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

MultiPL-E 3.0 release #145

Closed
13 tasks done
arjunguha opened this issue Jul 5, 2024 · 2 comments
Closed
13 tasks done

MultiPL-E 3.0 release #145

arjunguha opened this issue Jul 5, 2024 · 2 comments

Comments

@arjunguha
Copy link
Member

arjunguha commented Jul 5, 2024

  • Update the MultiPL-E container
    • Edited Dockerfile to exclude proof assistants. For future, put these in a different container.
    • Rebuilt container on Boa
    • Test executions waiting for completions
    • Update on GitHub
    • Update container for BigCode organization
  • Update the nuprl/MultiPL-E dataset
    • Remove requirement for trust_remote_code=True.
    • Temporary dataset at nuprl-staging/MultiPL-E
    • Publish a diff. Script on Discovery
    • Upload new MBPP
    • Generate completions for select languages. Running on Boa
  • Sort out Elixir Fix Elixir install #147
@arjunguha
Copy link
Member Author

For the record, I'm testing DeepSeekCoderv2-Lite and SC2-15B for the release:

base) [a.guha@d0130 data]$ column -s, -t passk.csv 
Dataset                                                  Pass@k  Estimate  NumProblems  MinCompletions  MaxCompletions
humaneval-clj-deepseekcoder_v2lite_base-0.2-reworded     1       0.16      161          50              50
humaneval-clj-starcoder2_15b-0.2-reworded                1       0.16      161          50              50
humaneval-cpp-deepseekcoder_v2lite_base-0.2-reworded     1       0.46      161          50              50
humaneval-cpp-starcoder2_15b-0.2-reworded                1       0.47      161          50              50
humaneval-cs-deepseekcoder_v2lite_base-0.2-reworded      1       0.23      158          50              50
humaneval-cs-starcoder2_15b-0.2-reworded                 1       0.32      158          50              50
humaneval-d-deepseekcoder_v2lite_base-0.2-reworded       1       0.16      156          50              50
humaneval-d-starcoder2_15b-0.2-reworded                  1       0.25      156          50              50
humaneval-elixir-deepseekcoder_v2lite_base-0.2-reworded  1       0.00      161          50              50
humaneval-elixir-starcoder2_15b-0.2-reworded             1       0.00      161          50              50
humaneval-go-deepseekcoder_v2lite_base-0.2-reworded      1       0.28      154          50              50
humaneval-go-starcoder2_15b-0.2-reworded                 1       0.26      154          50              50
humaneval-hs-deepseekcoder_v2lite_base-0.2-reworded      1       0.20      156          50              50
humaneval-hs-starcoder2_15b-0.2-reworded                 1       0.17      156          50              50
humaneval-java-deepseekcoder_v2lite_base-0.2-reworded    1       0.36      158          50              50
humaneval-java-starcoder2_15b-0.2-reworded               1       0.40      158          50              50
humaneval-jl-deepseekcoder_v2lite_base-0.2-reworded      1       0.33      159          50              50
humaneval-jl-starcoder2_15b-0.2-reworded                 1       0.32      159          50              50
humaneval-js-deepseekcoder_v2lite_base-0.2-reworded      1       0.45      161          50              50
humaneval-js-starcoder2_15b-0.2-reworded                 1       0.45      161          50              50
humaneval-lua-deepseekcoder_v2lite_base-0.2-reworded     1       0.38      161          50              50
humaneval-lua-starcoder2_15b-0.2-reworded                1       0.44      161          50              50
humaneval-ml-deepseekcoder_v2lite_base-0.2-reworded      1       0.17      155          50              50
humaneval-ml-starcoder2_15b-0.2-reworded                 1       0.24      155          50              50
humaneval-php-deepseekcoder_v2lite_base-0.2-reworded     1       0.43      161          50              50
humaneval-php-starcoder2_15b-0.2-reworded                1       0.34      161          50              50
humaneval-pl-deepseekcoder_v2lite_base-0.2-reworded      1       0.37      161          50              50
humaneval-pl-starcoder2_15b-0.2-reworded                 1       0.38      161          50              50
humaneval-rb-deepseekcoder_v2lite_base-0.2-reworded      1       0.40      161          50              50
humaneval-rb-starcoder2_15b-0.2-reworded                 1       0.42      161          50              50
humaneval-r-deepseekcoder_v2lite_base-0.2-reworded       1       0.31      161          50              50
humaneval-rkt-deepseekcoder_v2lite_base-0.2-reworded     1       0.22      161          50              50
humaneval-rkt-starcoder2_15b-0.2-reworded                1       0.26      161          50              50
humaneval-rs-deepseekcoder_v2lite_base-0.2-reworded      1       0.38      156          50              50
humaneval-rs-starcoder2_15b-0.2-reworded                 1       0.39      156          50              50
humaneval-r-starcoder2_15b-0.2-reworded                  1       0.25      161          50              50
humaneval-scala-deepseekcoder_v2lite_base-0.2-reworded   1       0.39      160          50              50
humaneval-scala-starcoder2_15b-0.2-reworded              1       0.41      160          50              50
humaneval-sh-deepseekcoder_v2lite_base-0.2-reworded      1       0.19      158          50              50
humaneval-sh-starcoder2_15b-0.2-reworded                 1       0.19      158          50              50
humaneval-swift-deepseekcoder_v2lite_base-0.2-reworded   1       0.36      158          50              50
humaneval-swift-starcoder2_15b-0.2-reworded              1       0.34      158          50              50
humaneval-ts-deepseekcoder_v2lite_base-0.2-reworded      1       0.47      159          50              50
humaneval-ts-starcoder2_15b-0.2-reworded                 1       0.43      159          50              50

@arjunguha
Copy link
Member Author

I am not able to get Clojure to work in a Singularity container with the Discovery HTTPS proxy. Getting it to work probably isn't worth it -- most people won't be using Singularity and HTTPS proxies. Anyway, for Discovery, I have a Conda environment conda activate /work/arjunguha-research-group/arjun/conda/clojure that should "just work". Assuming you create a file ~/.m2/settings.xml with this content:

<settings>
  <proxies>
    <proxy>
      <id>example-proxy</id>
      <active>true</active>
      <protocol>https</protocol>
      <host>10.99.0.130 </host>
      <port>3128</port>
      <nonProxyHosts>localhost|127.0.0.1</nonProxyHosts>
    </proxy>
  </proxies>
</settings>

@arjunguha arjunguha changed the title Preparing for next release MultiPL-E 3.0 release Jul 8, 2024
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

No branches or pull requests

1 participant