amazon-science/cceval

Issues

• Do you have a C++ testing set?

  #23 opened 4 months ago by czq99972
  0

• Where is the code for static analysis?

  #22 opened 7 months ago by ramsey-coding
  2

• How do I specify different Settings in the code generation script? baseline, retrieval, retrieval w/ ref

  #21 opened 8 months ago by kartikzheng
  1

• opensource-codeofconduct@amazon.com is the email for getting raw repo data?

  #19 opened a year ago by yiyepiaoling0715
  1

• Tokenizer used for BM25

  #18 opened a year ago by AaboutL
  1

• Retriever performance evaluation

  #14 opened a year ago by bsobolev
  3

• Is there a way to reproduce the gpt-3.5-turbo results with this repo?

  #5 opened a year ago by madhavatreplit
  11

• The model can complete even if it only sees the contents of the input file

  #4 opened a year ago by 18liumin
  2

• Is current crossfile_context provided correctly?

  #9 opened a year ago by leiwen83
  2

• Regarding relevant data to benchmark the performance of the retriever

  #13 opened a year ago by anmolagarwal999
  1

• It seems only python tree-sitter is used.

  #12 opened a year ago by starlitsky2010
  2

• I can't reproduce the benchmark results in the paper for the model codegen-350

  #10 opened a year ago by starlitsky2010
  9

• How is the data set collected? Can you provide some code?

  #2 opened a year ago by 18liumin
  4

• Is the maximum sequence length of CodeGen2.5 really 4,096?

  #7 opened a year ago by njuweicheng
  2

• Raw code projects of the curated data

  #8 opened a year ago by ganler
  1

• Supporting deepseek coder

  #11 opened a year ago by ganler
  1

• how to implement the RG approach?

  #3 opened a year ago by 18liumin
  1