datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
image-to-code-v2code-image-to-text
Code Snippet Image → Text
A multimodal dataset for fine-tuning vision-language models (VLMs) on the task of
transcribing an image of a code snippet back into its source text — syntax-aware OCR.
Each example pairs a syntax-highlighted PNG of code with the exact code text that
produced it. It spans 8 programming languages and deliberately mixes two capture types:
block — a complete function / unit (6–45 lines).
fragment — a contiguous partial view (3–14 lines) that may start or… See the full description on the dataset page: https://huggingface.co/datasets/anisiraj/code-image-to-text.image-to-code-v1