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

tear down by unloading from workers #39

Merged
merged 1 commit into from
Dec 18, 2023
Merged

Conversation

edknv
Copy link
Contributor

@edknv edknv commented Dec 17, 2023

No description provided.

@edknv edknv force-pushed the unload-model branch 3 times, most recently from 5a53cf3 to 4c72b19 Compare December 18, 2023 00:58
Copy link

Benchmark comparison for 1f7d20a (main) vs 4c72b199 (PR)


---------------------------------------------------------------------------------------------- benchmark: 2 tests ----------------------------------------------------------------------------------------------
Name (time in s)                                   Min                 Max                Mean            StdDev              Median               IQR            Outliers     OPS            Rounds  Iterations
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_fiqa_all_MiniLM_L6_v2 (0001_main)         98.3329 (1.0)      100.8454 (1.0)       99.3778 (1.0)      1.3279 (1.0)       98.4575 (1.0)      2.4145 (1.20)          2;0  0.0101 (1.0)           5           1
test_fiqa_all_MiniLM_L6_v2 (0002_4c72b19)     185.3408 (1.88)     189.6884 (1.88)     187.0673 (1.88)     1.6573 (1.25)     186.5039 (1.89)     2.0135 (1.0)           2;0  0.0053 (0.53)          5           1
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Legend:
  Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile.
  OPS: Operations Per Second, computed as 1 / Mean

Copy link

Benchmark comparison for 1f7d20a (main) vs 8e6b892d (PR)


--------------------------------------------------------------------------------------------- benchmark: 2 tests --------------------------------------------------------------------------------------------
Name (time in s)                                  Min                 Max               Mean            StdDev             Median               IQR            Outliers     OPS            Rounds  Iterations
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_fiqa_all_MiniLM_L6_v2 (0002_8e6b892)     97.0827 (1.0)      101.6896 (1.0)      99.4667 (1.01)     1.6997 (1.0)      99.3585 (1.01)     2.1483 (1.04)          2;0  0.0101 (0.99)          5           1
test_fiqa_all_MiniLM_L6_v2 (0001_main)        97.3289 (1.00)     102.0789 (1.00)     98.8616 (1.0)      1.8922 (1.11)     98.2761 (1.0)      2.0591 (1.0)           1;0  0.0101 (1.0)           5           1
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Legend:
  Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile.
  OPS: Operations Per Second, computed as 1 / Mean

@edknv edknv marked this pull request as ready for review December 18, 2023 06:36
@edknv edknv merged commit 05accee into rapidsai:main Dec 18, 2023
10 checks passed
@edknv edknv deleted the unload-model branch December 18, 2023 06:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant