Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature] TensorDict.tolist() #1229

Merged
merged 1 commit into from
Feb 21, 2025
Merged

[Feature] TensorDict.tolist() #1229

merged 1 commit into from
Feb 21, 2025

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 21, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 21, 2025
ghstack-source-id: 81f7fe4c26628b5b7fc738f7f9c09a10deb87e94
Pull Request resolved: #1229
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 21, 2025
@vmoens vmoens merged commit e6b8dfb into gh/vmoens/48/base Feb 21, 2025
36 checks passed
vmoens added a commit that referenced this pull request Feb 21, 2025
"$(git log -1 --format=%b origin/gh/vmoens/48/orig)"
@vmoens vmoens deleted the gh/vmoens/48/head branch February 21, 2025 13:30
vmoens added a commit that referenced this pull request Feb 21, 2025
Copy link

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}21$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 43.0900μs 20.5139μs 48.7474 KOps/s 47.7119 KOps/s $\color{#35bf28}+2.17\%$
test_plain_set_stack_nested 51.4060μs 20.6341μs 48.4634 KOps/s 48.4366 KOps/s $\color{#35bf28}+0.06\%$
test_plain_set_nested_inplace 73.2570μs 22.5122μs 44.4203 KOps/s 44.0450 KOps/s $\color{#35bf28}+0.85\%$
test_plain_set_stack_nested_inplace 86.5220μs 22.3741μs 44.6945 KOps/s 44.3501 KOps/s $\color{#35bf28}+0.78\%$
test_items 32.1700μs 4.1947μs 238.3977 KOps/s 234.4762 KOps/s $\color{#35bf28}+1.67\%$
test_items_nested 0.5056ms 0.4051ms 2.4686 KOps/s 2.4831 KOps/s $\color{#d91a1a}-0.58\%$
test_items_nested_locked 0.4927ms 0.4044ms 2.4730 KOps/s 2.4878 KOps/s $\color{#d91a1a}-0.59\%$
test_items_nested_leaf 0.1600ms 75.7799μs 13.1961 KOps/s 13.0206 KOps/s $\color{#35bf28}+1.35\%$
test_items_stack_nested 0.7088ms 0.4082ms 2.4495 KOps/s 2.4404 KOps/s $\color{#35bf28}+0.37\%$
test_items_stack_nested_leaf 0.1357ms 79.2184μs 12.6233 KOps/s 12.5714 KOps/s $\color{#35bf28}+0.41\%$
test_items_stack_nested_locked 0.7455ms 0.4085ms 2.4478 KOps/s 2.4711 KOps/s $\color{#d91a1a}-0.94\%$
test_keys 31.7790μs 3.4322μs 291.3542 KOps/s 285.5378 KOps/s $\color{#35bf28}+2.04\%$
test_keys_nested 0.2244ms 0.1647ms 6.0719 KOps/s 5.9936 KOps/s $\color{#35bf28}+1.31\%$
test_keys_nested_locked 0.7119ms 0.1711ms 5.8434 KOps/s 5.8219 KOps/s $\color{#35bf28}+0.37\%$
test_keys_nested_leaf 0.2812ms 0.1439ms 6.9505 KOps/s 6.8755 KOps/s $\color{#35bf28}+1.09\%$
test_keys_stack_nested 0.2840ms 0.1629ms 6.1379 KOps/s 5.9892 KOps/s $\color{#35bf28}+2.48\%$
test_keys_stack_nested_leaf 0.2589ms 0.1408ms 7.1006 KOps/s 6.8667 KOps/s $\color{#35bf28}+3.41\%$
test_keys_stack_nested_locked 0.2652ms 0.1697ms 5.8928 KOps/s 5.7974 KOps/s $\color{#35bf28}+1.64\%$
test_values 9.2334μs 1.0372μs 964.1170 KOps/s 915.8121 KOps/s $\textbf{\color{#35bf28}+5.27\%}$
test_values_nested 0.1403ms 62.3614μs 16.0356 KOps/s 15.6862 KOps/s $\color{#35bf28}+2.23\%$
test_values_nested_locked 0.2025ms 62.0722μs 16.1103 KOps/s 16.0497 KOps/s $\color{#35bf28}+0.38\%$
test_values_nested_leaf 0.1362ms 71.2983μs 14.0256 KOps/s 13.4086 KOps/s $\color{#35bf28}+4.60\%$
test_values_stack_nested 0.1305ms 63.8698μs 15.6568 KOps/s 15.9779 KOps/s $\color{#d91a1a}-2.01\%$
test_values_stack_nested_leaf 0.1569ms 70.3452μs 14.2156 KOps/s 14.0325 KOps/s $\color{#35bf28}+1.30\%$
test_values_stack_nested_locked 0.1340ms 63.6383μs 15.7138 KOps/s 15.9561 KOps/s $\color{#d91a1a}-1.52\%$
test_membership 14.4570μs 0.8436μs 1.1855 MOps/s 1.0969 MOps/s $\textbf{\color{#35bf28}+8.07\%}$
test_membership_nested 47.7090μs 2.9295μs 341.3583 KOps/s 340.5671 KOps/s $\color{#35bf28}+0.23\%$
test_membership_nested_leaf 28.8440μs 2.9260μs 341.7609 KOps/s 342.0937 KOps/s $\color{#d91a1a}-0.10\%$
test_membership_stacked_nested 46.6670μs 2.9048μs 344.2628 KOps/s 342.7079 KOps/s $\color{#35bf28}+0.45\%$
test_membership_stacked_nested_leaf 30.2660μs 2.9355μs 340.6535 KOps/s 344.5154 KOps/s $\color{#d91a1a}-1.12\%$
test_membership_nested_last 46.8570μs 4.3277μs 231.0711 KOps/s 225.0887 KOps/s $\color{#35bf28}+2.66\%$
test_membership_nested_leaf_last 22.3220μs 4.3542μs 229.6652 KOps/s 223.4677 KOps/s $\color{#35bf28}+2.77\%$
test_membership_stacked_nested_last 33.5530μs 4.3273μs 231.0935 KOps/s 223.0402 KOps/s $\color{#35bf28}+3.61\%$
test_membership_stacked_nested_leaf_last 34.5450μs 4.3606μs 229.3282 KOps/s 223.0053 KOps/s $\color{#35bf28}+2.84\%$
test_nested_getleaf 62.5870μs 10.7972μs 92.6164 KOps/s 93.9008 KOps/s $\color{#d91a1a}-1.37\%$
test_nested_get 39.6740μs 10.2205μs 97.8428 KOps/s 99.2400 KOps/s $\color{#d91a1a}-1.41\%$
test_stacked_getleaf 46.0860μs 10.6843μs 93.5955 KOps/s 96.0111 KOps/s $\color{#d91a1a}-2.52\%$
test_stacked_get 65.7950μs 10.1447μs 98.5741 KOps/s 99.4892 KOps/s $\color{#d91a1a}-0.92\%$
test_nested_getitemleaf 34.7250μs 11.1537μs 89.6560 KOps/s 89.6341 KOps/s $\color{#35bf28}+0.02\%$
test_nested_getitem 35.8770μs 10.6218μs 94.1458 KOps/s 93.2901 KOps/s $\color{#35bf28}+0.92\%$
test_stacked_getitemleaf 52.0870μs 11.1332μs 89.8215 KOps/s 88.6456 KOps/s $\color{#35bf28}+1.33\%$
test_stacked_getitem 37.7510μs 10.6435μs 93.9539 KOps/s 93.8293 KOps/s $\color{#35bf28}+0.13\%$
test_lock_nested 0.5456ms 0.4059ms 2.4637 KOps/s 2.4411 KOps/s $\color{#35bf28}+0.93\%$
test_lock_stack_nested 0.7389ms 0.4159ms 2.4047 KOps/s 2.3486 KOps/s $\color{#35bf28}+2.39\%$
test_unlock_nested 0.5949ms 0.3293ms 3.0369 KOps/s 2.9460 KOps/s $\color{#35bf28}+3.08\%$
test_unlock_stack_nested 0.5372ms 0.3365ms 2.9718 KOps/s 2.8678 KOps/s $\color{#35bf28}+3.63\%$
test_flatten_speed 0.2484ms 99.3820μs 10.0622 KOps/s 9.7066 KOps/s $\color{#35bf28}+3.66\%$
test_unflatten_speed 0.7261ms 0.5257ms 1.9024 KOps/s 1.8978 KOps/s $\color{#35bf28}+0.24\%$
test_common_ops 4.4867ms 0.8465ms 1.1813 KOps/s 1.2136 KOps/s $\color{#d91a1a}-2.66\%$
test_creation 65.5220μs 2.5071μs 398.8632 KOps/s 400.2693 KOps/s $\color{#d91a1a}-0.35\%$
test_creation_empty 40.6260μs 11.9807μs 83.4672 KOps/s 84.1618 KOps/s $\color{#d91a1a}-0.83\%$
