CubicalRipser: Persistent Homology for 1D Time Series, 2D Images, and 3D/4D Volumes

Authors: Takeki Sudo, Kazushi Ahara (Meiji University), Shizuo Kaji (Kyoto University)

CubicalRipser is an adaptation of Ripser by Ulrich Bauer, specialized in fast computation of persistent homology for cubical complexes.

Overview

Key Features

  • Support cubical complexes up to 4D
  • High performance for 3D
  • C++ command-line binaries and Python modules
  • PyTorch integration for differentiable workflows
  • Utility helpers for loading, plotting, and vectorization
  • Filtrations with both V- and T-constructions
  • Binary coefficients (field F2)
  • Optional creator/destroyer locations in outputs

Citation

If you use this software in research, please cite:

@misc{2005.12692,
  author = {Shizuo Kaji and Takeki Sudo and Kazushi Ahara},
  title = {Cubical Ripser: Software for computing persistent homology of image and volume data},
  year = {2020},
  eprint = {arXiv:2005.12692}
}

Contents

Getting Started

Try Online

Quickstart (Python)

pip install -U cripser

import numpy as np
import cripser

arr = np.load("sample/2d_hole.npy")
ph = cripser.compute_ph(arr, filtration="V", maxdim=2)
print(ph[:5])

Quickstart (CLI)

Build binaries first (see Installation), then:

./build/cubicalripser --maxdim 2 --output out.csv sample/3dimsample.txt
./build/tcubicalripser --maxdim 2 --output out_t.csv sample/3dimsample.txt

Installation

Using pip (recommended)

pip install -U cripser

If wheel compatibility is an issue on your platform:

pip uninstall -y cripser
pip install --no-binary cripser cripser

Building from source

Requirements:

  • Python >= 3.9 (Python 3.8 is no longer supported)
  • CMake >= 3.21
  • C++17+ compiler (GCC, Clang, MSVC; src/Makefile defaults to C++20)
  • nanobind (installed automatically as a build dependency via pyproject.toml)

Clone the repository:

git clone https://github.com/shizuo-kaji/CubicalRipser.git
cd CubicalRipser

Build CLI binaries:

cmake -S . -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build -j

Outputs:

  • build/cubicalripser (V-construction)
  • build/tcubicalripser (T-construction)

Install the Python package from source:

pip install .

Legacy alternative (from src/):

cd src
make all

Python Usage

CubicalRipser works on 1D/2D/3D/4D NumPy arrays (dtype convertible to float64).

Core APIs

  • cripser.computePH(...): low-level binding
  • cripser.compute_ph(...): convenience wrapper (converts essential deaths to np.inf)

Both support:

  • filtration="V" or filtration="T"
  • maxdim, top_dim, embedded, location

Example (V-construction):

import numpy as np
import cripser

arr = np.load("sample/4d_hole.npy")
ph = cripser.compute_ph(arr, filtration="V", maxdim=3)

Output format

For 1D-3D input, each row is typically:

dim, birth, death, x1, y1, z1, x2, y2, z2

For 4D input:

dim, birth, death, x1, y1, z1, w1, x2, y2, z2, w2

See Creator and Destroyer Cells for interpretation of coordinates.

Notes:

  • computePH(...) and CLI may represent essential deaths as DBL_MAX.
  • compute_ph(...) converts essential deaths to np.inf.

GUDHI conversion helpers

dgms = cripser.to_gudhi_diagrams(ph)
persistence = cripser.to_gudhi_persistence(ph)

GUDHI plotting example:

import gudhi as gd
gd.plot_persistence_diagram(diagrams=dgms)

Differentiable PyTorch wrapper

import torch
import cripser

x = torch.rand(32, 32, requires_grad=True)
ph = cripser.compute_ph_torch(x, maxdim=1, filtration="V")
loss = cripser.finite_lifetimes(ph, dim=0).sum()
loss.backward()

The gradient is propagated through birth/death values to creator/destroyer voxel locations. Pairing changes are discrete, so the gradient is piecewise-defined.

