datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
doocs-leetcode-solutions
Doocs LeetCode Solutions
A comprehensive dataset of LeetCode problems and solutions created from the Doocs LeetCode repository. This dataset is designed for fine-tuning large language models to understand programming problems and generate code solutions.
Description
Repository: Doocs LeetCode Solutions
Total Problems: 3500+
Total Solutions: 15,000+ (across multiple languages)
Size: ~60 MB (Parquet format)
Languages:
C
Cangjie
C++
C#
Dart
Go
Java
JavaScript
Kotlin
Nim
PHP… See the full description on the dataset page: https://huggingface.co/datasets/olegshulyakov/doocs-leetcode-solutions.leetcode-python-solutions-with-exaplanationsleetcode-assembly
LeetCode Assembly Dataset
441 LeetCode problems solved in C, compiled to assembly across 4 architectures, 2 compilers, and 4 optimization levels using GCC and Clang via the Godbolt Compiler Explorer API.
Dataset Summary
Stat
Value
Total rows
14,112
Unique problems
441
Architectures
x86-64, AArch64, MIPS64, RISC-V 64
Compilers
GCC 15.2, Clang 21.1.0
Optimization levels
-O0, -O1, -O2, -O3
Compilation success rate
100%
Difficulty split
Easy: 98… See the full description on the dataset page: https://huggingface.co/datasets/ronantakizawa/leetcode-assembly.task956_leetcode_420_strong_password_check
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task956_leetcode_420_strong_password_check
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task956_leetcode_420_strong_password_check.leetcode-codegen-python
LeetCode Code-Gen Dataset — Python
2522 rows. Given a problem statement, its input/output examples, and a
required algorithm/technique, generate a correct Python solution.
Part of a 4-language collection built from the same source: see the sibling
Python,
Java,
C++, and
JavaScript
datasets.
Verification
Every row was extracted, then executed in a sandboxed subprocess against the problem's own stated examples. Only rows that passed all examples are included -- this… See the full description on the dataset page: https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-python.leetcode2000-rl
LeetCode2000-RL
LeetCode2000-RL contains 2000 LeetCode programming tasks with one Python solution per task, normalized correctness test cases, and speed-test inputs with measured runtimes. It is a curated derivative of newfacade/LeetCodeDataset. New LLM-generated solutions and runtime-focused test cases were used to benchmark candidate solutions, replace slower solutions when faster correct variants were found, and select the final solution for each task. The final dataset was then… See the full description on the dataset page: https://huggingface.co/datasets/avidanborisov/leetcode2000-rl.leetcode-codegen-javascript
LeetCode Code-Gen Dataset — JavaScript
631 rows. Given a problem statement, its input/output examples, and a
required algorithm/technique, generate a correct JavaScript solution.
Part of a 4-language collection built from the same source: see the sibling
Python,
Java,
C++, and
JavaScript
datasets.
Verification
Not execution-verified. There is currently no compiler/runtime harness for this language in the build pipeline (only Python has one). Rows are extracted… See the full description on the dataset page: https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-javascript.leetcode-codegen-java
LeetCode Code-Gen Dataset — Java
4068 rows. Given a problem statement, its input/output examples, and a
required algorithm/technique, generate a correct Java solution.
Part of a 4-language collection built from the same source: see the sibling
Python,
Java,
C++, and
JavaScript
datasets.
Verification
Not execution-verified. There is currently no compiler/runtime harness for this language in the build pipeline (only Python has one). Rows are extracted directly from… See the full description on the dataset page: https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-java.leetcode-codegen-cpp
LeetCode Code-Gen Dataset — C++
4025 rows. Given a problem statement, its input/output examples, and a
required algorithm/technique, generate a correct C++ solution.
Part of a 4-language collection built from the same source: see the sibling
Python,
Java,
C++, and
JavaScript
datasets.
Verification
Not execution-verified. There is currently no compiler/runtime harness for this language in the build pipeline (only Python has one). Rows are extracted directly from… See the full description on the dataset page: https://huggingface.co/datasets/AmareshHebbar/leetcode-codegen-cpp.LeetCode_Python_SolutionsLeetCode_with_Solutionsdatasets:
[LimYeri/LeetCode_YT_CC_CoT_Summary] (https://huggingface.co/datasets/LimYeri/LeetCode_YT_CC_CoT_Summary)
[kreimben/leetcode_user_submissions] (https://huggingface.co/datasets/kreimben/leetcode_user_submissions)
[greengerong/leetcode] (https://huggingface.co/datasets/greengerong/leetcode)
leetcode_with_youtube_captionsleetcode_contests_unique_solutions
Dataset Card for "unique_solutions_leetcode_contests"
Dataset Summary
The purpose of this dataset is to provide a robust problem-solving foundation for Large Language Models (LLMs).
The dataset comprises top accepted solutions to Leetcode's contests.
For each contest problem and language, up to 15 submissions are chosen to be included in the dataset.
The dataset has been meticulously cleaned to ensure consistent formatting, accuracy of the numerical
values… See the full description on the dataset page: https://huggingface.co/datasets/Nan-Do/leetcode_contests_unique_solutions.LeetCodeDataset_rectifiedUsing o4-mini to rectify the typos or missing information during data curation (https://github.com/newfacade/LeetCodeDataset/issues/1).
One can use the field problem_description_corrected (but you should check to make sure it's not None or an empty string ""; very few cases always went wrong.) to replace the problem_description
Warning: I haven't manually checked all the corrected texts, so the quality is not fully guaranteed.
LeetCode_YouTube_CCLeetCode Information & YouTube Captions
Original data -> LimYeri/leetcode_with_youtube_captions
The original ['cc_content'] column had many repeated sentences, making the data too long.
To remove the repetitions, we used precise regular expressions to eliminate the repeated sentences. -> new column ['content']
Additionally, we also removed unnecessary strings (e.g., '[Music]').
vgrout-leetcode-teacher-demos
vGROUT LeetCode teacher demonstrations
Cached teacher demonstrations used to warm up the
vGROUT gradient-routing experiments on the
ariahw/rl-rewardhacking LeetCode
environment. Each row is a full problem-specific completion. The kind column gives the
two demonstration types:
hack (215 rows): verified exploits of the run_tests loophole (hacked=True,
gt_pass=False).
solve (126 rows): correct solutions verified against the ground-truth tests
(gt_pass=True).
Why fewer… See the full description on the dataset page: https://huggingface.co/datasets/wassname/vgrout-leetcode-teacher-demos.LeetCode_YT_CC_CoT_SummaryLeetCode Information & YouTube Captions with CoT Summaries
Original data -> LimYeri/leetcode_with_youtube_captions
The original ['cc_content'] column had tokens that were too long and contained a lot of repetition, which necessitated summarization.
Consequently, our team (Project Team: CodeMind) summarized the ['cc_content'] column data using the Chain of Thought (CoT) technique with the gpt-3.5-turbo-0125 & gpt-4-turbo-2024-04-09 model. -> new column ['Summary']