}-0.13\%$
test_func_call_runtime[False-compile-overhead]	0.4113ms	0.3644ms	2.7443 KOps/s	2.7321 KOps/s	$\color{#35bf28}+0.45\%$
test_func_call_runtime[True-eager]	1.1035ms	0.8974ms	1.1143 KOps/s	1.0833 KOps/s	$\color{#35bf28}+2.87\%$
test_func_call_runtime[True-compile]	1.0119ms	0.7908ms	1.2645 KOps/s	1.2623 KOps/s	$\color{#35bf28}+0.17\%$
test_func_call_runtime[True-compile-overhead]	0.4733ms	0.3833ms	2.6088 KOps/s	2.5992 KOps/s	$\color{#35bf28}+0.37\%$
test_func_call_cm_runtime[False-eager]	0.8424ms	0.7285ms	1.3726 KOps/s	1.3090 KOps/s	$\color{#35bf28}+4.86\%$
test_func_call_cm_runtime[False-compile]	0.8160ms	0.7691ms	1.3002 KOps/s	1.2844 KOps/s	$\color{#35bf28}+1.23\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5080ms	0.3709ms	2.6959 KOps/s	2.7278 KOps/s	$\color{#d91a1a}-1.17\%$
test_func_call_cm_runtime[True-eager]	1.1184ms	0.9940ms	1.0061 KOps/s	977.6972 Ops/s	$\color{#35bf28}+2.90\%$
test_func_call_cm_runtime[True-compile]	0.9058ms	0.8197ms	1.2200 KOps/s	1.2168 KOps/s	$\color{#35bf28}+0.26\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5186ms	0.4118ms	2.4287 KOps/s	2.4152 KOps/s	$\color{#35bf28}+0.56\%$
test_vmap_func_call_cm_runtime[eager]	2.5283ms	2.0738ms	482.2042 Ops/s	470.2081 Ops/s	$\color{#35bf28}+2.55\%$
test_vmap_func_call_cm_runtime[compile]	0.9328ms	0.8311ms	1.2032 KOps/s	1.1849 KOps/s	$\color{#35bf28}+1.54\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4813ms	0.4160ms	2.4036 KOps/s	2.4041 KOps/s	$\color{#d91a1a}-0.02\%$
test_distributed	2.2664ms	0.1737ms	5.7568 KOps/s	8.7754 KOps/s	$\textbf{\color{#d91a1a}-34.40\%}$
test_tdmodule	25.0900μs	14.9257μs	66.9987 KOps/s	58.9057 KOps/s	$\textbf{\color{#35bf28}+13.74\%}$
test_tdmodule_dispatch	51.2110μs	29.2049μs	34.2409 KOps/s	31.6014 KOps/s	$\textbf{\color{#35bf28}+8.35\%}$
test_tdseq	37.4100μs	16.1632μs	61.8688 KOps/s	57.5410 KOps/s	$\textbf{\color{#35bf28}+7.52\%}$
test_tdseq_dispatch	53.0110μs	32.6724μs	30.6069 KOps/s	28.4447 KOps/s	$\textbf{\color{#35bf28}+7.60\%}$
test_instantiation_functorch	2.0469ms	1.8977ms	526.9472 Ops/s	521.5826 Ops/s	$\color{#35bf28}+1.03\%$
test_exec_functorch	0.3007ms	0.2083ms	4.8010 KOps/s	4.6369 KOps/s	$\color{#35bf28}+3.54\%$
test_exec_functional_call	0.2523ms	0.2121ms	4.7152 KOps/s	4.6145 KOps/s	$\color{#35bf28}+2.18\%$
test_exec_td_decorator	0.4299ms	0.2568ms	3.8944 KOps/s	3.7902 KOps/s	$\color{#35bf28}+2.75\%$
test_vmap_mlp_speed_decorator[True-True]	0.8156ms	0.6692ms	1.4944 KOps/s	1.4626 KOps/s	$\color{#35bf28}+2.17\%$
test_vmap_mlp_speed_decorator[True-False]	0.7828ms	0.6722ms	1.4877 KOps/s	1.4699 KOps/s	$\color{#35bf28}+1.21\%$
test_vmap_mlp_speed_decorator[False-True]	0.7028ms	0.5874ms	1.7025 KOps/s	1.6784 KOps/s	$\color{#35bf28}+1.44\%$
test_vmap_mlp_speed_decorator[False-False]	0.6882ms	0.5875ms	1.7020 KOps/s	1.6132 KOps/s	$\textbf{\color{#35bf28}+5.51\%}$
test_vmap_transformer_speed_decorator[True-True]	19.6594ms	19.5232ms	51.2211 Ops/s	50.7545 Ops/s	$\color{#35bf28}+0.92\%$
test_vmap_transformer_speed_decorator[True-False]	20.1779ms	19.5744ms	51.0871 Ops/s	50.3599 Ops/s	$\color{#35bf28}+1.44\%$
test_vmap_transformer_speed_decorator[False-True]	19.5110ms	19.4055ms	51.5318 Ops/s	51.0769 Ops/s	$\color{#35bf28}+0.89\%$
test_vmap_transformer_speed_decorator[False-False]	19.5935ms	19.3942ms	51.5618 Ops/s	51.1667 Ops/s	$\color{#35bf28}+0.77\%$
test_to_module_speed[True]	1.3669ms	0.9196ms	1.0874 KOps/s	1.0723 KOps/s	$\color{#35bf28}+1.41\%$
test_to_module_speed[False]	1.3077ms	0.8994ms	1.1119 KOps/s	1.0843 KOps/s	$\color{#35bf28}+2.54\%$
test_tc_init	61.8010μs	35.7694μs	27.9569 KOps/s	28.3168 KOps/s	$\color{#d91a1a}-1.27\%$
test_tc_init_nested	0.1140ms	71.6895μs	13.9491 KOps/s	14.0528 KOps/s	$\color{#d91a1a}-0.74\%$
test_tc_first_layer_tensor	3.9343μs	0.6893μs	1.4507 MOps/s	1.4227 MOps/s	$\color{#35bf28}+1.97\%$
test_tc_first_layer_nontensor	29.8810μs	2.3190μs	431.2116 KOps/s	433.0428 KOps/s	$\color{#d91a1a}-0.42\%$
test_tc_second_layer_tensor	16.7253μs	1.4086μs	709.9246 KOps/s	701.0436 KOps/s	$\color{#35bf28}+1.27\%$
test_tc_second_layer_nontensor	27.2910μs	3.0111μs	332.1095 KOps/s	331.2535 KOps/s	$\color{#35bf28}+0.26\%$
test_unbind	0.1910s	11.7028ms	85.4500 Ops/s	104.5206 Ops/s	$\textbf{\color{#d91a1a}-18.25\%}$
test_full_like	0.6591ms	0.5759ms	1.7363 KOps/s	1.7455 KOps/s	$\color{#d91a1a}-0.53\%$
test_zeros_like	0.2599ms	0.1979ms	5.0537 KOps/s	5.0517 KOps/s	$\color{#35bf28}+0.04\%$
test_ones_like	0.3411ms	0.1978ms	5.0564 KOps/s	5.0562 KOps/s	$+0.00\%$
test_clone	0.4692ms	0.4148ms	2.4105 KOps/s	2.4107 KOps/s	$-0.01\%$
test_squeeze	34.1510μs	9.3002μs	107.5251 KOps/s	107.0335 KOps/s	$\color{#35bf28}+0.46\%$
test_unsqueeze	0.2061ms	69.3604μs	14.4174 KOps/s	13.6399 KOps/s	$\textbf{\color{#35bf28}+5.70\%}$
test_split	0.3922ms	0.1620ms	6.1718 KOps/s	6.0536 KOps/s	$\color{#35bf28}+1.95\%$
test_permute	0.2299ms	0.1761ms	5.6770 KOps/s	5.5349 KOps/s	$\color{#35bf28}+2.57\%$
test_stack	1.2473ms	0.8819ms	1.1339 KOps/s	1.1601 KOps/s	$\color{#d91a1a}-2.26\%$
test_cat	1.2637ms	1.2315ms	812.0447 Ops/s	811.9639 Ops/s	$+0.01\%$