Additional utilities

  • plotting: cripser.plot_diagrams(...) (requires matplotlib)
  • vectorization: cripser.persistence_image(...), cripser.create_PH_histogram_volume(...)
  • OT distance: cripser.wasserstein_distance(...) (requires torch and POT/ot)

Helper Python script (demo/cr.py)

A convenience wrapper for quick experiments without writing Python code.

Typical capabilities:

  • Accepts a single file (.npy, image, DICOM, etc.) or a directory of slices
  • Builds 1D-4D arrays from files
  • Chooses V- or T-construction
  • Computes PH up to a chosen max dimension
  • Writes CSV or .npy outputs
  • Optional sorting for DICOM/sequence inputs

Help:

python demo/cr.py -h

Examples:

# single NumPy array
python demo/cr.py sample/2d_hole.npy -o ph.csv

# increase max dimension, use T-construction
python demo/cr.py sample/bonsai128.npy -o ph.csv --maxdim 3 --filtration T

# directory of DICOM files (sorted)
python demo/cr.py dicom/ --sort -it dcm -o ph.csv

# directory of PNG slices
python demo/cr.py slices/ -it png -o ph.csv

# save PH as NumPy for reuse
python demo/cr.py sample/bonsai128.npy -o ph.npy

# invert intensity sign
python demo/cr.py sample/bonsai128.npy -o ph.csv --negative

Selected options (see -h for full list):

  • --maxdim k
  • --filtration V|T
  • --sort
  • -it EXT
  • -o FILE
  • --embedded
  • --negative
  • --transform ..., --threshold ..., --threshold_upper_limit ...

Command-Line Usage

Basic examples

Perseus-style text input:

./build/cubicalripser --print --maxdim 2 --output out.csv sample/3dimsample.txt

NumPy input (1D-4D):

./build/cubicalripser --maxdim 3 --output result.csv sample/bonsai128.npy

T-construction:

./build/tcubicalripser --maxdim 3 --output volume_ph.csv sample/bonsai128.npy

Common options (cubicalripser --help)

  • --maxdim, -m <k>: compute up to dimension k (default 3)
  • --threshold, -t <value>: threshold for births
  • --print, -p: print pairs to stdout
  • --embedded, -e: Alexander dual interpretation
  • --top_dim: top-dimensional computation using Alexander duality
  • --location, -l yes|none: include or omit creator/destroyer coordinates
  • --algorithm, -a link_find|compute_pairs: 0-dimensional PH method
  • --cache_size, -c <n>: cache limit
  • --min_recursion_to_cache, -mc <n>: recursion threshold for caching
  • --output, -o <FILE>: write .csv, .npy, or DIPHA-style persistence binary
  • --verbose, -v

Notes:

  • CLI does not use --filtration; V/T are separate binaries.
  • Use --output none to suppress output file creation.

Input Formats

Supported input formats (CLI)

Image-to-array conversion (demo/img2npy.py)

A helper utility converts images/volumes between multiple formats.

# image -> .npy
python demo/img2npy.py image.jpg output.npy

# image series glob -> volume .npy (shell expansion)
python demo/img2npy.py input*.jpg volume.npy

# explicit files -> volume .npy
python demo/img2npy.py input00.dcm input01.dcm input02.dcm volume.npy

DICOM volume conversion:

python demo/img2npy.py dicom/*.dcm output.npy

Direct DICOM PH computation:

python demo/cr.py dicom/ --sort -it dcm -o output.csv

DIPHA conversions:

# NumPy -> DIPHA complex
python demo/img2npy.py img.npy img.complex

# DIPHA complex -> NumPy
python demo/img2npy.py img.complex img.npy

# DIPHA persistence output (.output/.diagram) -> NumPy
python demo/img2npy.py result.output result.npy

1D time series

A scalar time series can be treated as a 1D image, so CubicalRipser can compute its persistent homology.

