Skip to content

schmidt-jo/emc_torch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

85 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

emc_torch_package

Package to run Bloch equation simulations using PyTorch. GPU acceleration can be enabled. The implementation is based on Ben-Eliezer et al. 2015 (https://doi.org/10.1002/mrm.25156).

Install

Recommended and tested is to clone git repository. From the repository root, run with a name of your choice:

conda env create -f environment.yml -n <env_name>

conda activate <env_name>

Usage

The Package contains simulation of dictionary databases as well as the fitting of data to such databases.

Simulation

To simulate with configuration found in the "example/simulate" folder run:

python simulate.py

To see all options run

python simulate.py --help

The simulation needs information about Gradients and RF events. The basic inputs can be seen in the "example/simulate/emc_params.json" file. Default configuration will create a database and some plots within the "example/simulate/" folder.

Fit

To fit with a configuration found in the "example/fit" folder run:

python fit.py

To see all options run

python fit.py --help

The fit can take additional input files and needs specification of data and database. The basic inputs can be seen in the "example/fit/fit_config.json" file. Default configuration will use the data within the "examples/fit" folder and b1 regularization.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages