pybhu 0.1.5

Some useful functions for Data Analysis

pybhu

pybhu currently exposes a desktop scientific image viewer for 2D and 3D array data. This README is intentionally focused on the public img_viewer interface.

Highlights

• Interactive PyQt6 viewer for 2D and 3D image stacks
• Support for .npy, .npz, and .pkl datasets
• Layer navigation, colormap selection, histogram-based contrast control, colorbar display, and FFT analysis

Repository Structure

.
|-- __init__.py
|-- img_viewer/
|   |-- api.py
|   |-- viewer.py
|   |-- loader.py
|   |-- state.py
|   |-- histogram_dialog.py
|   |-- colorbar_dialog.py
|   |-- fft.py
|   `-- fft_dialog.py
`-- README.md

Public Interface

img_viewer

The main application in this repository is a PyQt6 + Matplotlib scientific viewer for array-based data. It is designed for quick inspection of 2D images and 3D stacks, with built-in tools for:

• browsing layers in a stack
• switching colormaps and inverting palettes
• adjusting intensity limits from a histogram window
• displaying a separate colorbar window
• computing forward or inverse 2D FFTs with selectable window functions

Requirements

The repository currently ships as source code and does not yet include packaging metadata such as pyproject.toml. Install dependencies manually before use.

The examples below assume the package is available as pybhu, either because the project folder is named pybhu or because you install/package it that way.

Core viewer dependencies:

pip install numpy scipy matplotlib pyqt6

Quick Start

Launch the image viewer from a NumPy array

import numpy as np
from pybhu import img_viewer

data = np.random.rand(256, 256, 16)
viewer = img_viewer(data, colormap_name="viridis", initial_layer=0)
viewer.app.exec()

If you are already inside an environment that owns the Qt event loop, you may not need to call viewer.app.exec().

Open data directly from file

from pybhu import img_viewer

viewer = img_viewer("example_stack.npz")
viewer.app.exec()

Supported viewer file types:

• .npy
• .npz
• .pkl

For .npz and .pkl, the loader searches recursively for 2D or 3D arrays and lets you choose among multiple datasets when needed.