PMRF Teachings
PMRF Teachings
A part of knowledge dissemination
Teaching a 12 weeks course on "Practical implementation of deep neural networks and transfer learning concept" .
This course is formulated to provide the insight and understanding to the participates about deep neural networks.
Along with an understanding of how we can anticipate the requirement of large training dataset to effectively implement these models into sciences and engineering problems having limited training data
With the practical demonstrations are given, starting from creating an environment in Anaconda to the application of popular python libraries for building various machine learning algorithms/architecture such as; ANN, CNN, RNN etc.
Taught a 12 weeks course on "A practical approach to seismic method using python and machine learning". This course is designed to provide the theoretical background of seismic method and its inversion process.
Along with the practical demonstrations are given, starting from creating an environment in Anaconda to the application of popular python libraries in seismic and well-logging method.
Took problem-solving sessions for the students registered in the above course.
During the sessions, topics related to the content of the course and assignments were discussed so that students could interact and ask their queries and doubts related to the course.
Basics of the seismic method
Seismic inversion
Seismic inversion techniques
ML in seismic inversion
Role of optimizers
Introduction to useful libraries for seismic & installation
Loading Seismic data in python & visualization
Well data plotting
Scatter 2D plot for well data to analyse trend/pattern
Scatter 3D plot for well data to analyse trend/pattern
To check the PMRF teaching videos, please click here
Taking sessions on "Application of artifical intelligence and machine learning" to the undergraduate and postgraduate students.
This course is blend of therotical understnading of machine learning techniques and its practical implementation of through lab sessions.
The lab sessions are based on building an artifical neural network (ANN) using the tensorflow, which includes python libraries installation, data pre-processing, designing architecture of ANN, its training and validation.