The dataset description:

• The dataset is composed by submissions to Leetcode's contests.
  • This data is of public access.
• For each contest, the first accepted solution for some popular languages is taken
  • In this way, we can be sure that the source code is warranted to be correct.
• In total, there are 4666 unique submissions and 916 unique problems:
  • Up to contest weekly-348.
  • Up to contest biweekly-105
  • The idea is to update keep it updated with latest contests.
  • This readme won't probably be updated when that happens.
• The included languages are:
  • C, C++, Java, JavaScript, Python 2, Python 3, Rust.
• The dataset has been cleaned to ensure formatting consistency and the correctness of the present numerical values.
• The dataset is in instruction, input, and output JSON format.
• The files have been updated to the jsonlines format

Files:

• submissions.jsonl: Contains the submissions as explained above.

Deprecated File: (This files may or not may exist on the repo)

• submissions_py.json: Only python submissions, it doesn't follow the Llama format, 10 submissions per problem.
• submissions_py_llama.json: Same as above but following the Llama input format
• submissions_py_llama_small.json: Same as above but with 3 submissions per problem.