WebCode2M: Webpage Screenshots with Code and Layout Data

WebCode2M Purified provides a real-world dataset for generating code from webpage designs. It pairs webpage screenshots with corresponding HTML/CSS code and layout bounding boxes. The dataset was created by xcodemind and was last updated on October 28, 2024.

Use Cases

• Train models for automated code generation based on webpage screenshots mentioned in the description.
• Develop layout recognition systems using the bounding box and hierarchy information.
• Build tools for converting visual webpage designs into functional HTML/CSS code.
• Research cross-modal understanding between images and text in the context of web development.

Strengths

• Includes multiple aligned data modalities: images, text code, and layout bounding boxes.
• Focuses on real-world webpage designs, as stated in the description.
• Last updated on 2024-10-28, indicating recent maintenance.

Limitations

• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The license is unspecified, which may restrict usage.

Provenance

Source

huggingface

Collection Method

Likely collected from real webpages, pairing rendered screenshots with source code.

Freshness

Last updated 2024-10-28 04:58:15.