quick_pp

A Python package for quick-look preliminary petrophysical estimations.

Installation

You can install quick_pp directly from PyPI:

pip install quick_pp

For development or to use the qpp_assistant, you'll need to clone the repository and install dependencies:

1. Clone the repository:

   git clone https://github.com/imranfadhil/quick_pp.git
   cd quick_pp

2. Create and activate a virtual environment (tested with Python 3.11):

   uv venv --python 3.11
   source .venv/bin/activate  # On Windows, use: .venv\Scripts\activate

3. Install the required packages:

   uv pip install -r requirements.txt

Quick Start

Jupyter Notebook Examples

More structured analysis/ examples are done in https://github.com/imranfadhil/pp_portfolio

The included notebooks demonstrate the core functionalities:

• 01_data_handler: Create a MOCK qppp project file.
• 02_EDA: Perform a quick exploratory data analysis.
• 03_*: Carry out petrophysical interpretation of the MOCK wells.

Note: For the API notebook, you need to run python main.py app before executing the cells.

qpp_assistant Setup

To use the qpp_assistant, follow these steps after the development installation:

1. Specify the required credentials in a .env file (you can use .env copy as a template).
2. Run Docker Compose: docker-compose up -d.
3. Build your flow in Langflow at http://localhost:7860.
4. Run the main application: python main.py app.
5. Test your flow in the qpp Assistant at http://localhost:8888/qpp_assistant.

CLI

Train a Machine Learning Model

Requirements:

• The input data must be a Parquet file located at /data/input/<data_hash>___.parquet.
• The Parquet file must contain the input and target features as specified in MODELLING_CONFIG in config.py.

Command:

quick_pp train <model_config> <data_hash>

quick_pp train mock mock

Run the MLflow Server

Command:

quick_pp mlflow-server

You can access the MLflow UI at http://localhost:5015.

Run Predictions

Note: Trained models must be registered in MLflow before running predictions.

quick_pp predict <model_config> <data_hash>

Example:

    quick_pp predict mock mock

Deploy Trained Models as an API

quick_pp model-deployment

You can access the deployed model's Swagger UI at http://localhost:5555/docs.

Start the Main Application

quick_pp app

• API Docs: http://localhost:8888/docs
• qpp_assistant: http://localhost:8888/qpp_assistant (you can log in with any username and password).

To use the mcp tools, you would need to first add the following SSE URLS through the interface; http://localhost:8888/mcp - quick_pp tools.

http://localhost:5555/mcp - quick_pp ML model prediction tools (need to run quick_pp model-deployment first).

Documentation

Documentation is available at: https://quick-pp.readthedocs.io/en/latest/index.html