caskade logo

caskade

CI CD codecov PyPI - Version Documentation Status DOI

Build scientific simulators, treating them as a directed acyclic graph. Handles argument passing for complex nested simulators.

Install

pip install caskade

This will give you the numpy version of caskade. More details on the docs page. if you want to use caskade with jax then run:

pip install caskade[jax]

if you want to use caskade with torch then run:

pip install caskade[torch]

Don't worry if you got the wrong one. Just install the appropriate jax/torch directly if you want to add the option.

Usage

Make a Module object which may have some Params. Define a forward method using the decorator.

from caskade import Module, Param, forward

class MySim(Module):
    def __init__(self, a, b=None):
        super().__init__()
        self.a = a
        self.b = Param("b", b)

    @forward
    def myfun(self, x, b=None):
        return x + self.a + b

We may now create instances of the simulator and pass the dynamic parameters.

import torch

sim = MySim(1.0)

params = [torch.tensor(2.0)]

print(sim.myfun(3.0, params=params))

Which will print 6 by automatically filling b with the value from params.

Why do this?

The above example is not very impressive, the real power comes from the fact that Module objects can be nested, making an arbitrarily complicated analysis graph. Some other features include:

  • Unroll parameters into 1D vector to interface with other packages (emcee, scipy.optimize, dynesty, etc.)
  • Link parameters by value or functional relationship
  • Reparametrize (e.g. between polar and cartesian) without modifying underlying code
  • Save and load sampling chains automatically in HDF5
  • Track metadata alongside parameters
  • And much more! Beginner tutorial and Advanced tutorial

Use different backends

caskade can be run with different backends for torch, numpy, and jax. See the Beginners Guide tutorial to learn more!

Documentation

The caskade interface has lots of flexibility, check out the docs to learn more. For a quick start, jump right to the Jupyter notebook tutorial!

The caustics package can serve as a project template utilizing the many features of caskade.

The caskade package maintains 100% coverage for unit testing, ensuring reliability as the backbone of a research project.

Project links

Data verified by PyPI on Jul 19, 2026
Data provided by the project maintainers, verified at the time the release was uploaded to PyPI.

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 ConnorStone from gravatar.com ConnorStone

Credits

Author: Connor Stone

GitHub Statistics

Data verified by PyPI on Jul 19, 2026
The GitHub source repository was provided by the project maintainers and verified by PyPI at the time of upload. Stars, forks, and open issues/PRs are derived from that repository and have not been independently verified.

License

MIT License (MIT License Copyright (c) 2024 Connor Stone, PhD Permission is hereby granted, free of charge, to ...)

Requires

Python >=3.11

Provides Extra

dev jax numpy torch

Tags

DAG caskade differentiable programming jax pytorch scientific python

Classifiers

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

caskade-1.1.5.tar.gz (283.6 kB view details)

Uploaded Source

Built Distribution

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

caskade-1.1.5-py3-none-any.whl (43.3 kB view details)

Uploaded Python 3

File details

Details for the file caskade-1.1.5.tar.gz.

File metadata

  • Download URL: caskade-1.1.5.tar.gz
  • Upload date:
  • Size: 283.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.0.1 CPython/3.12.8

File hashes

Hashes for caskade-1.1.5.tar.gz
Algorithm Hash digest
SHA256 cf901badfca33a27a13c219e3a88640867b6ca9a1abfda7f428ad32986631d78
MD5 5c46734cafe6176b06c5e5d3fb239927
BLAKE2b-256 06f4c1912d689a7f43daf2f61d14831505bef63fb043d6a0d5df0c8e06b9cb46

See more details on using hashes here.

Provenance

The following attestation bundles were made for caskade-1.1.5.tar.gz:

Publisher: cd.yml on ConnorStoneAstro/caskade

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

File details

Details for the file caskade-1.1.5-py3-none-any.whl.

File metadata

  • Download URL: caskade-1.1.5-py3-none-any.whl
  • Upload date:
  • Size: 43.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.0.1 CPython/3.12.8

File hashes

Hashes for caskade-1.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 89c54775efd1be83ba5f699583a7476e7c8a96c0f50e62a10b5cccb485f413be
MD5 aa0e1620f7e4bee4aa273f1cd4d85fd4
BLAKE2b-256 f2cb2abc4a1469d5c375c8b2fcfebefcb35ef1f63df9c1b917065618557504d6

See more details on using hashes here.

Provenance

The following attestation bundles were made for caskade-1.1.5-py3-none-any.whl:

Publisher: cd.yml on ConnorStoneAstro/caskade

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

1.1.5 This release

2 files

1.1.4

2 files

1.1.3

2 files

1.1.2

2 files

1.1.1

2 files

1.1.0

2 files

1.0.0

2 files

0.15.1

2 files

0.15.0

2 files

0.14.1

2 files

0.14.0

2 files

0.13.0

2 files

0.12.1

2 files

0.12.0

2 files

0.11.0

2 files

0.10.5

2 files

0.10.4

2 files

0.10.3

2 files

0.10.2

2 files

0.10.1

2 files

0.10.0

2 files

0.9.3

2 files

0.9.2

2 files

0.9.1

2 files

0.9.0

2 files

0.8.3

2 files

0.8.2

2 files

0.8.1

2 files

0.8.0

2 files

0.7.5

2 files

0.7.4

2 files

0.7.3

2 files

0.7.2

2 files

0.7.1

2 files

0.7.0

2 files

0.6.3

2 files

0.6.2

2 files

0.6.1

2 files

0.6.0

2 files

0.5.1

2 files

0.5.0

2 files

0.4.2

2 files

0.4.1

2 files

0.4.0

2 files

0.3.0

2 files

0.2.0

2 files

0.1.0

2 files

0.0.3

2 files

0.0.2

2 files

0.0.1

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page