This repository contains the code and resources for a conceptual project demonstrating the integration of multimodal geological data (text reports, core images, and well logs) for enhanced subsurface characterization using machine learning.

The workflow covers:

1. Text Data Processing: Extracting key information from geological reports (conceptual using NER).
2. Image Analysis: Conceptually processing core photos for facies classification (e.g., Sandstone/Carbonate).
3. Well Log Data Analysis: Loading, visualizing, and engineering features from synthetic well log data.
4. Multimodal Data Integration: Combining information from all sources into a unified dataset.
5. Machine Learning: Developing a conceptual ML model for lithology prediction based on the integrated data.

• data/synthetic_well_log_1.csv: Synthetic well log data used for demonstration.
• mini_reports/: Example mini reports (conceptual text data).
• images/: Placeholder for core images and generated plots.

1. Clone the repository:

   git clone [https://github.com/YourUsername/geological-data-integration.git](https://github.com/YourUsername/geological-data-integration.git)
   cd geological-data-integration

2. Create a virtual environment (recommended):

   python -m venv venv
   # On Windows:
   .\venv\Scripts\activate
   # On Linux/macOS:
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the main script:

   python main.py

See requirements.txt for a list of Python packages.

Emmanouil Amygdalas / mamigdalas