For this special case, other software may be more efficient.

A related frequency-regression example is demonstrated in the TutorialTopologicalDataAnalysis repository.

V and T Constructions

  • V-construction: pixels/voxels represent 0-cells (4-neighborhood in 2D)
  • T-construction: pixels/voxels represent top-cells (8-neighborhood in 2D)

Use:

  • Python: filtration="V" or filtration="T"
  • CLI: cubicalripser (V), tcubicalripser (T)

By Alexander duality, the following are closely related:

./build/cubicalripser input.npy
./build/tcubicalripser --embedded input.npy

The difference is in the sign of filtration and treatment of permanent cycles. Here, --embedded converts input I to -I^infty in the paper's notation.

For details, see Duality in Persistent Homology of Images by Adelie Garin et al.

Creator and Destroyer Cells

The creator of a cycle is the cell that gives birth to the cycle. For example, in 0-dimensional homology, the voxel with lower filtration in a component creates that class, and a connecting voxel can destroy the class with higher birth time.

Creator and destroyer cells are not unique, but they are useful for localizing cycles.

For finite lifetime in the default convention:

arr[x2,y2,z2] - arr[x1,y1,z1] = death - birth = lifetime

where (x1,y1,z1) is creator and (x2,y2,z2) is destroyer.

With --embedded, creator and destroyer roles are swapped:

arr[x1,y1,z1] - arr[x2,y2,z2] = death - birth = lifetime

Thanks to Nicholas Byrne for suggesting this convention and providing test code.

Representative Homology Cycles

The Python APIs can optionally return a representative homology cycle for every persistence interval. This uses direct boundary reduction over F2, so it is intended for inspection and visualization rather than high-throughput PH calculation.

pairs, cycles = cripser.compute_ph(image, maxdim=1, representatives=True)

# cycles[i] is the cycle for persistence row pairs[i].
# Each cell is [x, y, z, cell_type] (or [x, y, z, w, cell_type] in 4D).

All listed cells have coefficient one in F2. cell_type identifies the cell orientation; for example, planar 1-cycles use 0 for x-edges and 1 for y-edges. Convert an individual chain to an array with np.asarray(cycles[i], dtype=np.uint32) if convenient.

Use plot_cycle for a single planar H₁ representative, or plot_cycles for several representatives. Passing the original 2-D image draws the cycles in the same (x, y) coordinates over the image; use overlay=False to omit the background.

ax = cripser.plot_cycles(
    [cycles[i] for i in selected_indices],
    image=image,
    labels=["H₁ #1", "H₁ #2"],
)

For an example with two figure-eight peaks, see the representative-cycles section of the main tutorial.

The option is disabled by default. In that mode CubicalRipser keeps using its existing optimized cohomology/coboundary reduction without allocating or tracking representative chains. It cannot be combined with top_dim=True, which uses a separate Alexander-duality shortcut.

Deep Learning Integration

The original project examples include lifetime-enhanced and histogram-style topological channels for CNNs.

Historical note: older documentation referenced demo/stackPH.py; equivalent functionality is now available in the Python vectorization APIs.

Example using current APIs:

import numpy as np
import cripser

arr = np.load("sample/2d_hole.npy")
ph = cripser.compute_ph(arr, maxdim=1)

# Persistence image (channels = selected homology dimensions)
pi = cripser.persistence_image(
    ph,
    homology_dims=(0, 1),
    n_birth_bins=32,
    n_life_bins=32,
)

# PH histogram volume (channels encode dim x life-bin x birth-bin)
hist = cripser.create_PH_histogram_volume(
    ph,
    image_shape=arr.shape,
    homology_dims=(0, 1),
    n_birth_bins=4,
    n_life_bins=4,
)

For practical CNN examples, see HomologyCNN.

Timing Comparisons

Timing and correctness comparison now use demo/compare_gudhi.py.

