Skip to main content

Get Started

Welcome to fpm-py! For a full walkthrough, please refer to the User Guide and API Reference. The steps below summarize the key features.

1. Installation​

Install the library using pip:

pip install fpm-py

2. Basic Usage​

Start by importing the necessary components:

import fpm_py as fpm
from fpm_py import ImageCapture, ImageSeries

3. Creating an Image Series​

Load your low-resolution images and their corresponding k-space vectors:

captures = [
    ImageCapture(image=img1, k_vector=k1),
    ImageCapture(image=img2, k_vector=k2),
    # Add more captures as needed
]

series = ImageSeries(captures=captures, magnification=4.0, pixel_size=6.5)

4. Reconstructing High-Resolution Images​

Use the reconstruct() function to generate a high-resolution image:

output = fpm.reconstruct(series)

5. Customization Options​

  • Iteration Terminators: Control when the reconstruction stops.
  • Optimizers: Customize how the image and pupil are updated.

For example, to set a custom iteration terminator:

output = fpm.reconstruct(series, iteration_terminator=lambda inputs: iter_ceil(inputs, max_iter=20))

6. Parallelism​

The algorithm automatically detects if a GPU is available and utilizes it for faster computations.

7. Explore Further​

This QuickStart covers the basics. For more detailed instructions and advanced features, refer to the full user guide.