test_creation_nested_1 44.2520μs 14.8814μs 67.1981 KOps/s 68.0328 KOps/s $\color{#d91a1a}-1.23\%$
test_creation_nested_2 61.1240μs 19.4347μs 51.4543 KOps/s 51.8743 KOps/s $\color{#d91a1a}-0.81\%$
test_clone 0.1102ms 13.3642μs 74.8270 KOps/s 75.1658 KOps/s $\color{#d91a1a}-0.45\%$
test_getitem[int] 0.7923ms 12.5114μs 79.9271 KOps/s 78.1541 KOps/s $\color{#35bf28}+2.27\%$
test_getitem[slice_int] 0.1249ms 23.9800μs 41.7013 KOps/s 40.2540 KOps/s $\color{#35bf28}+3.60\%$
test_getitem[range] 0.1590ms 50.4231μs 19.8322 KOps/s 19.0574 KOps/s $\color{#35bf28}+4.07\%$
test_getitem[tuple] 0.1350ms 19.9003μs 50.2505 KOps/s 49.2143 KOps/s $\color{#35bf28}+2.11\%$
test_getitem[list] 0.1648ms 45.3747μs 22.0387 KOps/s 21.3026 KOps/s $\color{#35bf28}+3.46\%$
test_setitem_dim[int] 59.1500μs 25.3378μs 39.4667 KOps/s 38.7238 KOps/s $\color{#35bf28}+1.92\%$
test_setitem_dim[slice_int] 0.1140ms 51.5881μs 19.3843 KOps/s 19.5384 KOps/s $\color{#d91a1a}-0.79\%$
test_setitem_dim[range] 0.1483ms 76.5727μs 13.0595 KOps/s 12.8006 KOps/s $\color{#35bf28}+2.02\%$
test_setitem_dim[tuple] 91.6910μs 40.7250μs 24.5550 KOps/s 24.1638 KOps/s $\color{#35bf28}+1.62\%$
test_setitem 0.1058ms 20.3146μs 49.2257 KOps/s 48.0153 KOps/s $\color{#35bf28}+2.52\%$
test_set 0.1906ms 19.8185μs 50.4580 KOps/s 49.7044 KOps/s $\color{#35bf28}+1.52\%$
test_set_shared 0.3927ms 0.1823ms 5.4857 KOps/s 5.3419 KOps/s $\color{#35bf28}+2.69\%$
test_update 0.1903ms 22.8580μs 43.7483 KOps/s 43.2543 KOps/s $\color{#35bf28}+1.14\%$
test_update_nested 0.2078ms 33.7871μs 29.5971 KOps/s 29.7790 KOps/s $\color{#d91a1a}-0.61\%$
test_update__nested 0.5729ms 33.7790μs 29.6042 KOps/s 28.7896 KOps/s $\color{#35bf28}+2.83\%$
test_set_nested 85.2190μs 22.0304μs 45.3917 KOps/s 44.1772 KOps/s $\color{#35bf28}+2.75\%$
test_set_nested_new 71.4840μs 27.0502μs 36.9683 KOps/s 37.0850 KOps/s $\color{#d91a1a}-0.31\%$
test_select 98.4840μs 43.8823μs 22.7883 KOps/s 23.1609 KOps/s $\color{#d91a1a}-1.61\%$
test_select_nested 0.1199ms 62.2775μs 16.0572 KOps/s 15.6016 KOps/s $\color{#35bf28}+2.92\%$
test_exclude_nested 0.3954ms 81.2185μs 12.3125 KOps/s 12.4060 KOps/s $\color{#d91a1a}-0.75\%$
test_empty[True] 0.5583ms 0.4157ms 2.4057 KOps/s 2.4413 KOps/s $\color{#d91a1a}-1.46\%$
test_empty[False] 11.2985μs 1.3688μs 730.5629 KOps/s 716.8171 KOps/s $\color{#35bf28}+1.92\%$
test_unbind_speed 0.3066ms 0.2642ms 3.7855 KOps/s 3.7045 KOps/s $\color{#35bf28}+2.19\%$
test_unbind_speed_stack0 0.4320ms 0.2596ms 3.8524 KOps/s 3.6750 KOps/s $\color{#35bf28}+4.83\%$
test_unbind_speed_stack1 0.1130s 0.7140ms 1.4005 KOps/s 1.1912 KOps/s $\textbf{\color{#35bf28}+17.57\%}$
test_split 0.1118s 1.7334ms 576.9011 Ops/s 631.6566 Ops/s $\textbf{\color{#d91a1a}-8.67\%}$
test_chunk 0.1182s 1.7447ms 573.1536 Ops/s 562.3844 Ops/s $\color{#35bf28}+1.91\%$
test_consolidate_njt[False-None] 8.7619ms 8.3457ms 119.8225 Ops/s 106.8119 Ops/s $\textbf{\color{#35bf28}+12.18\%}$
test_creation[device0] 4.6684ms 94.7574μs 10.5533 KOps/s 10.6992 KOps/s $\color{#d91a1a}-1.36\%$
test_creation_from_tensor 0.3119ms 98.5806μs 10.1440 KOps/s 10.1987 KOps/s $\color{#d91a1a}-0.54\%$
test_add_one[memmap_tensor0] 0.1598ms 4.9737μs 201.0591 KOps/s 201.0079 KOps/s $\color{#35bf28}+0.03\%$
test_contiguous[memmap_tensor0] 14.0060μs 0.4993μs 2.0030 MOps/s 1.8852 MOps/s $\textbf{\color{#35bf28}+6.25\%}$
test_stack[memmap_tensor0] 58.0590μs 3.3954μs 294.5160 KOps/s 290.7922 KOps/s $\color{#35bf28}+1.28\%$
test_memmaptd_index 1.2712ms 0.2287ms 4.3720 KOps/s 4.2984 KOps/s $\color{#35bf28}+1.71\%$
test_memmaptd_index_astensor 0.4880ms 0.3157ms 3.1676 KOps/s 3.1093 KOps/s $\color{#35bf28}+1.87\%$
test_memmaptd_index_op 0.7951ms 0.5786ms 1.7283 KOps/s 1.7188 KOps/s $\color{#35bf28}+0.55\%$
test_serialize_model 0.2380s 0.1323s 7.5599 Ops/s 8.4817 Ops/s $\textbf{\color{#d91a1a}-10.87\%}$
test_serialize_model_pickle 0.4812s 0.3910s 2.5575 Ops/s 2.5323 Ops/s $\color{#35bf28}+1.00\%$
test_serialize_weights 0.1287s 0.1174s 8.5158 Ops/s 8.6795 Ops/s $\color{#d91a1a}-1.89\%$
test_serialize_weights_returnearly 0.1799s 0.1624s 6.1574 Ops/s 6.2025 Ops/s $\color{#d91a1a}-0.73\%$
test_serialize_weights_pickle 0.5753s 0.4406s 2.2697 Ops/s 2.5020 Ops/s $\textbf{\color{#d91a1a}-9.28\%}$
test_serialize_weights_filesystem 0.1520s 0.1476s 6.7737 Ops/s 6.8983 Ops/s $\color{#d91a1a}-1.81\%$
test_serialize_model_filesystem 0.1562s 0.1455s 6.8741 Ops/s 6.4798 Ops/s $\textbf{\color{#35bf28}+6.09\%}$
test_reshape_pytree 61.0430μs 25.8221μs 38.7265 KOps/s 37.7650 KOps/s $\color{#35bf28}+2.55\%$
test_reshape_td 92.9830μs 33.0268μs 30.2784 KOps/s 30.1720 KOps/s $\color{#35bf28}+0.35\%$
test_view_pytree 83.0850μs 26.0004μs 38.4610 KOps/s 38.3864 KOps/s $\color{#35bf28}+0.19\%$
test_view_td 0.1015ms 39.5958μs 25.2552 KOps/s 24.6958 KOps/s $\color{#35bf28}+2.27\%$
test_unbind_pytree 88.7350μs 29.2041μs 34.2418 KOps/s 33.5336 KOps/s $\color{#35bf28}+2.11\%$
test_unbind_td 0.3798ms 39.5285μs 25.2982 KOps/s 25.0529 KOps/s $\color{#35bf28}+0.98\%$
test_split_pytree 75.4810μs 28.6431μs 34.9124 KOps/s 34.3292 KOps/s $\color{#35bf28}+1.70\%$
test_split_td 0.2389ms 44.7523μs 22.3452 KOps/s 21.8281 KOps/s $\color{#35bf28}+2.37\%$
test_add_pytree 95.6190μs 35.6216μs 28.0729 KOps/s 28.0642 KOps/s $\color{#35bf28}+0.03\%$
test_add_td 0.1534ms 55.3177μs 18.0774 KOps/s 16.5103 KOps/s $\textbf{\color{#35bf28}+9.49\%}$
test_compile_add_one_nested[tensordict-compile] 0.1475ms 67.2074μs 14.8793 KOps/s 14.3497 KOps/s $\color{#35bf28}+3.69\%$
test_compile_add_one_nested[tensordict-eager] 0.4469ms 0.1745ms 5.7314 KOps/s 5.8014 KOps/s $\color{#d91a1a}-1.21\%$
test_compile_add_one_nested[pytree-compile] 0.1110ms 45.4200μs 22.0167 KOps/s 20.9861 KOps/s $\color{#35bf28}+4.91\%$
test_compile_add_one_nested[pytree-eager] 0.2648ms 0.1171ms 8.5415 KOps/s 8.3710 KOps/s $\color{#35bf28}+2.04\%$
test_compile_copy_nested[tensordict-compile] 83.8970μs 28.8045μs 34.7168 KOps/s 33.6916 KOps/s $\color{#35bf28}+3.04\%$
test_compile_copy_nested[tensordict-eager] 0.1215ms 57.4607μs 17.4032 KOps/s 17.1537 KOps/s $\color{#35bf28}+1.45\%$
test_compile_copy_nested[pytree-compile] 0.1476ms 79.5870μs 12.5649 KOps/s 12.3131 KOps/s $\color{#35bf28}+2.04\%$
test_compile_copy_nested[pytree-eager] 0.1411ms 65.8851μs 15.1779 KOps/s 14.8690 KOps/s $\color{#35bf28}+2.08\%$
test_compile_add_one_flat[tensordict-compile] 0.2013ms 0.1074ms 9.3110 KOps/s 9.2066 KOps/s $\color{#35bf28}+1.13\%$
test_compile_add_one_flat[tensordict-eager] 0.3841ms 0.2175ms 4.5967 KOps/s 4.5908 KOps/s $\color{#35bf28}+0.13\%$
test_compile_add_one_flat[tensorclass-compile] 0.1098ms 47.2630μs 21.1582 KOps/s 20.4896 KOps/s $\color{#35bf28}+3.26\%$
test_compile_add_one_flat[tensorclass-eager] 0.2348ms 67.3148μs 14.8556 KOps/s 14.7711 KOps/s $\color{#35bf28}+0.57\%$
test_compile_add_one_flat[pytree-compile] 0.1719ms 0.1004ms 9.9567 KOps/s 9.8304 KOps/s $\color{#35bf28}+1.28\%$
test_compile_add_one_flat[pytree-eager] 0.4158ms 0.2047ms 4.8859 KOps/s 5.0242 KOps/s $\color{#d91a1a}-2.75\%$
test_compile_add_self_flat[tensordict-eager] 0.4204ms 0.2351ms 4.2526 KOps/s 4.2553 KOps/s $\color{#d91a1a}-0.06\%$
test_compile_add_self_flat[tensordict-compile] 0.2593ms 0.1098ms 9.1074 KOps/s 9.0841 KOps/s $\color{#35bf28}+0.26\%$
test_compile_add_self_flat[tensorclass-eager] 0.1521ms 63.0437μs 15.8620 KOps/s 15.8473 KOps/s $\color{#35bf28}+0.09\%$
test_compile_add_self_flat[tensorclass-compile] 0.1379ms 49.4000μs 20.2429 KOps/s 19.1017 KOps/s $\textbf{\color{#35bf28}+5.97\%}$
test_compile_add_self_flat[pytree-eager] 0.2806ms 0.1584ms 6.3146 KOps/s 6.3950 KOps/s $\color{#d91a1a}-1.26\%$
test_compile_add_self_flat[pytree-compile] 0.2080ms 0.1019ms 9.8104 KOps/s 9.7233 KOps/s $\color{#35bf28}+0.90\%$
test_compile_copy_flat[tensordict-compile] 74.9300μs 21.8050μs 45.8611 KOps/s 44.7411 KOps/s $\color{#35bf28}+2.50\%$
test_compile_copy_flat[tensordict-eager] 0.1291ms 66.2216μs 15.1008 KOps/s 15.1411 KOps/s $\color{#d91a1a}-0.27\%$
test_compile_copy_flat[pytree-compile] 0.1594ms 83.4020μs 11.9901 KOps/s 12.0985 KOps/s $\color{#d91a1a}-0.90\%$
test_compile_copy_flat[pytree-eager] 0.1277ms 67.0061μs 14.9240 KOps/s 14.7174 KOps/s $\color{#35bf28}+1.40\%$
test_compile_assign_and_add[tensordict-compile] 0.3056ms 0.2161ms 4.6280 KOps/s 4.5900 KOps/s $\color{#35bf28}+0.83\%$
test_compile_assign_and_add[tensordict-eager] 2.3290ms 1.3954ms 716.6626 Ops/s 704.7332 Ops/s $\color{#35bf28}+1.69\%$
test_compile_assign_and_add[pytree-compile] 0.2910ms 0.2067ms 4.8373 KOps/s 4.7243 KOps/s $\color{#35bf28}+2.39\%$
test_compile_assign_and_add[pytree-eager] 1.0329ms 0.8147ms 1.2274 KOps/s 1.2077 KOps/s $\color{#35bf28}+1.64\%$
test_compile_assign_and_add_stack[compile] 0.5983ms 0.4512ms 2.2162 KOps/s 2.1766 KOps/s $\color{#35bf28}+1.82\%$
test_compile_assign_and_add_stack[eager] 3.4004ms 2.7334ms 365.8440 Ops/s 369.6227 Ops/s $\color{#d91a1a}-1.02\%$
test_compile_indexing[tensor-tensordict-compile] 0.1064ms 38.6020μs 25.9054 KOps/s 24.6883 KOps/s $\color{#35bf28}+4.93\%$
test_compile_indexing[tensor-tensordict-eager] 0.6713ms 32.5433μs 30.7283 KOps/s 30.2212 KOps/s $\color{#35bf28}+1.68\%$
test_compile_indexing[tensor-tensorclass-compile] 0.1144ms 30.2401μs 33.0686 KOps/s 31.1357 KOps/s $\textbf{\color{#35bf28}+6.21\%}$
test_compile_indexing[tensor-tensorclass-eager] 0.1050ms 23.2985μs 42.9213 KOps/s 44.1824 KOps/s $\color{#d91a1a}-2.85\%$
test_compile_indexing[tensor-pytree-compile] 0.1080ms 31.1918μs 32.0597 KOps/s 30.6782 KOps/s $\color{#35bf28}+4.50\%$
test_compile_indexing[tensor-pytree-eager] 0.1417ms 23.5653μs 42.4353 KOps/s 43.6011 KOps/s $\color{#d91a1a}-2.67\%$
test_compile_indexing[slice-tensordict-compile] 0.1737ms 53.3991μs 18.7269 KOps/s 18.2791 KOps/s $\color{#35bf28}+2.45\%$
test_compile_indexing[slice-tensordict-eager] 0.3825ms 19.7324μs 50.6782 KOps/s 48.8373 KOps/s $\color{#35bf28}+3.77\%$
test_compile_indexing[slice-tensorclass-compile] 0.1650ms 44.5417μs 22.4509 KOps/s 21.3766 KOps/s $\textbf{\color{#35bf28}+5.03\%}$
test_compile_indexing[slice-tensorclass-eager] 76.4300μs 18.2480μs 54.8004 KOps/s 53.3619 KOps/s $\color{#35bf28}+2.70\%$
test_compile_indexing[slice-pytree-compile] 0.1139ms 46.0232μs 21.7282 KOps/s 20.7465 KOps/s $\color{#35bf28}+4.73\%$
test_compile_indexing[slice-pytree-eager] 53.1290μs 18.2785μs 54.7092 KOps/s 53.2639 KOps/s $\color{#35bf28}+2.71\%$
test_compile_indexing[int-tensordict-compile] 0.1255ms 54.2280μs 18.4407 KOps/s 17.8112 KOps/s $\color{#35bf28}+3.53\%$
test_compile_indexing[int-tensordict-eager] 1.0830ms 19.4972μs 51.2894 KOps/s 48.8137 KOps/s $\textbf{\color{#35bf28}+5.07\%}$
test_compile_indexing[int-tensorclass-compile] 0.1212ms 46.0280μs 21.7259 KOps/s 20.8372 KOps/s $\color{#35bf28}+4.27\%$
test_compile_indexing[int-tensorclass-eager] 69.4900μs 18.3027μs 54.6366 KOps/s 52.9897 KOps/s $\color{#35bf28}+3.11\%$
test_compile_indexing[int-pytree-compile] 0.1117ms 45.6635μs 21.8993 KOps/s 20.7763 KOps/s $\textbf{\color{#35bf28}+5.41\%}$
test_compile_indexing[int-pytree-eager] 64.0290μs 18.2794μs 54.7064 KOps/s 53.6552 KOps/s $\color{#35bf28}+1.96\%$
test_mod_add[eager] 98.4940μs 35.2414μs 28.3757 KOps/s 26.9849 KOps/s $\textbf{\color{#35bf28}+5.15\%}$
test_mod_add[compile] 0.1836ms 66.5902μs 15.0172 KOps/s 14.5832 KOps/s $\color{#35bf28}+2.98\%$
test_mod_add[compile-overhead] 0.1300ms 64.9539μs 15.3955 KOps/s 14.3893 KOps/s $\textbf{\color{#35bf28}+6.99\%}$
test_mod_wrap[eager] 0.3355ms 0.2242ms 4.4603 KOps/s 4.4062 KOps/s $\color{#35bf28}+1.23\%$
test_mod_wrap[compile] 2.1302ms 0.2293ms 4.3619 KOps/s 4.1819 KOps/s $\color{#35bf28}+4.30\%$
test_mod_wrap[compile-overhead] 0.3692ms 0.2244ms 4.4567 KOps/s 4.3226 KOps/s $\color{#35bf28}+3.10\%$
test_mod_wrap_and_backward[eager] 12.5393ms 11.0707ms 90.3287 Ops/s 72.9216 Ops/s $\textbf{\color{#35bf28}+23.87\%}$
test_mod_wrap_and_backward[compile] 12.7481ms 11.1015ms 90.0781 Ops/s 85.6028 Ops/s $\textbf{\color{#35bf28}+5.23\%}$
test_mod_wrap_and_backward[compile-overhead] 12.7026ms 11.1417ms 89.7527 Ops/s 88.2031 Ops/s $\color{#35bf28}+1.76\%$
test_seq_add[eager] 0.2011ms 0.1179ms 8.4810 KOps/s 8.0912 KOps/s $\color{#35bf28}+4.82\%$
test_seq_add[compile] 0.1742ms 79.9048μs 12.5149 KOps/s 11.9834 KOps/s $\color{#35bf28}+4.44\%$
test_seq_add[compile-overhead] 0.1733ms 77.1783μs 12.9570 KOps/s 12.5637 KOps/s $\color{#35bf28}+3.13\%$
test_seq_wrap[eager] 0.8358ms 0.4575ms 2.1858 KOps/s 2.1233 KOps/s $\color{#35bf28}+2.94\%$
test_seq_wrap[compile] 0.4401ms 0.2474ms 4.0415 KOps/s 4.0178 KOps/s $\color{#35bf28}+0.59\%$
test_seq_wrap[compile-overhead] 0.4657ms 0.2496ms 4.0064 KOps/s 3.9911 KOps/s $\color{#35bf28}+0.39\%$
test_func_call_runtime[False-eager] 0.8308ms 0.5350ms 1.8691 KOps/s 1.8506 KOps/s $\color{#35bf28}+1.00\%$
test_func_call_runtime[False-compile] 0.6156ms 0.4464ms 2.2400 KOps/s 2.1868 KOps/s $\color{#35bf28}+2.43\%$
test_func_call_runtime[False-compile-overhead] 0.6147ms 0.4488ms 2.2281 KOps/s 2.2118 KOps/s $\color{#35bf28}+0.74\%$
test_func_call_runtime[True-eager] 1.0959ms 0.7576ms 1.3200 KOps/s 1.3167 KOps/s $\color{#35bf28}+0.25\%$
test_func_call_runtime[True-compile] 0.7522ms 0.4674ms 2.1394 KOps/s 2.1098 KOps/s $\color{#35bf28}+1.40\%$
test_func_call_runtime[True-compile-overhead] 0.8446ms 0.4744ms 2.1081 KOps/s 2.0895 KOps/s $\color{#35bf28}+0.89\%$
test_func_call_cm_runtime[False-eager] 0.7827ms 0.5373ms 1.8611 KOps/s 1.8499 KOps/s $\color{#35bf28}+0.60\%$
test_func_call_cm_runtime[False-compile] 0.5815ms 0.4566ms 2.1902 KOps/s 2.2309 KOps/s $\color{#d91a1a}-1.82\%$
test_func_call_cm_runtime[False-compile-overhead] 0.6224ms 0.4487ms 2.2286 KOps/s 2.2051 KOps/s $\color{#35bf28}+1.07\%$
test_func_call_cm_runtime[True-eager] 1.3825ms 0.9045ms 1.1056 KOps/s 1.1037 KOps/s $\color{#35bf28}+0.17\%$
test_func_call_cm_runtime[True-compile] 1.0467ms 0.8021ms 1.2467 KOps/s 1.2265 KOps/s $\color{#35bf28}+1.65\%$
test_func_call_cm_runtime[True-compile-overhead] 1.2803ms 0.8109ms 1.2332 KOps/s 1.2069 KOps/s $\color{#35bf28}+2.18\%$
test_vmap_func_call_cm_runtime[eager] 2.7860ms 1.9202ms 520.7861 Ops/s 514.7588 Ops/s $\color{#35bf28}+1.17\%$
test_vmap_func_call_cm_runtime[compile] 0.7593ms 0.5477ms 1.8258 KOps/s 1.8297 KOps/s $\color{#d91a1a}-0.21\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.7516ms 0.5504ms 1.8170 KOps/s 1.8214 KOps/s $\color{#d91a1a}-0.24\%$
test_distributed 0.2498ms 0.1246ms 8.0266 KOps/s 7.6679 KOps/s $\color{#35bf28}+4.68\%$
test_tdmodule 58.4590μs 28.2308μs 35.4223 KOps/s 33.9340 KOps/s $\color{#35bf28}+4.39\%$
test_tdmodule_dispatch 80.0590μs 50.7941μs 19.6873 KOps/s 19.3961 KOps/s $\color{#35bf28}+1.50\%$
test_tdseq 73.2760μs 30.0899μs 33.2338 KOps/s 32.5507 KOps/s $\color{#35bf28}+2.10\%$
test_tdseq_dispatch 92.1020μs 56.2543μs 17.7764 KOps/s 17.5482 KOps/s $\color{#35bf28}+1.30\%$
test_instantiation_functorch 2.1277ms 1.5428ms 648.1629 Ops/s 631.1859 Ops/s $\color{#35bf28}+2.69\%$
test_exec_functorch 0.3395ms 0.1798ms 5.5625 KOps/s 5.4768 KOps/s $\color{#35bf28}+1.57\%$
test_exec_functional_call 0.3331ms 0.1723ms 5.8026 KOps/s 5.7206 KOps/s $\color{#35bf28}+1.43\%$
test_exec_td_decorator 0.5157ms 0.2362ms 4.2332 KOps/s 4.2427 KOps/s $\color{#d91a1a}-0.22\%$
test_vmap_mlp_speed_decorator[True-True] 0.8294ms 0.6582ms 1.5193 KOps/s 1.4966 KOps/s $\color{#35bf28}+1.52\%$
test_vmap_mlp_speed_decorator[True-False] 1.0598ms 0.6676ms 1.4980 KOps/s 1.5030 KOps/s $\color{#d91a1a}-0.33\%$
test_vmap_mlp_speed_decorator[False-True] 0.8716ms 0.5360ms 1.8657 KOps/s 1.8468 KOps/s $\color{#35bf28}+1.02\%$
test_vmap_mlp_speed_decorator[False-False] 0.9417ms 0.5366ms 1.8636 KOps/s 1.8424 KOps/s $\color{#35bf28}+1.15\%$
test_to_module_speed[True] 2.1944ms 1.3426ms 744.8141 Ops/s 743.4786 Ops/s $\color{#35bf28}+0.18\%$
test_to_module_speed[False] 1.8788ms 1.3030ms 767.4839 Ops/s 750.7431 Ops/s $\color{#35bf28}+2.23\%$
test_tc_init 82.8750μs 46.7886μs 21.3727 KOps/s 21.2816 KOps/s $\color{#35bf28}+0.43\%$
test_tc_init_nested 0.1910ms 90.8167μs 11.0112 KOps/s 10.8947 KOps/s $\color{#35bf28}+1.07\%$
test_tc_first_layer_tensor 16.3000μs 1.5093μs 662.5394 KOps/s 644.3362 KOps/s $\color{#35bf28}+2.83\%$
test_tc_first_layer_nontensor 30.2870μs 4.6477μs 215.1584 KOps/s 204.4022 KOps/s $\textbf{\color{#35bf28}+5.26\%}$
test_tc_second_layer_tensor 23.9440μs 2.8023μs 356.8471 KOps/s 346.7791 KOps/s $\color{#35bf28}+2.90\%$
test_tc_second_layer_nontensor 39.1220μs 5.9418μs 168.2985 KOps/s 164.3296 KOps/s $\color{#35bf28}+2.42\%$
test_unbind 0.2554s 13.9350ms 71.7619 Ops/s 65.7394 Ops/s $\textbf{\color{#35bf28}+9.16\%}$
test_full_like 11.0787ms 8.7162ms 114.7283 Ops/s 101.5576 Ops/s $\textbf{\color{#35bf28}+12.97\%}$
test_zeros_like 11.0329ms 5.5516ms 180.1275 Ops/s 295.3502 Ops/s $\textbf{\color{#d91a1a}-39.01\%}$
test_ones_like 6.2309ms 3.8616ms 258.9596 Ops/s 177.9484 Ops/s $\textbf{\color{#35bf28}+45.53\%}$
test_clone 7.5683ms 5.9230ms 168.8342 Ops/s 167.4963 Ops/s $\color{#35bf28}+0.80\%$
test_squeeze 65.6230μs 12.5589μs 79.6246 KOps/s 74.5982 KOps/s $\textbf{\color{#35bf28}+6.74\%}$
test_unsqueeze 0.4493ms 94.4333μs 10.5895 KOps/s 10.3046 KOps/s $\color{#35bf28}+2.76\%$
test_split 0.3270ms 0.1911ms 5.2320 KOps/s 5.0412 KOps/s $\color{#35bf28}+3.78\%$
test_permute 0.3401ms 0.2052ms 4.8744 KOps/s 4.9478 KOps/s $\color{#d91a1a}-1.48\%$
test_stack 35.0437ms 26.7495ms 37.3839 Ops/s 36.7157 Ops/s $\color{#35bf28}+1.82\%$
test_cat 33.0697ms 26.6343ms 37.5456 Ops/s 37.1603 Ops/s $\color{#35bf28}+1.04\%$

Copy link

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}35$. Worsened: $\large\color{#d91a1a}17$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 42.4410μs 12.9473μs 77.2361 KOps/s 82.1069 KOps/s $\textbf{\color{#d91a1a}-5.93\%}$
test_plain_set_stack_nested 41.2400μs 13.0755μs 76.4792 KOps/s 81.2972 KOps/s $\textbf{\color{#d91a1a}-5.93\%}$