CLI timing example:

python demo/compare_gudhi.py \
  --methods cli \
  --sample-datasets sample/bonsai128.npy \
  --cubicalripser-bin build/cubicalripser \
  --tcubicalripser-bin build/tcubicalripser \
  --runs 5 --warmup 1 \
  --csv-out demo/logs/timing_bonsai128.csv

This writes:

  • run rows: one per timed iteration (elapsed_seconds)
  • summary rows: aggregate metrics per (binary, dataset)
  • statistics: mean_seconds, std_seconds, min_seconds, max_seconds
  • metadata: timestamp_utc, git_commit, maxdim, binary_path, input_path

Reference comparison example:

python demo/compare_gudhi.py \
  --methods cli \
  --sample-datasets bonsai128 bonsai256 4d_hole \
  --cubicalripser-bin build/cubicalripser \
  --tcubicalripser-bin build/tcubicalripser \
  --reference-csv demo/logs/reference_timing.csv \
  --max-slowdown 1.10 \
  --fail-on-regression \
  --csv-out demo/logs/timing_current_vs_reference.csv

Testing and Regression Checks

Run Python tests:

pytest

Timing regression against a saved reference:

python demo/compare_gudhi.py \
  --methods cli \
  --sample-datasets bonsai128 bonsai256 4d_hole \
  --cubicalripser-bin build/cubicalripser \
  --tcubicalripser-bin build/tcubicalripser \
  --reference-csv demo/logs/reference_timing.csv \
  --runs 3 --warmup 1 \
  --csv-out demo/logs/timing_current_vs_reference.csv

More details: demo/COMPARE.md.

Other Software for Cubical Complex PH

The following notes are based on limited understanding and tests and may be incomplete.

  • Cubicle by Hubert Wagner

    • V-construction
    • parallelized algorithms can be faster on multicore machines
    • chunked input handling can reduce memory usage

  • HomcCube by Ippei Obayashi

    • V-construction
    • integrated into HomCloud

  • DIPHA by Ulrich Bauer and Michael Kerber

    • V-construction
    • MPI-parallelized for cluster use
    • commonly used; memory footprint can be relatively large

  • GUDHI (INRIA)

    • V- and T-construction in arbitrary dimensions
    • strong documentation and usability
    • generally emphasizes usability over raw performance

  • diamorse

    • V-construction

  • Perseus by Vidit Nanda

    • V-construction

Release Notes

  • v0.0.35: Added support for computation of cycle representatives (homology cycles) for each persistence interval.
  • v0.0.34: Switched the Python binding layer from pybind11 to nanobind so a single cp312-abi3 wheel covers Python 3.12+. Python 3.8 is dropped (nanobind requires ≥ 3.9).
  • v0.0.31: Changed module structure (hopefully, backward compatible)
  • v0.0.30: Improved cache resulting in large speedup
  • v0.0.24: Repository renamed from CubicalRipser_3dim to CubicalRipser.
    • update old remote if needed: 
      git remote set-url origin https://github.com/shizuo-kaji/CubicalRipser.git
  • v0.0.23: Added torch integration
  • v0.0.22: Changed birth coordinates for T-construction to better match GUDHI for permanent cycles
  • v0.0.19: Added support for 4D cubical complexes
  • v0.0.15: Added support for Fortran-indexed numpy arrays (F_CONTIGUOUS arrays): Up to v0.0.14, C_CONTIGUOUS was assumed, which caused incorrect results for Fortran-indexed arrays.
  • v0.0.8: Fixed memory leak in Python bindings (pointed out by Nicholas Byrne)
  • v0.0.7: Speed improvements
  • v0.0.6: Changed birth/death location definition
  • up to v0.0.5, differences from the original version:
    • optimized implementation (lower memory footprint and faster on some data)
    • improved Python usability
    • much larger practical input sizes
    • cache control
    • Alexander duality option for highest-degree PH
    • both V and T constructions
    • birth/death location output

License

Distributed under GNU Lesser General Public License v3.0 or later. See LICENSE.

Project links

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for shizuo-kaji from gravatar.com shizuo-kaji

Credits

Author: Shizuo Kaji

License

GNU GENERAL PUBLIC LICENSE

Requires

Python >=3.9

Tags

persistent homology TDA cubical complex image volume

Report project as malware

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cripser-0.0.35.tar.gz (131.0 kB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

cripser-0.0.35-cp312-abi3-win_amd64.whl (258.4 kB view details)

Uploaded CPython 3.12+Windows x86-64

cripser-0.0.35-cp312-abi3-musllinux_1_2_x86_64.whl (1.3 MB view details)

Uploaded CPython 3.12+musllinux: musl 1.2+ x86-64

cripser-0.0.35-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl (393.6 kB view details)

Uploaded CPython 3.12+manylinux: glibc 2.24+ x86-64manylinux: glibc 2.28+ x86-64

cripser-0.0.35-cp312-abi3-macosx_11_0_x86_64.whl (277.8 kB view details)

Uploaded CPython 3.12+macOS 11.0+ x86-64

cripser-0.0.35-cp312-abi3-macosx_11_0_universal2.whl (466.2 kB view details)

Uploaded CPython 3.12+macOS 11.0+ universal2 (ARM64, x86-64)

cripser-0.0.35-cp312-abi3-macosx_11_0_arm64.whl (248.8 kB view details)

Uploaded CPython 3.12+macOS 11.0+ ARM64

cripser-0.0.35-cp311-cp311-win_amd64.whl (262.7 kB view details)

Uploaded CPython 3.11Windows x86-64

cripser-0.0.35-cp311-cp311-musllinux_1_2_x86_64.whl (1.3 MB view details)

Uploaded CPython 3.11musllinux: musl 1.2+ x86-64

cripser-0.0.35-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl (402.6 kB view details)

Uploaded CPython 3.11manylinux: glibc 2.24+ x86-64manylinux: glibc 2.28+ x86-64

cripser-0.0.35-cp311-cp311-macosx_11_0_x86_64.whl (282.2 kB view details)

Uploaded CPython 3.11macOS 11.0+ x86-64

cripser-0.0.35-cp311-cp311-macosx_11_0_universal2.whl (475.8 kB view details)

Uploaded CPython 3.11macOS 11.0+ universal2 (ARM64, x86-64)

cripser-0.0.35-cp311-cp311-macosx_11_0_arm64.whl (253.8 kB view details)

Uploaded CPython 3.11macOS 11.0+ ARM64

cripser-0.0.35-cp310-cp310-win_amd64.whl (263.1 kB view details)

Uploaded CPython 3.10Windows x86-64

cripser-0.0.35-cp310-cp310-musllinux_1_2_x86_64.whl (1.3 MB view details)

Uploaded CPython 3.10musllinux: musl 1.2+ x86-64

cripser-0.0.35-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl (403.4 kB view details)

Uploaded CPython 3.10manylinux: glibc 2.24+ x86-64manylinux: glibc 2.28+ x86-64

cripser-0.0.35-cp310-cp310-macosx_11_0_x86_64.whl (282.9 kB view details)

Uploaded CPython 3.10macOS 11.0+ x86-64

cripser-0.0.35-cp310-cp310-macosx_11_0_universal2.whl (477.2 kB view details)

Uploaded CPython 3.10macOS 11.0+ universal2 (ARM64, x86-64)

cripser-0.0.35-cp310-cp310-macosx_11_0_arm64.whl (254.3 kB view details)

Uploaded CPython 3.10macOS 11.0+ ARM64

cripser-0.0.35-cp39-cp39-win_amd64.whl (263.6 kB view details)

Uploaded CPython 3.9Windows x86-64

cripser-0.0.35-cp39-cp39-musllinux_1_2_x86_64.whl (1.3 MB view details)

Uploaded CPython 3.9musllinux: musl 1.2+ x86-64

cripser-0.0.35-cp39-cp39-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl (403.8 kB view details)

