MIRTorch
A differentiable PyTorch toolbox for medical-image reconstruction, developed at the University of Michigan. MIRTorch provides composable linear maps, proximal operators, iterative solvers, and MRI and SPECT system models.
Documentation · Examples · API
New
Non-Cartesian and B0-informed MRI now use FINUFFT on supported non-macOS CPU systems and cuFINUFFT on CUDA when installed. In warm NVIDIA A10 benchmarks, the new paths measured up to 5.2× faster NUFFT, 9.4× faster Toeplitz normal operations, and 12.3× faster iterative solvers. These are workload-specific measurements, not universal speedups.
Install
Install PyTorch for your platform, then:
pip install MIRTorch
CUDA users can install cuFINUFFT with:
pip install "MIRTorch[cufinufft]"
For local development:
pip install -e ".[dev]"
Backends and compilation
NuSense, NuSenseGram, Gmri, and GmriGram use an installed FINUFFT or
cuFINUFFT library when the device supports it, then fall back to torchkbnufft.
Base macOS, Apple Metal, and Linux ARM installs therefore work without a
native library. Set backend="torchkbnufft" or backend="finufft" to
override the automatic choice.
Real-valued CUDA runs of Diff2dgram, FISTA, and POGM use torch.compile
automatically when PyTorch provides it. Other inputs stay eager; pass
compile=False to disable compilation explicitly.
Examples
The notebooks choose CUDA, Apple Metal, or CPU at runtime:
demo_mri.ipynb: CG-SENSE and B0-informed PWLSdemo_3d.ipynb: 3D non-Cartesian MRI and Toeplitz embeddingdemo_cs.ipynb: compressed-sensing MRIdemo_mlem.ipynb: SPECT reconstructiondemo_mnist.ipynb: CG, FISTA, and POGMdemo_dl.ipynb: dictionary learning
Citation and acknowledgments
MIRTorch is inspired by MIRT, MIRT.jl, SigPy, and PyLops. See the documentation for the MIRTorch, BJORK, and SPECT citations.
MIRTorch is distributed under the BSD 3-Clause License.
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mirtorch-0.2.0.tar.gz.
File metadata
- Download URL: mirtorch-0.2.0.tar.gz
- Upload date:
- Size: 83.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.14
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
6d9342405063128b88581779da902d67ef864d3ba725ea22dd0fd4e8d7e31603
|
|
| MD5 |
bd104a81c1c0215b49a6f80c016a4653
|
|
| BLAKE2b-256 |
bf96916fe89b3127dfe4b1e9caa3f38a38c4e1382858e2944deb7d5c123b884e
|
Provenance
The following attestation bundles were made for mirtorch-0.2.0.tar.gz:
Publisher:
python-release.yml on guanhuaw/MIRTorch
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
mirtorch-0.2.0.tar.gz -
Subject digest:
6d9342405063128b88581779da902d67ef864d3ba725ea22dd0fd4e8d7e31603 - Sigstore transparency entry: 2266406735
- Sigstore integration time:
-
Permalink:
guanhuaw/MIRTorch@f8857eceb9c3da13851718313dea331f8bc033de -
Branch / Tag:
refs/heads/master - Owner: https://github.com/guanhuaw
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
python-release.yml@f8857eceb9c3da13851718313dea331f8bc033de -
Trigger Event:
push
-
Statement type:
File details
Details for the file mirtorch-0.2.0-py3-none-any.whl.
File metadata
- Download URL: mirtorch-0.2.0-py3-none-any.whl
- Upload date:
- Size: 100.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.14
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
2e106c297716bbf34b6b78b40db4e4d758ad1d060898d89107f1ef39c1fdc794
|
|
| MD5 |
8aa755583b33fa2502e11be10486ba97
|
|
| BLAKE2b-256 |
293b7aaa421dafafe594570107044deb55ea4c515dc459f632d4a42ee24fd634
|
Provenance
The following attestation bundles were made for mirtorch-0.2.0-py3-none-any.whl:
Publisher:
python-release.yml on guanhuaw/MIRTorch
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
mirtorch-0.2.0-py3-none-any.whl -
Subject digest:
2e106c297716bbf34b6b78b40db4e4d758ad1d060898d89107f1ef39c1fdc794 - Sigstore transparency entry: 2266406975
- Sigstore integration time:
-
Permalink:
guanhuaw/MIRTorch@f8857eceb9c3da13851718313dea331f8bc033de -
Branch / Tag:
refs/heads/master - Owner: https://github.com/guanhuaw
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
python-release.yml@f8857eceb9c3da13851718313dea331f8bc033de -
Trigger Event:
push
-
Statement type:
