Skip to content

Ptychography reconstruction with the Bilinear Hessian

Notifications You must be signed in to change notification settings

nikitinvv/BH-ptychography

Repository files navigation

BH-ptychography

Simple demonstration of ptychography reconstruction using the Bilinear Hessian (BH) method. Implementation on GPU using cuPy library

Jupyter Notebooks

demo_object_probe.ipynb - ptychography reconstruction of the object and probe with the BH-CG (Conjugate gradient) method

demo_object_probe_positions.ipynb - ptychography reconstruction of the object, probe, and positions with the BH-CG method

demo_object_probeQN.ipynb - ptychography reconstruction of the object and probe with the BH-QN (Quasi-Newton) method and convergence comparison with the BH-CG

Google Colab

The code can be quickly tested in Google Colab, without installing any packages. See video guidance https://anl.box.com/s/gk9k8fjb7zsmy1qkshvnetdytiotpja0

Installation on a machine with GPU

conda create -n BH-ptychography -c conda-forge cupy notebook pandas

conda activate BH-ptychography

About

Ptychography reconstruction with the Bilinear Hessian

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published