Uploaded CPython 3.9manylinux: glibc 2.24+ x86-64manylinux: glibc 2.28+ x86-64

cripser-0.0.35-cp39-cp39-macosx_11_0_x86_64.whl (283.2 kB view details)

Uploaded CPython 3.9macOS 11.0+ x86-64

cripser-0.0.35-cp39-cp39-macosx_11_0_universal2.whl (477.9 kB view details)

Uploaded CPython 3.9macOS 11.0+ universal2 (ARM64, x86-64)

cripser-0.0.35-cp39-cp39-macosx_11_0_arm64.whl (254.7 kB view details)

Uploaded CPython 3.9macOS 11.0+ ARM64

File details

Details for the file cripser-0.0.35.tar.gz.

File metadata

  • Download URL: cripser-0.0.35.tar.gz
  • Upload date:
  • Size: 131.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for cripser-0.0.35.tar.gz
Algorithm Hash digest
SHA256 f4119aadd3b1de125db8d7ea29147d9f36126bf902ade27d343997c78712f468
MD5 011cea95c2e01d8eaf2ff0384ad24f97
BLAKE2b-256 c875f5c14192a1e4ad9ed44169658208ef6c8027bf0c5d22b9ad6290082ccc14

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35.tar.gz:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-win_amd64.whl.

File metadata

  • Download URL: cripser-0.0.35-cp312-abi3-win_amd64.whl
  • Upload date:
  • Size: 258.4 kB
  • Tags: CPython 3.12+, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for cripser-0.0.35-cp312-abi3-win_amd64.whl
Algorithm Hash digest
SHA256 887b84d941e0b948be26229b336043e708ab9a3c4bbba2efb6067d20e5f2b464
MD5 774e8882be40f9e1957aa8739b736f18
BLAKE2b-256 91b18c59cfbcbc27aeb8f4d1d9eb325a9080275fcf43609511123182cee1e00b

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-win_amd64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp312-abi3-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 c7308101a34109499af2a117caa3c45248f500640481639d97707cc34c5aebd0
MD5 8d3f16deba75a0b569f719466a923ea9
BLAKE2b-256 09604bb43e84f634fff4c7ab716512716bd3528a09da3b6d66538684948296aa

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-musllinux_1_2_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 c2850912026f52e26f1ad9cf3b463c5940d0baac4ddffdf8795e168362412a73
MD5 4d99fa6ee93528e01cc2d7dbc4f5872b
BLAKE2b-256 e0958e3835fcbb95bd047aff9144e9ee61c49ad81cfed08e323f204c5d937363

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp312-abi3-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 9ca9c061f20df125680c6ae96ce0bf56cd8448461403a1a4821de7c281293e08
MD5 935f0dd2948e607e4b25f7802b76da42
BLAKE2b-256 16ac5ea2a8991ce3ff535fc458fb29b25ce94aaa58afb7cd8acd4f9cc542e87b

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-macosx_11_0_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp312-abi3-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 c810208573745272949000e5cbc7588c1a8d01543bc069037d8e175fde361b80
MD5 5cf646c009c1b4469d21d0afd68003e7
BLAKE2b-256 859afdcdc1115ab94144d0c61ecb2200a4d5d41dd36702ef755fd35f550e5aa6

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-macosx_11_0_universal2.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp312-abi3-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp312-abi3-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 748b3f9f55ed0521cb02cf13269224b5d071ae261f602feacdc042f73137c4ec
MD5 361e4d1f10fb8c94b12b2cadd3df1275
BLAKE2b-256 86c9638702c08605ee5237da81a3ce66e109373fc25e8b5f8819d46f88ad8fd3

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp312-abi3-macosx_11_0_arm64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-win_amd64.whl.