test_plain_set_nested_inplace 46.8310μs 14.1269μs 70.7868 KOps/s 74.9224 KOps/s $\textbf{\color{#d91a1a}-5.52\%}$
test_plain_set_stack_nested_inplace 45.9310μs 13.9597μs 71.6349 KOps/s 75.5762 KOps/s $\textbf{\color{#d91a1a}-5.22\%}$
test_items 29.2400μs 2.8779μs 347.4743 KOps/s 347.2678 KOps/s $\color{#35bf28}+0.06\%$
test_items_nested 0.4427ms 0.3655ms 2.7363 KOps/s 2.7283 KOps/s $\color{#35bf28}+0.29\%$
test_items_nested_locked 0.4326ms 0.3667ms 2.7267 KOps/s 2.7399 KOps/s $\color{#d91a1a}-0.48\%$
test_items_nested_leaf 0.1111ms 60.3494μs 16.5702 KOps/s 16.4128 KOps/s $\color{#35bf28}+0.96\%$
test_items_stack_nested 0.4037ms 0.3638ms 2.7485 KOps/s 2.7320 KOps/s $\color{#35bf28}+0.60\%$
test_items_stack_nested_leaf 90.8810μs 61.8287μs 16.1737 KOps/s 16.2567 KOps/s $\color{#d91a1a}-0.51\%$
test_items_stack_nested_locked 0.4222ms 0.3649ms 2.7402 KOps/s 2.7209 KOps/s $\color{#35bf28}+0.71\%$
test_keys 36.2010μs 3.4152μs 292.8103 KOps/s 290.9762 KOps/s $\color{#35bf28}+0.63\%$
test_keys_nested 0.2453ms 88.8748μs 11.2518 KOps/s 11.3805 KOps/s $\color{#d91a1a}-1.13\%$
test_keys_nested_locked 0.7229ms 94.5386μs 10.5777 KOps/s 10.6354 KOps/s $\color{#d91a1a}-0.54\%$
test_keys_nested_leaf 0.1074ms 79.3613μs 12.6006 KOps/s 12.6612 KOps/s $\color{#d91a1a}-0.48\%$
test_keys_stack_nested 0.1309ms 89.0176μs 11.2337 KOps/s 11.3035 KOps/s $\color{#d91a1a}-0.62\%$
test_keys_stack_nested_leaf 0.1609ms 80.4155μs 12.4354 KOps/s 12.5701 KOps/s $\color{#d91a1a}-1.07\%$
test_keys_stack_nested_locked 0.1323ms 95.1855μs 10.5058 KOps/s 10.5310 KOps/s $\color{#d91a1a}-0.24\%$
test_values 4.9535μs 0.8531μs 1.1722 MOps/s 1.1744 MOps/s $\color{#d91a1a}-0.18\%$
test_values_nested 68.1600μs 37.2055μs 26.8778 KOps/s 27.0871 KOps/s $\color{#d91a1a}-0.77\%$
test_values_nested_locked 69.2910μs 40.0546μs 24.9659 KOps/s 25.8114 KOps/s $\color{#d91a1a}-3.28\%$
test_values_nested_leaf 84.3510μs 42.2594μs 23.6634 KOps/s 23.6128 KOps/s $\color{#35bf28}+0.21\%$
test_values_stack_nested 69.7900μs 37.9125μs 26.3765 KOps/s 26.6431 KOps/s $\color{#d91a1a}-1.00\%$
test_values_stack_nested_leaf 70.4200μs 42.5369μs 23.5090 KOps/s 23.4741 KOps/s $\color{#35bf28}+0.15\%$
test_values_stack_nested_locked 66.3410μs 40.1199μs 24.9253 KOps/s 25.3988 KOps/s $\color{#d91a1a}-1.86\%$
test_membership 1.6150μs 0.5007μs 1.9974 MOps/s 1.9942 MOps/s $\color{#35bf28}+0.16\%$
test_membership_nested 15.7400μs 1.9956μs 501.1022 KOps/s 499.7053 KOps/s $\color{#35bf28}+0.28\%$
test_membership_nested_leaf 14.0550μs 1.9948μs 501.3142 KOps/s 496.2357 KOps/s $\color{#35bf28}+1.02\%$
test_membership_stacked_nested 29.1900μs 2.0720μs 482.6207 KOps/s 479.9690 KOps/s $\color{#35bf28}+0.55\%$
test_membership_stacked_nested_leaf 41.1410μs 2.0880μs 478.9242 KOps/s 475.0712 KOps/s $\color{#35bf28}+0.81\%$
test_membership_nested_last 30.3900μs 3.0775μs 324.9422 KOps/s 329.6866 KOps/s $\color{#d91a1a}-1.44\%$
test_membership_nested_leaf_last 33.1800μs 3.0788μs 324.8007 KOps/s 327.8268 KOps/s $\color{#d91a1a}-0.92\%$
test_membership_stacked_nested_last 52.2810μs 3.0266μs 330.4071 KOps/s 325.9607 KOps/s $\color{#35bf28}+1.36\%$
test_membership_stacked_nested_leaf_last 32.3900μs 3.0463μs 328.2674 KOps/s 329.3716 KOps/s $\color{#d91a1a}-0.34\%$
test_nested_getleaf 34.7700μs 6.2157μs 160.8835 KOps/s 161.7774 KOps/s $\color{#d91a1a}-0.55\%$
test_nested_get 38.7610μs 5.9585μs 167.8269 KOps/s 168.8545 KOps/s $\color{#d91a1a}-0.61\%$
test_stacked_getleaf 75.2300μs 6.1469μs 162.6837 KOps/s 163.6826 KOps/s $\color{#d91a1a}-0.61\%$
test_stacked_get 36.8010μs 5.8273μs 171.6071 KOps/s 173.2578 KOps/s $\color{#d91a1a}-0.95\%$
test_nested_getitemleaf 37.1100μs 6.4701μs 154.5578 KOps/s 155.4923 KOps/s $\color{#d91a1a}-0.60\%$
test_nested_getitem 35.9600μs 6.1662μs 162.1752 KOps/s 165.3959 KOps/s $\color{#d91a1a}-1.95\%$
test_stacked_getitemleaf 37.8200μs 6.3943μs 156.3895 KOps/s 157.3554 KOps/s $\color{#d91a1a}-0.61\%$
test_stacked_getitem 39.7010μs 5.9797μs 167.2325 KOps/s 167.8932 KOps/s $\color{#d91a1a}-0.39\%$
test_lock_nested 0.3964ms 0.3370ms 2.9674 KOps/s 2.8654 KOps/s $\color{#35bf28}+3.56\%$
test_lock_stack_nested 0.3808ms 0.3440ms 2.9073 KOps/s 2.8266 KOps/s $\color{#35bf28}+2.86\%$
test_unlock_nested 0.3571ms 0.2843ms 3.5175 KOps/s 3.4142 KOps/s $\color{#35bf28}+3.02\%$
test_unlock_stack_nested 0.3419ms 0.2848ms 3.5110 KOps/s 3.4019 KOps/s $\color{#35bf28}+3.21\%$
test_flatten_speed 0.1145ms 77.0069μs 12.9858 KOps/s 12.7422 KOps/s $\color{#35bf28}+1.91\%$
test_unflatten_speed 0.3736ms 0.3200ms 3.1249 KOps/s 3.0754 KOps/s $\color{#35bf28}+1.61\%$
test_common_ops 0.7488ms 0.6253ms 1.5993 KOps/s 1.6016 KOps/s $\color{#d91a1a}-0.14\%$
test_creation 59.0810μs 1.7002μs 588.1681 KOps/s 571.8884 KOps/s $\color{#35bf28}+2.85\%$
test_creation_empty 40.0500μs 9.2531μs 108.0714 KOps/s 128.5504 KOps/s $\textbf{\color{#d91a1a}-15.93\%}$
test_creation_nested_1 45.4500μs 10.9246μs 91.5365 KOps/s 106.5388 KOps/s $\textbf{\color{#d91a1a}-14.08\%}$
test_creation_nested_2 32.8900μs 13.5067μs 74.0373 KOps/s 82.7286 KOps/s $\textbf{\color{#d91a1a}-10.51\%}$
test_clone 30.4600μs 10.3971μs 96.1807 KOps/s 85.7913 KOps/s $\textbf{\color{#35bf28}+12.11\%}$
test_getitem[int] 1.1569ms 10.4169μs 95.9977 KOps/s 90.9073 KOps/s $\textbf{\color{#35bf28}+5.60\%}$
test_getitem[slice_int] 0.1108ms 20.6476μs 48.4317 KOps/s 45.5889 KOps/s $\textbf{\color{#35bf28}+6.24\%}$
test_getitem[range] 0.1373ms 37.1506μs 26.9175 KOps/s 25.4275 KOps/s $\textbf{\color{#35bf28}+5.86\%}$
test_getitem[tuple] 0.1054ms 17.7939μs 56.1990 KOps/s 52.9626 KOps/s $\textbf{\color{#35bf28}+6.11\%}$
test_getitem[list] 0.1208ms 33.6572μs 29.7113 KOps/s 28.6919 KOps/s $\color{#35bf28}+3.55\%$
test_setitem_dim[int] 38.8100μs 19.1940μs 52.0996 KOps/s 47.7122 KOps/s $\textbf{\color{#35bf28}+9.20\%}$
test_setitem_dim[slice_int] 62.2210μs 38.7780μs 25.7878 KOps/s 24.3369 KOps/s $\textbf{\color{#35bf28}+5.96\%}$
test_setitem_dim[range] 86.1500μs 53.8706μs 18.5630 KOps/s 18.2684 KOps/s $\color{#35bf28}+1.61\%$
test_setitem_dim[tuple] 67.4100μs 32.7989μs 30.4888 KOps/s 29.5790 KOps/s $\color{#35bf28}+3.08\%$
test_setitem 82.6510μs 15.5318μs 64.3839 KOps/s 63.0476 KOps/s $\color{#35bf28}+2.12\%$
test_set 74.4900μs 15.2481μs 65.5820 KOps/s 63.9710 KOps/s $\color{#35bf28}+2.52\%$
test_set_shared 0.6095ms 0.1579ms 6.3314 KOps/s 6.1650 KOps/s $\color{#35bf28}+2.70\%$
test_update 0.4321ms 18.6941μs 53.4929 KOps/s 55.0318 KOps/s $\color{#d91a1a}-2.80\%$
test_update_nested 85.5710μs 24.3766μs 41.0229 KOps/s 42.5963 KOps/s $\color{#d91a1a}-3.69\%$
test_update__nested 0.4677ms 25.1550μs 39.7535 KOps/s 37.9752 KOps/s $\color{#35bf28}+4.68\%$
test_set_nested 85.2410μs 16.6335μs 60.1197 KOps/s 59.7823 KOps/s $\color{#35bf28}+0.56\%$
test_set_nested_new 83.2800μs 18.6391μs 53.6506 KOps/s 51.8797 KOps/s $\color{#35bf28}+3.41\%$
test_select 61.8800μs 30.3506μs 32.9483 KOps/s 32.7000 KOps/s $\color{#35bf28}+0.76\%$
test_select_nested 72.3510μs 43.2949μs 23.0974 KOps/s 22.7087 KOps/s $\color{#35bf28}+1.71\%$
test_exclude_nested 98.3510μs 62.2705μs 16.0590 KOps/s 15.5931 KOps/s $\color{#35bf28}+2.99\%$
test_empty[True] 0.3537ms 0.2929ms 3.4137 KOps/s 3.3822 KOps/s $\color{#35bf28}+0.93\%$
test_empty[False] 3.3310μs 0.8241μs 1.2135 MOps/s 1.1873 MOps/s $\color{#35bf28}+2.21\%$
test_to 87.3400μs 56.0704μs 17.8347 KOps/s 17.3326 KOps/s $\color{#35bf28}+2.90\%$
test_to_nonblocking 80.1710μs 47.4707μs 21.0656 KOps/s 20.8722 KOps/s $\color{#35bf28}+0.93\%$
test_unbind_speed 0.3013ms 0.2386ms 4.1913 KOps/s 3.9254 KOps/s $\textbf{\color{#35bf28}+6.77\%}$
test_unbind_speed_stack0 0.2870ms 0.2414ms 4.1419 KOps/s 3.9847 KOps/s $\color{#35bf28}+3.94\%$
test_unbind_speed_stack1 93.8256ms 0.7379ms 1.3553 KOps/s 1.3170 KOps/s $\color{#35bf28}+2.91\%$
test_split 94.9586ms 1.5945ms 627.1508 Ops/s 601.5373 Ops/s $\color{#35bf28}+4.26\%$
test_chunk 97.1292ms 1.6108ms 620.8248 Ops/s 602.1929 Ops/s $\color{#35bf28}+3.09\%$
test_consolidate[False-None] 96.9469ms 3.0475ms 328.1386 Ops/s 357.9728 Ops/s $\textbf{\color{#d91a1a}-8.33\%}$
test_consolidate[default-None] 1.8280ms 1.7178ms 582.1477 Ops/s 553.9440 Ops/s $\textbf{\color{#35bf28}+5.09\%}$
test_consolidate[reduce-overhead-None] 1.8365ms 1.7609ms 567.8864 Ops/s 542.5081 Ops/s $\color{#35bf28}+4.68\%$
test_consolidate_njt[False-None] 0.2990s 8.5374ms 117.1322 Ops/s 147.1212 Ops/s $\textbf{\color{#d91a1a}-20.38\%}$
test_to[False-False-None] 1.9169ms 1.7365ms 575.8585 Ops/s 555.8849 Ops/s $\color{#35bf28}+3.59\%$
test_to[True-False-None] 1.5811ms 1.3342ms 749.5263 Ops/s 703.8173 Ops/s $\textbf{\color{#35bf28}+6.49\%}$
test_to[within-False-None] 4.2540ms 4.1764ms 239.4425 Ops/s 229.6034 Ops/s $\color{#35bf28}+4.29\%$
test_to[True-default-None] 5.5548ms 5.3500ms 186.9172 Ops/s 183.6667 Ops/s $\color{#35bf28}+1.77\%$
test_to_njt[False-False-None] 7.0518ms 6.9406ms 144.0799 Ops/s 140.7829 Ops/s $\color{#35bf28}+2.34\%$
test_to_njt[True-False-None] 5.9258ms 5.5194ms 181.1805 Ops/s 180.7354 Ops/s $\color{#35bf28}+0.25\%$
test_to_njt[within-False-None] 12.2543ms 12.0997ms 82.6467 Ops/s 79.1606 Ops/s $\color{#35bf28}+4.40\%$
test_creation[device0] 0.4580ms 81.2535μs 12.3072 KOps/s 12.1969 KOps/s $\color{#35bf28}+0.90\%$
test_creation_from_tensor 0.4705ms 84.4361μs 11.8433 KOps/s 11.7775 KOps/s $\color{#35bf28}+0.56\%$
test_add_one[memmap_tensor0] 0.5592ms 6.8223μs 146.5787 KOps/s 138.4580 KOps/s $\textbf{\color{#35bf28}+5.87\%}$
test_contiguous[memmap_tensor0] 1.8420μs 0.4223μs 2.3679 MOps/s 2.3607 MOps/s $\color{#35bf28}+0.31\%$
test_stack[memmap_tensor0] 36.7600μs 4.2627μs 234.5939 KOps/s 199.8985 KOps/s $\textbf{\color{#35bf28}+17.36\%}$
test_memmaptd_index 1.6184ms 0.2440ms 4.0984 KOps/s 3.8298 KOps/s $\textbf{\color{#35bf28}+7.01\%}$
test_memmaptd_index_astensor 0.4340ms 0.3027ms 3.3035 KOps/s 3.1147 KOps/s $\textbf{\color{#35bf28}+6.06\%}$
test_memmaptd_index_op 0.7348ms 0.5948ms 1.6812 KOps/s 1.6548 KOps/s $\color{#35bf28}+1.60\%$
test_serialize_model 0.1311s 0.1305s 7.6607 Ops/s 7.6885 Ops/s $\color{#d91a1a}-0.36\%$
test_serialize_model_pickle 1.3496s 1.2124s 0.8248 Ops/s 0.8258 Ops/s $\color{#d91a1a}-0.12\%$
test_serialize_weights 0.2736s 0.1507s 6.6367 Ops/s 7.7209 Ops/s $\textbf{\color{#d91a1a}-14.04\%}$
test_serialize_weights_returnearly 0.3358s 54.5353ms 18.3367 Ops/s 14.3548 Ops/s $\textbf{\color{#35bf28}+27.74\%}$
test_serialize_weights_pickle 1.3974s 1.2202s 0.8196 Ops/s 0.8373 Ops/s $\color{#d91a1a}-2.12\%$
test_reshape_pytree 53.3600μs 22.0457μs 45.3604 KOps/s 43.8497 KOps/s $\color{#35bf28}+3.45\%$
test_reshape_td 64.8100μs 26.8934μs 37.1839 KOps/s 35.6497 KOps/s $\color{#35bf28}+4.30\%$
test_view_pytree 55.1510μs 22.0100μs 45.4340 KOps/s 44.7874 KOps/s $\color{#35bf28}+1.44\%$
test_view_td 58.8400μs 30.8270μs 32.4391 KOps/s 30.1721 KOps/s $\textbf{\color{#35bf28}+7.51\%}$
test_unbind_pytree 62.0010μs 27.8686μs 35.8826 KOps/s 33.9435 KOps/s $\textbf{\color{#35bf28}+5.71\%}$
test_unbind_td 0.5745ms 36.9867μs 27.0368 KOps/s 25.6974 KOps/s $\textbf{\color{#35bf28}+5.21\%}$
test_split_pytree 76.4310μs 29.6784μs 33.6945 KOps/s 32.4407 KOps/s $\color{#35bf28}+3.86\%$
test_split_td 0.7102ms 38.2555μs 26.1400 KOps/s 24.8686 KOps/s $\textbf{\color{#35bf28}+5.11\%}$
test_add_pytree 74.0800μs 36.3927μs 27.4780 KOps/s 27.6077 KOps/s $\color{#d91a1a}-0.47\%$
test_add_td 0.1026ms 51.5359μs 19.4040 KOps/s 20.2402 KOps/s $\color{#d91a1a}-4.13\%$
test_compile_add_one_nested[tensordict-compile] 0.1737ms 0.1204ms 8.3052 KOps/s 7.7502 KOps/s $\textbf{\color{#35bf28}+7.16\%}$
test_compile_add_one_nested[tensordict-eager] 0.2335ms 0.1337ms 7.4779 KOps/s 7.4313 KOps/s $\color{#35bf28}+0.63\%$
test_compile_add_one_nested[pytree-compile] 0.1500ms 95.5183μs 10.4692 KOps/s 10.0731 KOps/s $\color{#35bf28}+3.93\%$
test_compile_add_one_nested[pytree-eager] 1.0753ms 0.1505ms 6.6438 KOps/s 6.4145 KOps/s $\color{#35bf28}+3.58\%$
test_compile_copy_nested[tensordict-compile] 59.0600μs 25.2455μs 39.6109 KOps/s 30.0015 KOps/s $\textbf{\color{#35bf28}+32.03\%}$
