Nicely Log Measurement Prediction Utilizing Neural Networks with Keras | by Andy McDonald | Oct, 2023
An instance of predicting bulk density (RHOB) with Keras and illustrating impacts of normalisation on prediction outcomes
21 hours in the past
Massive quantities of information are acquired each day from wells around the globe. Nonetheless, the standard of that information can range considerably from lacking information to information impacted by sensor failure and borehole circumstances. This may have knock-on penalties on different elements of a subsurface venture, similar to delays and inaccurate assumptions and conclusions.
As lacking information is without doubt one of the most typical points we face with properly log information high quality, quite a few strategies and methods have been developed to estimate values and fill within the gaps. This consists of the applying of machine studying know-how — which has elevated in reputation over the previous few a long time with libraries similar to TensorFlow and PyTorch.
On this tutorial, we can be utilizing Keras, which is a high-level neural networks API that runs on high of TensorFlow. We’ll use it for example the method of constructing a machine-learning mannequin to permit predictions of bulk density (RHOB). This can be a generally acquired logging measurement, nonetheless, it may be considerably impacted by unhealthy gap circumstances or, in some circumstances, instruments can fail, leading to no measurements over key intervals.
We’ll begin with a quite simple mannequin, that doesn’t account for normalising the inputs, a typical step within the machine studying workflow. Then, we’ll then construct a second mannequin with normalised inputs and illustrate its influence on the ultimate prediction outcome.
Step one on this tutorial is to import the libraries we can be working with.
For this tutorial, we want 4 libraries:
These are imported as follows:
import pandas as pd
from…