File metadata

  • Download URL: cripser-0.0.35-cp311-cp311-win_amd64.whl
  • Upload date:
  • Size: 262.7 kB
  • Tags: CPython 3.11, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for cripser-0.0.35-cp311-cp311-win_amd64.whl
Algorithm Hash digest
SHA256 ac0e4f4e73da7c6fc7753b14c1695e269e064c1d42639afea85de0593fdf5153
MD5 8738185b1d73b322f513112877cc39cf
BLAKE2b-256 25d9b1b7d9f33419299d6b008e768848145aa79a8c8d2e3da0bbb19c9fc95154

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-win_amd64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp311-cp311-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 06a1e71d4c55657a401c3f833c50771c8cf91ecacb4d6653cd317fad70eb355a
MD5 980ef351b4b10ff6e0cf1fc61e01f6d1
BLAKE2b-256 636333d3a3c99924526ab8d8581076a47af0c37c2c444546b050809498ad1d01

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-musllinux_1_2_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 0202e762bcf353571b3429423c3fdd079cbd685ba82a4f60f767a036e951aab8
MD5 23c22c919fb608a5a5c533acbaacf6da
BLAKE2b-256 b194e48a1dd4d471cfd758d62a58c72f1a66788b3b332b091fb329020f4d8400

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp311-cp311-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 b83b1f9375b51f2cc46d1327c14887323cf9a05bd374a148de941f7cf0f9d0fa
MD5 4548f08f3d605d082be264463c338c92
BLAKE2b-256 3f5f8aeb228ae3b021af54b62be747c6128f0396a762ba440cf5b0186ab5969e

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-macosx_11_0_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp311-cp311-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 78495d75d05cac43b1138f9ceb377d863524d781e1ae941405edab3266259eec
MD5 71969d717a71e079a3897a75ab4353e4
BLAKE2b-256 ac77ad46782183b18f13a616b6dea896da183f1c47a10511d26b873c0c5f11ac

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-macosx_11_0_universal2.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp311-cp311-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp311-cp311-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 f15f6962d74989f104360eb48fcb462e0f8e8e341c4dcc9bf97a2e82650adec9
MD5 e164e9d9757f5f71a5b3504efa0f43ed
BLAKE2b-256 7821b5b56f4aba910e37407c1b199be7a53330334bd01fb17914c776088d6e29

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp311-cp311-macosx_11_0_arm64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-win_amd64.whl.

File metadata

  • Download URL: cripser-0.0.35-cp310-cp310-win_amd64.whl
  • Upload date:
  • Size: 263.1 kB
  • Tags: CPython 3.10, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for cripser-0.0.35-cp310-cp310-win_amd64.whl
Algorithm Hash digest
SHA256 aa3b51bfa1f3a6fb21970f26877ddc0cb46aa51e46c359b775618202e6d53945
MD5 98e559f6d722c5d6227caf808267b3b9
BLAKE2b-256 91a9761bc4b1161eab882003081d369fb9ed58083b00fc82be3f39416404c9e6

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-win_amd64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp310-cp310-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 74b52105013c716529ea9ab03e7d6fa270a2960f090876488f324694ecf7c2ef
MD5 a8ac04d809a94e982acb549bcc154605
BLAKE2b-256 c38da738506700a5d1c6b4e376c1ecad4f65d42cc490b5a210de52157b6f2000

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-musllinux_1_2_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 8deeeed76d5fa0acfad5f9852afe50d6aff662da278df646261cdbfc3cf386b3
MD5 0313000f4a8cbaf8415c3c38b517c686
BLAKE2b-256 abaea268dae6c14e638797fbee5a9f4e98a9bacf6bddcd3c7c0accd845e15bfe

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp310-cp310-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 bcebf2ab5210982f5228a2276f48991e2e39da6c31f49ce40295d69caeda97bc
MD5 e1c0699ee22e6a64a64352f069b70e5b
BLAKE2b-256 8157bbc7d7f3ae52a20b92b96cf966491d03b9b2630a313ac5530ac833d74a30

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-macosx_11_0_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp310-cp310-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 bc65262643c9feca8faca19d89c81184755ad847d99c29fe2ecbb5a5f0bdec77
MD5 152537193662a713fe23a3fe2278bc9c
BLAKE2b-256 0a217ef0bae853f1c6ae33f4a67fabbe4f00ebb9daa7ec9cc6cf6618aee1de0c

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-macosx_11_0_universal2.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp310-cp310-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp310-cp310-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 eb5eecb77884c804bcba15983b0e0ab89c102c22b5455f092071e2e33a95322c
MD5 854bbc2a033322d09dd191a29bc5894d
BLAKE2b-256 360ed650d21d5263572ab97faf4d94f9a295bc8601224b5abba55373d3c6358d

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp310-cp310-macosx_11_0_arm64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-win_amd64.whl.

File metadata

  • Download URL: cripser-0.0.35-cp39-cp39-win_amd64.whl
  • Upload date:
  • Size: 263.6 kB
  • Tags: CPython 3.9, Windows x86-64
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for cripser-0.0.35-cp39-cp39-win_amd64.whl
Algorithm Hash digest
SHA256 8c2828bfdc9b1011b7067b662b0239b1cbb4510b20c823c6f1019e489bd3180b
MD5 e47660193b2ddac31bac52daa79e5981
BLAKE2b-256 87e554cf5bfd578bf2c4ed27e7c808e1afc2fad4cec0767d1ae2733011c7967d

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-win_amd64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp39-cp39-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 e4bbfeb469ffe0d082b7cea306fba34f2178a2ddbbcf435c3efbbd1bf931a78b
MD5 6c829fadaabfdccd7f2431d10118e225
BLAKE2b-256 7486c55ad20580dec62b0d2e2aa601c0bdfbf2e7cd3bb433214dcc1f902e7b58

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-musllinux_1_2_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp39-cp39-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 4689052ec993a89aa650f1fa137f545696a04bc9823aa376b54ac43654278a42
MD5 570c5c7fd80bf0744c0469e1553cd531
BLAKE2b-256 6561f5292422c08d2c1d1ba8a9141379d4dbfbc5a9d7a0e0578ef6e659faa927

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp39-cp39-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 194087cb2dc9d0725d9019c429b55a9d57e45576d3851809b5afd36530c0ce38
MD5 57ed3f04cc19b366eeb9709aff0b6d43
BLAKE2b-256 4250c65af44bc012e009421dfbdc5222568657c5ae432edfaab294b691d63b16

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-macosx_11_0_x86_64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp39-cp39-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 0ea6bc8671bf443604542047d606ead808f77c49514f8f29da688b7e7c452ccd
MD5 5bbf8841287e2a21f4f6eea73504ac09
BLAKE2b-256 3fb80ca4427c099e5d7a1f6b9fe3fdea7241509182d15786ee64113c84e1cb88

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-macosx_11_0_universal2.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file cripser-0.0.35-cp39-cp39-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for cripser-0.0.35-cp39-cp39-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 63b304fa79f17fe01c74992a0705a7ee34d53c6cc21ec0e18032c9dad42e0e0b
MD5 a5f015446ffdf46a2b0eae09e5550cdd
BLAKE2b-256 2e9e2b1557d81506bab1646ae543e6f8a6e695dd62413a7c8b2f238a902f0760

See more details on using hashes here.

Provenance

The following attestation bundles were made for cripser-0.0.35-cp39-cp39-macosx_11_0_arm64.whl:

Publisher: build.yml on shizuo-kaji/CubicalRipser

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.0.35 This release

25 files

0.0.34

25 files

0.0.33

43 files

0.0.32

43 files

0.0.25

43 files

0.0.24

43 files

0.0.23

41 files

0.0.22

41 files

0.0.21

41 files

0.0.19

41 files

0.0.16

41 files

0.0.15

41 files

0.0.14

41 files

0.0.13

13 files

0.0.12

5 files

0.0.11

5 files

0.0.10

5 files

0.0.9

2 files

0.0.8

1 file

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page