pygid 0.2.7

Fast Preprocessing of Grazing Incidence Diffraction Data

pygid

Fast Preprocessing of Grazing Incidence Diffraction (GID) Data

pygid is a Python-based package for fast conversion of 2D detector images into reciprocal (Cartesian and polar) coordinates. Although the package focuses on the grazing-incidence geometry, it can also be used for transmission (SAXS/WAXS) experiments.

Key features:

• Supports grazing-incidence and transmission geometries.
• Converts area detector images to Cartesian, polar, and pseudopolar coordinates.
• Based on a widely used detector geometry description (PONI files).
• Performs radial and azimuthal profiling.
• Handles single frames, multiple frames, and batch processing.
• Provides a wide range of intensity corrections, including polarization, solid angle, absorption, Lorentz, and detector corrections.
• Utilizes simulation of GIWAXS peak positions using CIF crystal structure files (via pygidSIM).
• Includes experimental and sample metadata management.
• Allows plotting of the conversion results with adjustable parameters.
• Reuses the coordinate maps for several images with the same geometry.
• Supports several interpolation techniques.
• Saves the results of conversion and metadata as a NXsas (NeXus) file.
• Can be used as a first step in the mlgid analysis pipeline.

Input

• Experimental geometry parameters, e.g. a PONI file — see Tutorial 1.

• One of the following:

  • Detector image provided as a 2D NumPy array, a 3D array (with axis 0 representing the image stack axis) or list of 2D arrays.
  • Path to the raw data file(s) (TIFF, EDF, HDF5) — see Tutorial 3.

Output

• Image converted to reciprocal-space coordinates, returned together with the corresponding coordinate axes as NumPy arrays — see Tutorial 4 and Tutorial 5.

• A standardized NeXus (HDF5) file — see File format.
  For example, for cylindrical GID coordinates:

  • img_gid_q — converted image stored as a 3D array (with dimension #0 representing the image stack axis) under /entry/data/img_gid_q
  • q_z — corresponding vertical axis (first dimension) stored as a 1D array under /entry/data/q_z
  • q_xy — corresponding horizontal axis (second dimension) stored as a 1D array under /entry/data/q_xy

• Image exported in a standard format such as PNG, TIFF, or JPEG.

Installation

Install using pip

pip install pygid

Install from source

First, clone the repository:

git clone https://github.com/mlgid-project/pygid.git

Then, to install all required modules, navigate to the cloned directory and execute:

cd pygid
pip install -e .

How to use (short version)

Below is a minimal working example demonstrating how to use pygid to convert 2D detector images to reciprocal space coordinates in a grazing-incidence diffraction (GID) geometry.

1. Download example dataset from Zenodo or set your own files:

import pygid
files = pygid.datasets.get_dataset("tutorial_00")
data_path = files["data"]
poni_path = files["poni"]
mask_path = files["mask"]
yaml_path = files["yaml"]

2. Load experimental parameters from the PONI file — see Tutorial 1

import pygid

params = pygid.ExpParams(
    poni_path=poni_path,        # path to the PONI file
    mask_path=mask_path,        # path to the mask file (EDF/ NPY/ TIFF)
    fliplr=True,                # horizontal flipping of the image
    flipud=True,                # vertical flipping of the image
    transp=False,               # 90 deg rotation of the image
    ai=0.075,                   # angle of incidence in degrees
)

3. Enter sample and experimental metadata — see Tutorial 6

Experimental Metadata (ExpMetadata)

Dictionary-like object containing experiment information:

exp_metadata = pygid.ExpMetadata(
    start_time="2024-03-29T15:51:41.343788",
    end_time="2024-07-12T08:26:22Z",
    source_type="synchrotron",
    source_name="ESRF ID10",
    instrument_name="ID10-surf",
    detector="eiger4m",
    monitor=1.1e5,              # Optional
    extend_fields=["monitor"]   # Fields appended instead of overwritten
)

Sample Metadata (SampleMetadata)

Load sample information from a YAML file:

smpl_metadata = pygid.SampleMetadata(
    path_to_load=yaml_path,
)

4. Create coordinate maps based on geometry and experimental setup — see Tutorial 2

matrix = pygid.CoordMaps(
    params,
    vert_positive=False,        # Cut the positive values for the vertical axis
    hor_positive=False,         # Cut the positive values for the horizontal axis
)

5. Initialize pygid.Conversion instance and load the detector image — see Tutorial 3

analysis = pygid.Conversion(
    matrix=matrix,
    path=data_path,             # path to the data file
    dataset = '/entry_0000/ESRF-ID10/eiger4m/data',
)

6. Perform GID geometry conversion and plot the result, returns the axes and the converted image (list of images) — see Tutorial 4

q_xy, q_z, img = analysis.det2q_gid(
    plot_result=True,                             # plot the result of conversion
    clims=(600, 1e5),                            # image color limits
    save_fig=True, path_to_save_fig="240124_PEN_DIP_polar.png",  # save figure
    return_result=True,                          # return arrays
    save_result=True, path_to_save="240124_PEN_DIP_result.h5",   # save data as a NXsas (NeXus) file
    overwrite_file=False,                        # overwrite the existing file
    smpl_metadata=smpl_metadata,
    exp_metadata=exp_metadata
)

For a detailed description of functionality, follow the tutorials.

Usage examples can be found in the Jupyter Notebook: example/pygid_example.ipynb, and on Zenodo with data collected from different sources.

Citation

Abukaev, A., Völter, C., Romodin, M., Schwartzkopff, S., Bertram, F., Konovalov, O., Hinderhofer, A., Lapkin, D. and Schreiber, F., 2026. pygid: a Python package for fast data reduction in grazing-incidence diffraction. J. Appl. Cryst., 59(1). doi: 10.1107/S1600576725010593