Description

Train AI models with true engineering precision

This dataset is a carefully crafted collection of structured civil and structural engineering knowledge, extracted from the book “Deep Surface: Unlocking N-M Biaxial Interaction in Rectangular Concrete Columns to Eurocode 2”, by Dr. Harshana S. P. Wattage.

Designed for AI and machine learning training, this dataset provides detailed coverage of:

• Eurocode 2 column biaxial bending theory.
• Structural behaviour of reinforced rectangular concrete columns.
• 2D and 3D N–M interaction principles.
• Material properties, stress-strain models, and axial-moment behaviour.
• High-end specialized AI training dataset.

Included:

• Fully structured JSONL dataset (DeepSurface.jsonl).
• 61+ high-quality detailed technical illustrations and cover image (/images/).
• Extracted formulas (LaTeX-ready).
• Clean tabular data.
• Full eBook reference (“Deep Surface” PDF copy).
• Comprehensive metadata (Dataset_card.json).
• 162 highly curated, fully complete AI-ready samples (sections).

Key Benefits:

• High-quality academic and professional engineering content.
• Fine-tuned for internal AI training or specialized model creation.
• Directly embeds Eurocode 2 standards knowledge.
• Perfect for LLMs, fine-tuning, and multimodal AI (text + images).

License Terms:

• Commercial AI training use permitted.
• Redistribution, resale, or public reproduction of dataset or eBook prohibited.
• No refunds (due to digital nature of product).

Format
Each entry in DeepSurface.jsonl is a JSON object with the following fields:

• source_id — Unique section identifier
• section_title — Title of the section
• content — Main text content (plain text + minimal LaTeX)
• formulas_latex — List of LaTeX-formatted formulas
• tables — Structured tabular data (if present)
• figures — Linked figure images (if present)
• Metadata fields (author, book title, publication year, etc.)

File Structure

Ideal for AI engineers, technical researchers, universities, corporate AI training teams, and structural simulation projects.

File Size: 22.2 MB

Published: May/2025

License file and full documentation included.

Empower your AI models with real engineering expertise.

Free Preview: JSONL Format (Includes First 12 Sections)