Python ecosystem functionality will seamlessly integrate with the application’s data stores and viewers, eliminating the need to export, reformat and reload data. With Python ecosystems, users can process data with pre-built or client-proprietary Python scripts or Jupyter notebooks, and store input and output data in either a HampsonRussell or Jason project database or a shared directory. HampsonRussell and Jason users, even those with limited expertise in machine learning or Python scripting, will now benefit from complete control over input data and analysis output. G&G experts and data scientists can use Ecosystem workflows pre-built by CGG or they can build their own new reservoir characterization workflows using the latest open source machine learning packages, such as Google’s TensorFlow. Scripts and workflows directly access well, horizon and seismic data for use in machine learning, deep learning, visualization and numerical analysis. Python ecosystems allow users to efficiently research and test various state-of-the-art machine learning workflows for proof-of-concept or commercial projects. Already attracting considerable industry interest in GeoSoftware’s PowerLog petrophysical software, Python ecosystems in HampsonRussell and Jason will let experts and data scientists completely customize machine learning and reservoir characterization workflows by using extensively available Python machine learning libraries and also their own proprietary code.

Back to Newsroom CGG GeoSoftware Adds Machine Learning Applications for Reservoir Characterization Using Python Ecosystem TechnologyĬGG GeoSoftware, part of CGG’s Geoscience division, has announced that machine learning technology in Python ecosystems will be available in upcoming releases of its flagship HampsonRussell and Jason reservoir characterization solutions.

Cgg hampson russell software