Rock physics templates, well-log QC and machine learning — right in your browser
Load data, QC it, generate synthetic logs and train a neural network — no install, no code.
Try it in one click:
What’s included & how the examples work
- Built-in Vp, pressure & well-log datasets
- QC plot, pairplot, heatmap, statistics
- Neural-network train & predict
Brine-sand Vp: Vp vs depth for 0–100% clay. Pore pressure: hydrostatic to 17 ppg. Well logs + lithology: GR & DTC QC, ready for NN. Each example loads a built-in dataset into the table below and draws a QC plot; then try the QC and Transform tools. You can also load a template or your own LAS/CSV data, generate synthetic logs from formula rules and train a neural network to predict a property.
Input Data
Edit the table, load a built-in dataset or import your own file
Enter your data manually by editing the table, or load predefined datasets using the "Load RPT Data" button. You can also import .csv or .las file using "Import Data" button.
| Depth | Rock_Property |
|---|---|
QC Data
Quality-control plots for the table above
Visualize your input data with quality control plots. Choose from QC Plot, Pairplot, Heatmap, Statistics or Missing Data QC to inspect data relationships and distributions.
Transform Data
Clean, synthesize or predict with a neural network
Apply transformations to your data for further analysis. You can remove missing data, create synthetic data, or use advanced techniques like neural network-based transformations.
How to reuse the last trained model
This model can be applied to new dataset for prediction purpose. In "Transform with Neural Networks", select the same input features as the last trained model and click "Predict with last model".
No metadata available.
