datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
adaption-python-code-generation-tasks
OSS-Instruct Coding Tasks (Augmented)
Coding problems inspired by open-source snippets, with solutions across several languages.
Rows
24,952
Domain
programming
Format
data.parquet, one row per example
Licence
other
Built for
supervised fine-tuning (SFT) experiments on Adaption AutoScientist
Columns
Column
Description
prompt
The prompt (user turn) as uploaded.
completion
The target response as uploaded.
enhanced_prompt… See the full description on the dataset page: https://huggingface.co/datasets/rodriguescarson/adaption-python-code-generation-tasks.adaption-python-code-generation-tasks-aug-81a8e6
OSS-Instruct Coding Tasks (Augmented)
Coding problems inspired by open-source snippets, with solutions across several languages.
Rows
24,952
Domain
programming
Format
data.parquet, one row per example
Licence
other
Built for
supervised fine-tuning (SFT) experiments on Adaption AutoScientist
Columns
Column
Description
original_prompt
The prompt (user turn) as uploaded.
original_completion
The target response as uploaded.… See the full description on the dataset page: https://huggingface.co/datasets/rodriguescarson/adaption-python-code-generation-tasks-aug-81a8e6.adaption-python-code-generation-tasks-aug
OSS-Instruct Coding Tasks (Augmented)
Coding problems inspired by open-source snippets, with solutions across several languages.
Rows
11,981
Domain
programming
Format
data.parquet, one row per example
Licence
other
Built for
supervised fine-tuning (SFT) experiments on Adaption AutoScientist
Columns
Column
Description
original_prompt
The prompt (user turn) as uploaded.
original_completion
The target response as uploaded.… See the full description on the dataset page: https://huggingface.co/datasets/rodriguescarson/adaption-python-code-generation-tasks-aug.Python_Code_Generationpython-code-generation-syntheticcf-cpp-to-python-code-generation
Dataset
The cf-llm-finetune uses a synthetic parallel dataset built from the Codeforces submissions and problems. C++ ICPC-style solutions are filtered, cleaned, and paired with problem statements to generate Python translations using GPT-4.1, creating a fine-tuning dataset for code translation.
The final dataset consists of C++ solutions from 2,000 unique problems, and synthetic Python answers, split into train (1,400), validation (300), and test (300) sets. For details on dataset… See the full description on the dataset page: https://huggingface.co/datasets/demoversion/cf-cpp-to-python-code-generation.