test_compile_copy_nested[tensordict-eager] 63.7200μs 29.4712μs 33.9314 KOps/s 33.7904 KOps/s $\color{#35bf28}+0.42\%$
test_compile_copy_nested[pytree-compile] 0.1337ms 63.4312μs 15.7651 KOps/s 15.3807 KOps/s $\color{#35bf28}+2.50\%$
test_compile_copy_nested[pytree-eager] 88.3010μs 48.6563μs 20.5523 KOps/s 20.1061 KOps/s $\color{#35bf28}+2.22\%$
test_compile_add_one_flat[tensordict-compile] 0.1961ms 0.1433ms 6.9806 KOps/s 6.9179 KOps/s $\color{#35bf28}+0.91\%$
test_compile_add_one_flat[tensordict-eager] 0.3374ms 0.2183ms 4.5816 KOps/s 4.6299 KOps/s $\color{#d91a1a}-1.04\%$
test_compile_add_one_flat[tensorclass-compile] 0.1508ms 98.6128μs 10.1407 KOps/s 10.0595 KOps/s $\color{#35bf28}+0.81\%$
test_compile_add_one_flat[tensorclass-eager] 0.1446ms 54.9015μs 18.2144 KOps/s 17.2229 KOps/s $\textbf{\color{#35bf28}+5.76\%}$
test_compile_add_one_flat[pytree-compile] 0.2454ms 0.1374ms 7.2781 KOps/s 7.2692 KOps/s $\color{#35bf28}+0.12\%$
test_compile_add_one_flat[pytree-eager] 0.5558ms 0.4878ms 2.0501 KOps/s 1.9790 KOps/s $\color{#35bf28}+3.59\%$
test_compile_add_self_flat[tensordict-eager] 0.3962ms 0.2613ms 3.8269 KOps/s 3.7995 KOps/s $\color{#35bf28}+0.72\%$
test_compile_add_self_flat[tensordict-compile] 0.1898ms 0.1452ms 6.8874 KOps/s 7.0152 KOps/s $\color{#d91a1a}-1.82\%$
test_compile_add_self_flat[tensorclass-eager] 0.1655ms 66.9530μs 14.9359 KOps/s 14.5863 KOps/s $\color{#35bf28}+2.40\%$
test_compile_add_self_flat[tensorclass-compile] 0.1614ms 0.1021ms 9.7927 KOps/s 10.0572 KOps/s $\color{#d91a1a}-2.63\%$
test_compile_add_self_flat[pytree-eager] 0.7659ms 0.4144ms 2.4132 KOps/s 2.3418 KOps/s $\color{#35bf28}+3.05\%$
test_compile_add_self_flat[pytree-compile] 0.1853ms 0.1363ms 7.3384 KOps/s 7.3467 KOps/s $\color{#d91a1a}-0.11\%$
test_compile_copy_flat[tensordict-compile] 56.1310μs 18.9235μs 52.8445 KOps/s 52.9804 KOps/s $\color{#d91a1a}-0.26\%$
test_compile_copy_flat[tensordict-eager] 0.1072ms 31.1341μs 32.1191 KOps/s 32.1674 KOps/s $\color{#d91a1a}-0.15\%$
test_compile_copy_flat[pytree-compile] 0.1771ms 69.5201μs 14.3843 KOps/s 14.6334 KOps/s $\color{#d91a1a}-1.70\%$
test_compile_copy_flat[pytree-eager] 85.1410μs 51.7291μs 19.3315 KOps/s 19.2392 KOps/s $\color{#35bf28}+0.48\%$
test_compile_assign_and_add[tensordict-compile] 1.6402ms 0.3943ms 2.5359 KOps/s 2.1355 KOps/s $\textbf{\color{#35bf28}+18.75\%}$
test_compile_assign_and_add[tensordict-eager] 2.9011ms 2.6828ms 372.7394 Ops/s 350.6977 Ops/s $\textbf{\color{#35bf28}+6.29\%}$
test_compile_assign_and_add[pytree-compile] 1.5930ms 0.3818ms 2.6190 KOps/s 2.1714 KOps/s $\textbf{\color{#35bf28}+20.61\%}$
test_compile_assign_and_add[pytree-eager] 3.0666ms 2.6610ms 375.8039 Ops/s 361.6411 Ops/s $\color{#35bf28}+3.92\%$
test_compile_indexing[tensor-tensordict-compile] 0.5338ms 0.1162ms 8.6050 KOps/s 8.2345 KOps/s $\color{#35bf28}+4.50\%$
test_compile_indexing[tensor-tensordict-eager] 0.5929ms 86.1244μs 11.6111 KOps/s 11.9350 KOps/s $\color{#d91a1a}-2.71\%$
test_compile_indexing[tensor-tensorclass-compile] 0.5400ms 0.1176ms 8.5066 KOps/s 8.8393 KOps/s $\color{#d91a1a}-3.76\%$
test_compile_indexing[tensor-tensorclass-eager] 0.4609ms 68.9560μs 14.5020 KOps/s 13.8015 KOps/s $\textbf{\color{#35bf28}+5.08\%}$
test_compile_indexing[tensor-pytree-compile] 0.5553ms 0.1155ms 8.6596 KOps/s 8.6876 KOps/s $\color{#d91a1a}-0.32\%$
test_compile_indexing[tensor-pytree-eager] 0.4910ms 73.5635μs 13.5937 KOps/s 13.9208 KOps/s $\color{#d91a1a}-2.35\%$
test_compile_indexing[slice-tensordict-compile] 0.5042ms 0.1032ms 9.6890 KOps/s 9.8639 KOps/s $\color{#d91a1a}-1.77\%$
test_compile_indexing[slice-tensordict-eager] 0.4320ms 20.3690μs 49.0943 KOps/s 51.5429 KOps/s $\color{#d91a1a}-4.75\%$
test_compile_indexing[slice-tensorclass-compile] 0.1427ms 0.1003ms 9.9729 KOps/s 10.3110 KOps/s $\color{#d91a1a}-3.28\%$
test_compile_indexing[slice-tensorclass-eager] 0.4111ms 18.0733μs 55.3303 KOps/s 61.2564 KOps/s $\textbf{\color{#d91a1a}-9.67\%}$
test_compile_indexing[slice-pytree-compile] 0.5101ms 0.1016ms 9.8403 KOps/s 9.3999 KOps/s $\color{#35bf28}+4.69\%$
test_compile_indexing[slice-pytree-eager] 0.4090ms 17.9971μs 55.5645 KOps/s 61.6225 KOps/s $\textbf{\color{#d91a1a}-9.83\%}$
test_compile_indexing[int-tensordict-compile] 0.5228ms 0.1095ms 9.1350 KOps/s 9.5225 KOps/s $\color{#d91a1a}-4.07\%$
test_compile_indexing[int-tensordict-eager] 0.5422ms 20.6330μs 48.4661 KOps/s 50.9475 KOps/s $\color{#d91a1a}-4.87\%$
test_compile_indexing[int-tensorclass-compile] 0.5239ms 0.1052ms 9.5099 KOps/s 9.4628 KOps/s $\color{#35bf28}+0.50\%$
test_compile_indexing[int-tensorclass-eager] 55.8200μs 18.1618μs 55.0605 KOps/s 62.1849 KOps/s $\textbf{\color{#d91a1a}-11.46\%}$
test_compile_indexing[int-pytree-compile] 0.5189ms 0.1053ms 9.4941 KOps/s 9.8768 KOps/s $\color{#d91a1a}-3.87\%$
test_compile_indexing[int-pytree-eager] 0.4138ms 18.0792μs 55.3121 KOps/s 57.6873 KOps/s $\color{#d91a1a}-4.12\%$
test_mod_add[eager] 0.4552ms 40.1310μs 24.9184 KOps/s 25.0001 KOps/s $\color{#d91a1a}-0.33\%$
test_mod_add[compile] 0.1256ms 82.1697μs 12.1699 KOps/s 11.8217 KOps/s $\color{#35bf28}+2.95\%$
test_mod_add[compile-overhead] 0.3242ms 0.1686ms 5.9312 KOps/s 5.4280 KOps/s $\textbf{\color{#35bf28}+9.27\%}$
test_mod_wrap[eager] 0.6701ms 0.2533ms 3.9475 KOps/s 3.7767 KOps/s $\color{#35bf28}+4.52\%$
test_mod_wrap[compile] 0.3410ms 0.2902ms 3.4456 KOps/s 3.3945 KOps/s $\color{#35bf28}+1.51\%$
test_mod_wrap[compile-overhead] 7.4575ms 3.8465ms 259.9785 Ops/s 262.9542 Ops/s $\color{#d91a1a}-1.13\%$
test_mod_wrap_and_backward[eager] 1.5971ms 1.3907ms 719.0693 Ops/s 685.2285 Ops/s $\color{#35bf28}+4.94\%$
test_mod_wrap_and_backward[compile] 1.3601ms 1.2851ms 778.1365 Ops/s 758.5909 Ops/s $\color{#35bf28}+2.58\%$
test_mod_wrap_and_backward[compile-overhead] 1.3754ms 0.9258ms 1.0801 KOps/s 1.0408 KOps/s $\color{#35bf28}+3.78\%$
test_seq_add[eager] 0.1714ms 0.1190ms 8.4046 KOps/s 7.8992 KOps/s $\textbf{\color{#35bf28}+6.40\%}$
test_seq_add[compile] 0.2247ms 91.0752μs 10.9799 KOps/s 10.3289 KOps/s $\textbf{\color{#35bf28}+6.30\%}$
test_seq_add[compile-overhead] 0.2285ms 0.1294ms 7.7286 KOps/s 7.2283 KOps/s $\textbf{\color{#35bf28}+6.92\%}$
test_seq_wrap[eager] 0.4946ms 0.4294ms 2.3291 KOps/s 2.1416 KOps/s $\textbf{\color{#35bf28}+8.75\%}$
test_seq_wrap[compile] 0.3630ms 0.3080ms 3.2463 KOps/s 3.1852 KOps/s $\color{#35bf28}+1.92\%$
test_seq_wrap[compile-overhead] 0.2781ms 0.2268ms 4.4089 KOps/s 4.3264 KOps/s $\color{#35bf28}+1.91\%$
test_func_call_runtime[False-eager] 0.8145ms 0.7520ms 1.3297 KOps/s 1.2874 KOps/s $\color{#35bf28}+3.28\%$
test_func_call_runtime[False-compile] 0.9195ms 0.7587ms 1.3181 KOps/s 1.2931 KOps/s $\color{#35bf28}+1.93\%$
test_func_call_runtime[False-compile-overhead] 0.4229ms 0.3657ms 2.7346 KOps/s 2.6999 KOps/s $\color{#35bf28}+1.28\%$
test_func_call_runtime[True-eager] 0.9810ms 0.9078ms 1.1015 KOps/s 1.0294 KOps/s $\textbf{\color{#35bf28}+7.00\%}$
test_func_call_runtime[True-compile] 0.8438ms 0.7816ms 1.2794 KOps/s 1.2133 KOps/s $\textbf{\color{#35bf28}+5.45\%}$
test_func_call_runtime[True-compile-overhead] 0.5101ms 0.3897ms 2.5660 KOps/s 2.5453 KOps/s $\color{#35bf28}+0.82\%$
test_func_call_cm_runtime[False-eager] 0.9298ms 0.7874ms 1.2700 KOps/s 1.3109 KOps/s $\color{#d91a1a}-3.12\%$
test_func_call_cm_runtime[False-compile] 0.8325ms 0.7569ms 1.3212 KOps/s 1.2946 KOps/s $\color{#35bf28}+2.06\%$
test_func_call_cm_runtime[False-compile-overhead] 0.4318ms 0.3704ms 2.6997 KOps/s 2.6843 KOps/s $\color{#35bf28}+0.57\%$
test_func_call_cm_runtime[True-eager] 1.1487ms 1.0227ms 977.7663 Ops/s 960.7579 Ops/s $\color{#35bf28}+1.77\%$
test_func_call_cm_runtime[True-compile] 1.5235ms 1.0291ms 971.6946 Ops/s 914.0805 Ops/s $\textbf{\color{#35bf28}+6.30\%}$
test_func_call_cm_runtime[True-compile-overhead] 1.1365ms 1.0280ms 972.7168 Ops/s 972.2771 Ops/s $\color{#35bf28}+0.05\%$
test_vmap_func_call_cm_runtime[eager] 2.5203ms 2.1266ms 470.2331 Ops/s 464.4808 Ops/s $\color{#35bf28}+1.24\%$
test_vmap_func_call_cm_runtime[compile] 1.0278ms 0.8398ms 1.1908 KOps/s 1.1788 KOps/s $\color{#35bf28}+1.02\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.5203ms 0.4185ms 2.3897 KOps/s 2.3409 KOps/s $\color{#35bf28}+2.08\%$
test_distributed 3.2312ms 0.1949ms 5.1302 KOps/s 8.6953 KOps/s $\textbf{\color{#d91a1a}-41.00\%}$
test_tdmodule 31.9700μs 20.8733μs 47.9082 KOps/s 47.3973 KOps/s $\color{#35bf28}+1.08\%$
test_tdmodule_dispatch 59.3210μs 37.4589μs 26.6960 KOps/s 27.9691 KOps/s $\color{#d91a1a}-4.55\%$
test_tdseq 45.1710μs 21.4475μs 46.6256 KOps/s 47.7505 KOps/s $\color{#d91a1a}-2.36\%$
test_tdseq_dispatch 81.5410μs 40.2693μs 24.8328 KOps/s 25.8168 KOps/s $\color{#d91a1a}-3.81\%$
test_instantiation_functorch 1.6088ms 1.5249ms 655.7649 Ops/s 620.7215 Ops/s $\textbf{\color{#35bf28}+5.65\%}$
test_exec_functorch 0.1865ms 0.1425ms 7.0158 KOps/s 6.7631 KOps/s $\color{#35bf28}+3.74\%$
test_exec_functional_call 0.2111ms 0.1375ms 7.2732 KOps/s 6.9756 KOps/s $\color{#35bf28}+4.27\%$
test_exec_td_decorator 0.3716ms 0.1895ms 5.2762 KOps/s 5.1319 KOps/s $\color{#35bf28}+2.81\%$
test_vmap_mlp_speed_decorator[True-True] 0.8113ms 0.6951ms 1.4387 KOps/s 1.4265 KOps/s $\color{#35bf28}+0.86\%$
test_vmap_mlp_speed_decorator[True-False] 0.8302ms 0.6945ms 1.4399 KOps/s 1.4273 KOps/s $\color{#35bf28}+0.89\%$
test_vmap_mlp_speed_decorator[False-True] 0.7145ms 0.6033ms 1.6574 KOps/s 1.6377 KOps/s $\color{#35bf28}+1.20\%$
test_vmap_mlp_speed_decorator[False-False] 0.7202ms 0.6055ms 1.6516 KOps/s 1.6370 KOps/s $\color{#35bf28}+0.89\%$
test_vmap_transformer_speed_decorator[True-True] 19.8965ms 19.5147ms 51.2434 Ops/s 50.9864 Ops/s $\color{#35bf28}+0.50\%$
test_vmap_transformer_speed_decorator[True-False] 19.5788ms 19.4937ms 51.2985 Ops/s 51.0114 Ops/s $\color{#35bf28}+0.56\%$
test_vmap_transformer_speed_decorator[False-True] 19.4242ms 19.3217ms 51.7554 Ops/s 51.3189 Ops/s $\color{#35bf28}+0.85\%$
test_vmap_transformer_speed_decorator[False-False] 19.8124ms 19.3431ms 51.6981 Ops/s 51.3617 Ops/s $\color{#35bf28}+0.66\%$
test_to_module_speed[True] 1.4429ms 0.9821ms 1.0182 KOps/s 1.0187 KOps/s $\color{#d91a1a}-0.05\%$
test_to_module_speed[False] 1.0343ms 0.9666ms 1.0346 KOps/s 1.0442 KOps/s $\color{#d91a1a}-0.92\%$
test_tc_init 65.4810μs 36.8882μs 27.1089 KOps/s 28.1651 KOps/s $\color{#d91a1a}-3.75\%$
test_tc_init_nested 0.1750ms 75.7878μs 13.1947 KOps/s 14.0747 KOps/s $\textbf{\color{#d91a1a}-6.25\%}$
test_tc_first_layer_tensor 20.5400μs 0.7998μs 1.2504 MOps/s 1.3960 MOps/s $\textbf{\color{#d91a1a}-10.43\%}$
test_tc_first_layer_nontensor 24.5400μs 2.2266μs 449.1173 KOps/s 446.9738 KOps/s $\color{#35bf28}+0.48\%$
test_tc_second_layer_tensor 9.8033μs 1.4298μs 699.4066 KOps/s 698.4161 KOps/s $\color{#35bf28}+0.14\%$
test_tc_second_layer_nontensor 32.4010μs 3.0064μs 332.6271 KOps/s 334.5884 KOps/s $\color{#d91a1a}-0.59\%$
test_unbind 0.2148s 11.8998ms 84.0353 Ops/s 143.2505 Ops/s $\textbf{\color{#d91a1a}-41.34\%}$
test_full_like 10.2990ms 9.2956ms 107.5783 Ops/s 107.6571 Ops/s $\color{#d91a1a}-0.07\%$
test_zeros_like 9.2567ms 7.2593ms 137.7538 Ops/s 137.2764 Ops/s $\color{#35bf28}+0.35\%$
test_ones_like 4.8533ms 4.3220ms 231.3728 Ops/s 231.4124 Ops/s $\color{#d91a1a}-0.02\%$
test_clone 6.8005ms 6.3960ms 156.3482 Ops/s 155.6658 Ops/s $\color{#35bf28}+0.44\%$
test_squeeze 63.8300μs 9.9175μs 100.8314 KOps/s 101.5363 KOps/s $\color{#d91a1a}-0.69\%$
test_unsqueeze 0.1323ms 75.4804μs 13.2485 KOps/s 13.3566 KOps/s $\color{#d91a1a}-0.81\%$
test_split 0.3725ms 0.1618ms 6.1805 KOps/s 6.1409 KOps/s $\color{#35bf28}+0.65\%$
test_permute 0.3361ms 0.1886ms 5.3013 KOps/s 5.3691 KOps/s $\color{#d91a1a}-1.26\%$
test_stack 50.9244ms 50.4311ms 19.8290 Ops/s 19.8367 Ops/s $\color{#d91a1a}-0.04\%$
test_cat 50.6767ms 49.8701ms 20.0521 Ops/s 19.6941 Ops/s $\color{#35bf28}+1.82\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants