# Entropy Coders for Research and Production
The `constriction` library provides a set of composable entropy coding algorithms with a
focus on correctness, versatility, ease of use, compression performance, and
computational efficiency. The goals of `constriction` are three-fold:
1. **to facilitate research on novel lossless and lossy compression methods** by
providing a *composable* set of primitives (e.g., you can can easily switch out a
Range Coder for an ANS coder without having to find a new library or change how you
represent exactly invertible entropy models);
2. **to simplify the transition from research code to deployed software** by providing
similar APIs and binary compatible entropy coders for both Python (for rapid
prototyping on research code) and Rust (for turning successful prototypes into
standalone binaries, libraries, or WebAssembly modules); and
3. **to serve as a teaching resource** by providing a variety of entropy coding
primitives within a single consistent framework. Check out our [additional teaching
material](https://robamler.github.io/teaching/compress21/) from a university course
on data compression, which contains some problem sets where you use `constriction`
(with solutions).
**More Information:** [project website](https://bamler-lab.github.io/constriction)
**Live demo:** [here's a web app](https://robamler.github.io/linguistic-flux-capacitor)
that started out as a machine-learning research project in Python and was later turned
into a web app by using `constriction` in a WebAssembly module).
## Quick Start
### Installing `constriction` for Python
```bash
pip install constriction~=0.4.1
```
### Hello, World
You'll mostly use the `stream` submodule, which provides stream codes (like Range
Coding or ANS). The following example shows a simple encoding-decoding round trip. More
complex entropy models and other entropy coders are also supported, see section
"More Examples" below.
```python
import constriction
import numpy as np
message = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)
# Define an i.i.d. entropy model (see below for more complex models):
entropy_model = constriction.stream.model.QuantizedGaussian(-50, 50, 3.2, 9.6)
# Let's use an ANS coder in this example. See below for a Range Coder example.
encoder = constriction.stream.stack.AnsCoder()
encoder.encode_reverse(message, entropy_model)
compressed = encoder.get_compressed()
print(f"compressed representation: {compressed}")
print(f"(in binary: {[bin(word) for word in compressed]})")
decoder = constriction.stream.stack.AnsCoder(compressed)
decoded = decoder.decode(entropy_model, 9) # (decodes 9 symbols)
assert np.all(decoded == message)
```
## More Examples
### Switching Out the Entropy Coding Algorithm
Let's take our "Hello, World" example from above and assume we want to
switch the entropy coding algorithm from ANS to Range Coding. But we don't want to
look for a new library or change how we represent entropy *models* and compressed data.
Luckily, we only have to modify a few lines of code:
```python
import constriction
import numpy as np
# Same representation of message and entropy model as in the previous example:
message = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)
entropy_model = constriction.stream.model.QuantizedGaussian(-50, 50, 3.2, 9.6)
# Let's use a Range coder now:
encoder = constriction.stream.queue.RangeEncoder() # <-- CHANGED LINE
encoder.encode(message, entropy_model) # <-- (slightly) CHANGED LINE
compressed = encoder.get_compressed()
print(f"compressed representation: {compressed}")
print(f"(in binary: {[bin(word) for word in compressed]})")
decoder = constriction.stream.queue.RangeDecoder(compressed) #<--CHANGED LINE
decoded = decoder.decode(entropy_model, 9) # (decodes 9 symbols)
assert np.all(decoded == message)
```
### Complex Entropy Models
This time, let's keep the entropy coding algorithm as it is but make the entropy *model*
more complex. We'll encode the first 5 symbols of the message again with a
`QuantizedGaussian` distribution, but this time we'll use individual model parameters
(means and standard deviations) for each of the 5 symbols. For the remaining 4 symbols,
we'll use a fixed categorical distribution, just to make it more interesting:
```python
import constriction
import numpy as np
# Same message as above, but a complex entropy model consisting of two parts:
message = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)
means = np.array([2.3, 6.1, -8.5, 4.1, 1.3], dtype=np.float32)
stds = np.array([6.2, 5.3, 3.8, 3.2, 4.7], dtype=np.float32)
entropy_model1 = constriction.stream.model.QuantizedGaussian(-50, 50)
entropy_model2 = constriction.stream.model.Categorical(
np.array([0.2, 0.5, 0.3], dtype=np.float32), # Probabilities of the symbols 0,1,2.
perfect=False
)
# Simply encode both parts in sequence with their respective models:
encoder = constriction.stream.queue.RangeEncoder()
encoder.encode(message[0:5], entropy_model1, means, stds) # per-symbol params.
encoder.encode(message[5:9], entropy_model2)
compressed = encoder.get_compressed()
print(f"compressed representation: {compressed}")
print(f"(in binary: {[bin(word) for word in compressed]})")
decoder = constriction.stream.queue.RangeDecoder(compressed)
decoded_part1 = decoder.decode(entropy_model1, means, stds)
decoded_part2 = decoder.decode(entropy_model2, 4)
assert np.all(np.concatenate((decoded_part1, decoded_part2)) == message)
```
You can define even more complex entropy models by providing an arbitrary Python
function for the cumulative distribution function (see
[`CustomModel`](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html#constriction.stream.model.CustomModel) and
[`ScipyModel`](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html#constriction.stream.model.CustomModel)). The
`constriction` library provides wrappers that turn your models into *exactly*
invertible fixed-point arithmetic since even tiny rounding errors could otherwise
completely break an entropy coding algorithm.
### Exercise
We've shown examples of ANS coding with a simple entropy model, of
Range Coding with the same simple entropy model, and of Range coding with a complex
entropy model. One combination is still missing: ANS coding
with the complex entropy model from the last example above. This should be no problem
now, so try it out yourself:
- In the last example above, change both "queue.RangeEncoder" and "queue.RangeDecoder"
to "stack.AnsCoder" (ANS uses the same data structure for both encoding and decoding).
- Then change both occurrences of `.encode(...)` to `.encode_reverse(...)` (since ANS
operates as a stack, i.e., last-in-first-out, we encode the symbols in reverse order
so that we can decode them in their normal order).
- Finally, there's one slightly subtle change: when encoding the message, switch the
order of the two lines that encode `message[0:5]` and `message[5:9]`, respectively.
Do *not* change the order of decoding though. This is again necessary because ANS
operates as a stack.
Congratulations, you've successfully implemented your first own compression scheme with
`constriction`.
## Further Reading
You can find links to more examples and tutorials on the [project
website](https://bamler-lab.github.io/constriction). Or just dive right into the
documentation of [range coding](https://bamler-lab.github.io/constriction/apidoc/python/stream/queue.html), [ANS](https://bamler-lab.github.io/constriction/apidoc/python/stream/stack.html), and
[entropy models](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html).
If you're still new to the concept of entropy coding then check out the [teaching
material](https://robamler.github.io/teaching/compress21/).
## Contributing
Pull requests and issue reports are welcome. Unless contributors explicitly state otherwise
at the time of contributing, all contributions will be assumed to be licensed under either
one of MIT license, Apache License Version 2.0, or Boost Software License Version 1.0, at
the choice of each licensee.
There's no official guide for contributions since nobody reads those anyway. Just be nice to
other people and act like a grown-up (i.e., it's OK to make mistakes as long as you strive
for improvement and are open to consider respectfully phrased opinions of other people).
## License
This work is licensed under the terms of the MIT license, Apache License Version 2.0, or
Boost Software License Version 1.0. You can choose between one of them if you use this work.
See the files whose name start with `LICENSE` in this directory. The compiled python
extension module is linked with a number of third party libraries. Binary distributions of
the `constriction` python extension module contain a file `LICENSE.html` that includes all
licenses of all dependencies (the file is also available
[online](https://bamler-lab.github.io/constriction/license.html)).
## What's With the Name?
Constriction is a library of compression primitives with bindings for Rust and Python.
[Pythons](https://en.wikipedia.org/wiki/Pythonidae) are a family of nonvenomous snakes that
subdue their prey by "compressing" it, a method known as
[constriction](https://en.wikipedia.org/wiki/Constriction).
Raw data
{
"_id": null,
"home_page": "https://bamler-lab.github.io/constriction/",
"name": "constriction",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "Robert Bamler <robert.bamler@uni-tuebingen.de>",
"keywords": "compression, entropy-coding, range-coding, python, machine-learning",
"author": "Robert Bamler <robert.bamler@uni-tuebingen.de>",
"author_email": "Robert Bamler <robert.bamler@uni-tuebingen.de>",
"download_url": null,
"platform": null,
"description": "# Entropy Coders for Research and Production\n\nThe `constriction` library provides a set of composable entropy coding algorithms with a\nfocus on correctness, versatility, ease of use, compression performance, and\ncomputational efficiency. The goals of `constriction` are three-fold:\n\n1. **to facilitate research on novel lossless and lossy compression methods** by\n providing a *composable* set of primitives (e.g., you can can easily switch out a\n Range Coder for an ANS coder without having to find a new library or change how you\n represent exactly invertible entropy models);\n2. **to simplify the transition from research code to deployed software** by providing\n similar APIs and binary compatible entropy coders for both Python (for rapid\n prototyping on research code) and Rust (for turning successful prototypes into\n standalone binaries, libraries, or WebAssembly modules); and\n3. **to serve as a teaching resource** by providing a variety of entropy coding\n primitives within a single consistent framework. Check out our [additional teaching\n material](https://robamler.github.io/teaching/compress21/) from a university course\n on data compression, which contains some problem sets where you use `constriction`\n (with solutions).\n\n**More Information:** [project website](https://bamler-lab.github.io/constriction)\n\n**Live demo:** [here's a web app](https://robamler.github.io/linguistic-flux-capacitor)\nthat started out as a machine-learning research project in Python and was later turned\ninto a web app by using `constriction` in a WebAssembly module).\n\n## Quick Start\n\n### Installing `constriction` for Python\n\n```bash\npip install constriction~=0.4.1\n```\n\n### Hello, World\n\nYou'll mostly use the `stream` submodule, which provides stream codes (like Range\nCoding or ANS). The following example shows a simple encoding-decoding round trip. More\ncomplex entropy models and other entropy coders are also supported, see section\n\"More Examples\" below.\n\n```python\nimport constriction\nimport numpy as np\n\nmessage = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)\n\n# Define an i.i.d. entropy model (see below for more complex models):\nentropy_model = constriction.stream.model.QuantizedGaussian(-50, 50, 3.2, 9.6)\n\n# Let's use an ANS coder in this example. See below for a Range Coder example.\nencoder = constriction.stream.stack.AnsCoder()\nencoder.encode_reverse(message, entropy_model)\n\ncompressed = encoder.get_compressed()\nprint(f\"compressed representation: {compressed}\")\nprint(f\"(in binary: {[bin(word) for word in compressed]})\")\n\ndecoder = constriction.stream.stack.AnsCoder(compressed)\ndecoded = decoder.decode(entropy_model, 9) # (decodes 9 symbols)\nassert np.all(decoded == message)\n```\n\n## More Examples\n\n### Switching Out the Entropy Coding Algorithm\n\nLet's take our \"Hello, World\" example from above and assume we want to\nswitch the entropy coding algorithm from ANS to Range Coding. But we don't want to\nlook for a new library or change how we represent entropy *models* and compressed data.\nLuckily, we only have to modify a few lines of code:\n\n```python\nimport constriction\nimport numpy as np\n\n# Same representation of message and entropy model as in the previous example:\nmessage = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)\nentropy_model = constriction.stream.model.QuantizedGaussian(-50, 50, 3.2, 9.6)\n\n# Let's use a Range coder now:\nencoder = constriction.stream.queue.RangeEncoder() # <-- CHANGED LINE\nencoder.encode(message, entropy_model) # <-- (slightly) CHANGED LINE\n\ncompressed = encoder.get_compressed()\nprint(f\"compressed representation: {compressed}\")\nprint(f\"(in binary: {[bin(word) for word in compressed]})\")\n\ndecoder = constriction.stream.queue.RangeDecoder(compressed) #<--CHANGED LINE\ndecoded = decoder.decode(entropy_model, 9) # (decodes 9 symbols)\nassert np.all(decoded == message)\n```\n\n### Complex Entropy Models\n\nThis time, let's keep the entropy coding algorithm as it is but make the entropy *model*\nmore complex. We'll encode the first 5 symbols of the message again with a\n`QuantizedGaussian` distribution, but this time we'll use individual model parameters\n(means and standard deviations) for each of the 5 symbols. For the remaining 4 symbols,\nwe'll use a fixed categorical distribution, just to make it more interesting:\n\n```python\nimport constriction\nimport numpy as np\n\n# Same message as above, but a complex entropy model consisting of two parts:\nmessage = np.array([6, 10, -4, 2, 5, 2, 1, 0, 2], dtype=np.int32)\nmeans = np.array([2.3, 6.1, -8.5, 4.1, 1.3], dtype=np.float32)\nstds = np.array([6.2, 5.3, 3.8, 3.2, 4.7], dtype=np.float32)\nentropy_model1 = constriction.stream.model.QuantizedGaussian(-50, 50)\nentropy_model2 = constriction.stream.model.Categorical(\n np.array([0.2, 0.5, 0.3], dtype=np.float32), # Probabilities of the symbols 0,1,2.\n perfect=False\n) \n\n# Simply encode both parts in sequence with their respective models:\nencoder = constriction.stream.queue.RangeEncoder()\nencoder.encode(message[0:5], entropy_model1, means, stds) # per-symbol params.\nencoder.encode(message[5:9], entropy_model2)\n\ncompressed = encoder.get_compressed()\nprint(f\"compressed representation: {compressed}\")\nprint(f\"(in binary: {[bin(word) for word in compressed]})\")\n\ndecoder = constriction.stream.queue.RangeDecoder(compressed)\ndecoded_part1 = decoder.decode(entropy_model1, means, stds)\ndecoded_part2 = decoder.decode(entropy_model2, 4)\nassert np.all(np.concatenate((decoded_part1, decoded_part2)) == message)\n```\n\nYou can define even more complex entropy models by providing an arbitrary Python\nfunction for the cumulative distribution function (see\n[`CustomModel`](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html#constriction.stream.model.CustomModel) and\n[`ScipyModel`](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html#constriction.stream.model.CustomModel)). The\n`constriction` library provides wrappers that turn your models into *exactly*\ninvertible fixed-point arithmetic since even tiny rounding errors could otherwise\ncompletely break an entropy coding algorithm.\n\n### Exercise\n\nWe've shown examples of ANS coding with a simple entropy model, of\nRange Coding with the same simple entropy model, and of Range coding with a complex\nentropy model. One combination is still missing: ANS coding\nwith the complex entropy model from the last example above. This should be no problem\nnow, so try it out yourself:\n\n- In the last example above, change both \"queue.RangeEncoder\" and \"queue.RangeDecoder\"\n to \"stack.AnsCoder\" (ANS uses the same data structure for both encoding and decoding).\n- Then change both occurrences of `.encode(...)` to `.encode_reverse(...)` (since ANS\n operates as a stack, i.e., last-in-first-out, we encode the symbols in reverse order\n so that we can decode them in their normal order).\n- Finally, there's one slightly subtle change: when encoding the message, switch the\n order of the two lines that encode `message[0:5]` and `message[5:9]`, respectively.\n Do *not* change the order of decoding though. This is again necessary because ANS\n operates as a stack.\n\nCongratulations, you've successfully implemented your first own compression scheme with\n`constriction`.\n\n## Further Reading\n\nYou can find links to more examples and tutorials on the [project\nwebsite](https://bamler-lab.github.io/constriction). Or just dive right into the\ndocumentation of [range coding](https://bamler-lab.github.io/constriction/apidoc/python/stream/queue.html), [ANS](https://bamler-lab.github.io/constriction/apidoc/python/stream/stack.html), and\n[entropy models](https://bamler-lab.github.io/constriction/apidoc/python/stream/model.html).\n\nIf you're still new to the concept of entropy coding then check out the [teaching\nmaterial](https://robamler.github.io/teaching/compress21/).\n\n## Contributing\n\nPull requests and issue reports are welcome. Unless contributors explicitly state otherwise\nat the time of contributing, all contributions will be assumed to be licensed under either\none of MIT license, Apache License Version 2.0, or Boost Software License Version 1.0, at\nthe choice of each licensee.\n\nThere's no official guide for contributions since nobody reads those anyway. Just be nice to\nother people and act like a grown-up (i.e., it's OK to make mistakes as long as you strive\nfor improvement and are open to consider respectfully phrased opinions of other people).\n\n## License\n\nThis work is licensed under the terms of the MIT license, Apache License Version 2.0, or\nBoost Software License Version 1.0. You can choose between one of them if you use this work.\nSee the files whose name start with `LICENSE` in this directory. The compiled python\nextension module is linked with a number of third party libraries. Binary distributions of\nthe `constriction` python extension module contain a file `LICENSE.html` that includes all\nlicenses of all dependencies (the file is also available\n[online](https://bamler-lab.github.io/constriction/license.html)).\n\n## What's With the Name?\n\nConstriction is a library of compression primitives with bindings for Rust and Python.\n[Pythons](https://en.wikipedia.org/wiki/Pythonidae) are a family of nonvenomous snakes that\nsubdue their prey by \"compressing\" it, a method known as\n[constriction](https://en.wikipedia.org/wiki/Constriction).\n\n",
"bugtrack_url": null,
"license": "various (see LICENSE.html)",
"summary": "Entropy coders for research and production (Rust and Python).",
"version": "0.4.1",
"project_urls": {
"Homepage": "https://bamler-lab.github.io/constriction/",
"changelog": "https://github.com/bamler-lab/constriction/releases",
"documentation": "https://bamler-lab.github.io/constriction/apidoc/python/",
"repository": "https://github.com/bamler-lab/constriction"
},
"split_keywords": [
"compression",
" entropy-coding",
" range-coding",
" python",
" machine-learning"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f48ecb5f62de673cba49f31c7ecca330e8359d5c94148965e6bd5d92f5e0fc47",
"md5": "c12a99e404e1bffd96879b14b2afca80",
"sha256": "36df35c8f7665a9bf9cb1cbe545b0275e93a3a8c6d74438f70b526d90388ce1c"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp310-cp310-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "c12a99e404e1bffd96879b14b2afca80",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8",
"size": 814415,
"upload_time": "2024-10-23T10:17:58",
"upload_time_iso_8601": "2024-10-23T10:17:58.941866Z",
"url": "https://files.pythonhosted.org/packages/f4/8e/cb5f62de673cba49f31c7ecca330e8359d5c94148965e6bd5d92f5e0fc47/constriction-0.4.1-cp310-cp310-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "18c62492d14cd804f131c2df5e4cc29820e51405a1b80ddf7e2c1f4f1ff03f9d",
"md5": "7c80dd8a3af814e6f4b5098511ff5a80",
"sha256": "158853a0f5f0290e8dd43f0c6773061d609b7d025e76d49c02402863a9db71a4"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp310-cp310-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "7c80dd8a3af814e6f4b5098511ff5a80",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8",
"size": 413312,
"upload_time": "2024-10-23T10:18:00",
"upload_time_iso_8601": "2024-10-23T10:18:00.145030Z",
"url": "https://files.pythonhosted.org/packages/18/c6/2492d14cd804f131c2df5e4cc29820e51405a1b80ddf7e2c1f4f1ff03f9d/constriction-0.4.1-cp310-cp310-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "2c82de9d2293d17548fed7cf2d32891d1c8bca5de57d2501b7e42bb9e495ccd4",
"md5": "950c245bc554078c4465e733bf379738",
"sha256": "418872bd9135a7113bc494ea1d2818e891663bcc8e8aa8400242bc7a19e076ac"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "950c245bc554078c4465e733bf379738",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8",
"size": 461579,
"upload_time": "2024-10-23T10:13:54",
"upload_time_iso_8601": "2024-10-23T10:13:54.334321Z",
"url": "https://files.pythonhosted.org/packages/2c/82/de9d2293d17548fed7cf2d32891d1c8bca5de57d2501b7e42bb9e495ccd4/constriction-0.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "8d5ead33ab67f64db8b0f60876da29e3f6d0c43393d5d2c1ea81158960e282ad",
"md5": "a1ba3fbba4a969c0e68b9a5fa0eb2ef9",
"sha256": "fc0c5e746f4154bb13a723e95673d5580974fb6a9422e45018fb1fe1f16c8a36"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp310-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "a1ba3fbba4a969c0e68b9a5fa0eb2ef9",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8",
"size": 301367,
"upload_time": "2024-10-23T10:16:45",
"upload_time_iso_8601": "2024-10-23T10:16:45.186712Z",
"url": "https://files.pythonhosted.org/packages/8d/5e/ad33ab67f64db8b0f60876da29e3f6d0c43393d5d2c1ea81158960e282ad/constriction-0.4.1-cp310-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "da6691437bae9efe39db4b2e1c98877db0f6e6713475394414eda0f4f723b8b7",
"md5": "1e1aea232afbb5ce7834d511259b4d70",
"sha256": "07b49e8158ef529406d2a978ee3f0d6a80c0c37719c9498cd1e2db62e654e60a"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp311-cp311-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "1e1aea232afbb5ce7834d511259b4d70",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8",
"size": 814299,
"upload_time": "2024-10-23T10:18:01",
"upload_time_iso_8601": "2024-10-23T10:18:01.536742Z",
"url": "https://files.pythonhosted.org/packages/da/66/91437bae9efe39db4b2e1c98877db0f6e6713475394414eda0f4f723b8b7/constriction-0.4.1-cp311-cp311-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "ea8fbca40e8e305ec0baebd754fb65ed6d94e5a408683b19a4934ca099d51bab",
"md5": "0ed5f6228f6bddad98b48ba61cb132a8",
"sha256": "9753302b8af085dadb85aa9f26f4b6d2fcd6e2e8c1d69d647404f70c721ae215"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp311-cp311-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "0ed5f6228f6bddad98b48ba61cb132a8",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8",
"size": 413195,
"upload_time": "2024-10-23T10:18:02",
"upload_time_iso_8601": "2024-10-23T10:18:02.721241Z",
"url": "https://files.pythonhosted.org/packages/ea/8f/bca40e8e305ec0baebd754fb65ed6d94e5a408683b19a4934ca099d51bab/constriction-0.4.1-cp311-cp311-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4f8cad7ac213e272e56162835951e7d7b4ccdac48c076daddc662f94125a7118",
"md5": "d20a93913541038583b7014db21a11a3",
"sha256": "5629e1a4785be4b59e0cf15f76408b3ac0587976db838737373a16f1e4191e03"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "d20a93913541038583b7014db21a11a3",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8",
"size": 461294,
"upload_time": "2024-10-23T10:13:56",
"upload_time_iso_8601": "2024-10-23T10:13:56.138634Z",
"url": "https://files.pythonhosted.org/packages/4f/8c/ad7ac213e272e56162835951e7d7b4ccdac48c076daddc662f94125a7118/constriction-0.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "94e63aad0b68b64e5551131b6572d7106c28975b10466906542d1156ae87ac27",
"md5": "a7fbe5a86146f63499d48ba3cde13dd0",
"sha256": "4c6a6657505533f462fe103a2a32b14bed7e09719074e2caff7f7aa06559697a"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp311-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "a7fbe5a86146f63499d48ba3cde13dd0",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8",
"size": 301385,
"upload_time": "2024-10-23T10:16:46",
"upload_time_iso_8601": "2024-10-23T10:16:46.589333Z",
"url": "https://files.pythonhosted.org/packages/94/e6/3aad0b68b64e5551131b6572d7106c28975b10466906542d1156ae87ac27/constriction-0.4.1-cp311-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "48a4aaac45be6506cc6a9fe00ff6e3a4481e697c5a39bacfecc8b473dbdbc740",
"md5": "604552317db661563b3e503428e9fbff",
"sha256": "3e1acbb1ab80297730aa8b60c7d28306f557149a298cae3c48ef974a63007862"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp312-cp312-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "604552317db661563b3e503428e9fbff",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8",
"size": 817349,
"upload_time": "2024-10-23T10:18:03",
"upload_time_iso_8601": "2024-10-23T10:18:03.807582Z",
"url": "https://files.pythonhosted.org/packages/48/a4/aaac45be6506cc6a9fe00ff6e3a4481e697c5a39bacfecc8b473dbdbc740/constriction-0.4.1-cp312-cp312-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "6defb4f3c1a58f9f169ed6513797f8ce4bd8aad807274a60c25a16f7d432e213",
"md5": "19763aa75f1f1f9b718e5094b80a22e4",
"sha256": "564f93bf485a624a8952c7701e3c5e384e8826e8e923d4b33f9b752b7f4efa1f"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp312-cp312-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "19763aa75f1f1f9b718e5094b80a22e4",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8",
"size": 412569,
"upload_time": "2024-10-23T10:18:05",
"upload_time_iso_8601": "2024-10-23T10:18:05.281371Z",
"url": "https://files.pythonhosted.org/packages/6d/ef/b4f3c1a58f9f169ed6513797f8ce4bd8aad807274a60c25a16f7d432e213/constriction-0.4.1-cp312-cp312-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "8b4b64a35c48a0b43f03d4be362877e25f469910f1d0585dd070546ccaa7d83d",
"md5": "25128f949097f0ee1478e2e1e1466ec3",
"sha256": "15d0e25689a0efaa02e8137829bc02895833205b142f03947283219fe5eb6d78"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "25128f949097f0ee1478e2e1e1466ec3",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8",
"size": 460495,
"upload_time": "2024-10-23T10:13:57",
"upload_time_iso_8601": "2024-10-23T10:13:57.838885Z",
"url": "https://files.pythonhosted.org/packages/8b/4b/64a35c48a0b43f03d4be362877e25f469910f1d0585dd070546ccaa7d83d/constriction-0.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "f43b3b3413058e1f507b3c6870c7c3f3b2d006529e2bd7f39f46b13c832d964c",
"md5": "b0dc802181e03f7451836caa9aff4913",
"sha256": "e9f2d0b9496d47f0888213e4ea72d7154f933efaeea76ef4549603a45327b61c"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp312-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "b0dc802181e03f7451836caa9aff4913",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8",
"size": 300363,
"upload_time": "2024-10-23T10:16:48",
"upload_time_iso_8601": "2024-10-23T10:16:48.049604Z",
"url": "https://files.pythonhosted.org/packages/f4/3b/3b3413058e1f507b3c6870c7c3f3b2d006529e2bd7f39f46b13c832d964c/constriction-0.4.1-cp312-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "58d4c84580411e2c63f9c29be195ba9d461960ce04936d09ec086aac8ec3ad29",
"md5": "cb4d575ff52cfc8e1fdfb07e3f9a4d1c",
"sha256": "582b728a57931f0fb995c0d56ad3ad57c7eea0ad8e216206acbc7da78aadcc12"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "cb4d575ff52cfc8e1fdfb07e3f9a4d1c",
"packagetype": "bdist_wheel",
"python_version": "cp313",
"requires_python": ">=3.8",
"size": 815671,
"upload_time": "2024-10-23T10:18:07",
"upload_time_iso_8601": "2024-10-23T10:18:07.412370Z",
"url": "https://files.pythonhosted.org/packages/58/d4/c84580411e2c63f9c29be195ba9d461960ce04936d09ec086aac8ec3ad29/constriction-0.4.1-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "498e211d84cbce742835a568976b7cc1b582138e08f5e55d65a5c4abe294f4d6",
"md5": "340a0470ff9e226d8b738a23df4e0650",
"sha256": "fd9076c942741460b7e5c999ca3aead4c72b8231c619d711c16a623815e21ebd"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp313-cp313-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "340a0470ff9e226d8b738a23df4e0650",
"packagetype": "bdist_wheel",
"python_version": "cp313",
"requires_python": ">=3.8",
"size": 411869,
"upload_time": "2024-10-23T10:18:08",
"upload_time_iso_8601": "2024-10-23T10:18:08.736054Z",
"url": "https://files.pythonhosted.org/packages/49/8e/211d84cbce742835a568976b7cc1b582138e08f5e55d65a5c4abe294f4d6/constriction-0.4.1-cp313-cp313-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "1381ef8a8276cb8a6cfc78e663127af0413e67c64ef69f2f3c70a584dd910e34",
"md5": "eea749aacb074fdf8a0b57ebc1635de4",
"sha256": "9d1ff5d05f7ba384a025acf6585fc78f79663d5f7379da5100c55af372b87a89"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "eea749aacb074fdf8a0b57ebc1635de4",
"packagetype": "bdist_wheel",
"python_version": "cp313",
"requires_python": ">=3.8",
"size": 459649,
"upload_time": "2024-10-23T10:13:59",
"upload_time_iso_8601": "2024-10-23T10:13:59.218743Z",
"url": "https://files.pythonhosted.org/packages/13/81/ef8a8276cb8a6cfc78e663127af0413e67c64ef69f2f3c70a584dd910e34/constriction-0.4.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e05ff91492ed4a1601392c79eca3ef65e8f9cc49db06168845724b5810867271",
"md5": "1426b36a2f1e823a7fa35712304f2713",
"sha256": "314c4e70a1322a839904a58d63c8ead373d42936ae4754fa496df9db452e8dd0"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp313-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "1426b36a2f1e823a7fa35712304f2713",
"packagetype": "bdist_wheel",
"python_version": "cp313",
"requires_python": ">=3.8",
"size": 299738,
"upload_time": "2024-10-23T10:16:49",
"upload_time_iso_8601": "2024-10-23T10:16:49.330623Z",
"url": "https://files.pythonhosted.org/packages/e0/5f/f91492ed4a1601392c79eca3ef65e8f9cc49db06168845724b5810867271/constriction-0.4.1-cp313-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "bb88b4c1ff72c148fefad1226b9a22600c42a718c487e3fb3d68ea6570243807",
"md5": "b7ba1efc3c4ef66b9c7c81f8801925a2",
"sha256": "2feb39a6bb55ab613ca5dad569ecb5cb2dd705de78a4e32c8c39dc8622c1ac5c"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp38-cp38-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "b7ba1efc3c4ef66b9c7c81f8801925a2",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.8",
"size": 814755,
"upload_time": "2024-10-23T10:18:09",
"upload_time_iso_8601": "2024-10-23T10:18:09.805614Z",
"url": "https://files.pythonhosted.org/packages/bb/88/b4c1ff72c148fefad1226b9a22600c42a718c487e3fb3d68ea6570243807/constriction-0.4.1-cp38-cp38-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "71e38954268f9c329b9479b937d839e50b3dc3c243e99065486a6fd6d4799c22",
"md5": "de35264998ad12f7147df2fb5d2ca19b",
"sha256": "fea787493a5a354dbd96b9f1e9a6a05cad29dca06d53fbf6c27c1c5a6c80cbb9"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp38-cp38-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "de35264998ad12f7147df2fb5d2ca19b",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.8",
"size": 413380,
"upload_time": "2024-10-23T10:18:10",
"upload_time_iso_8601": "2024-10-23T10:18:10.959960Z",
"url": "https://files.pythonhosted.org/packages/71/e3/8954268f9c329b9479b937d839e50b3dc3c243e99065486a6fd6d4799c22/constriction-0.4.1-cp38-cp38-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e8cfae3746a7f75d9f4673f6f49c6d47db21ea7825cd90d475427cf2e1de9cb4",
"md5": "62f578c54ab8a902efc051d1f81e1764",
"sha256": "40af92426ff4231134086785d6930c625074e939dbb44ab5ec3e58a4d8a3d99a"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "62f578c54ab8a902efc051d1f81e1764",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.8",
"size": 461682,
"upload_time": "2024-10-23T10:14:00",
"upload_time_iso_8601": "2024-10-23T10:14:00.894327Z",
"url": "https://files.pythonhosted.org/packages/e8/cf/ae3746a7f75d9f4673f6f49c6d47db21ea7825cd90d475427cf2e1de9cb4/constriction-0.4.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e5434ea1737a98350157b7e7a27a707d9209596c966bac3901f54fac189ea647",
"md5": "e6f20f3bdaf8275e9a5d1c4e59161824",
"sha256": "d51c9fa611e5374948cce35d6a0829d3a40fcfa443eeef6022114230ad19e903"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp38-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "e6f20f3bdaf8275e9a5d1c4e59161824",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.8",
"size": 301557,
"upload_time": "2024-10-23T10:16:51",
"upload_time_iso_8601": "2024-10-23T10:16:51.235619Z",
"url": "https://files.pythonhosted.org/packages/e5/43/4ea1737a98350157b7e7a27a707d9209596c966bac3901f54fac189ea647/constriction-0.4.1-cp38-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a603bedb0d6ef40748bd6704b9bda7ad8aa25fd1043bb52c6ae1fd00f99dbd79",
"md5": "0e68ac601ca40a18e7916093453bef21",
"sha256": "686aabfd0fd839097c627211310f4f6de213d85ead53b1035e10fe884c71610b"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp39-cp39-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"has_sig": false,
"md5_digest": "0e68ac601ca40a18e7916093453bef21",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8",
"size": 815312,
"upload_time": "2024-10-23T10:18:12",
"upload_time_iso_8601": "2024-10-23T10:18:12.686902Z",
"url": "https://files.pythonhosted.org/packages/a6/03/bedb0d6ef40748bd6704b9bda7ad8aa25fd1043bb52c6ae1fd00f99dbd79/constriction-0.4.1-cp39-cp39-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "d88ae1eb0cb4edfdcb798596fefed596f882ea16d6ff1277de06ec770a06df0a",
"md5": "958f6c1c4f1dc6b188befce5d045d5c1",
"sha256": "5f29dd82aef092a6792e6c1897a39d40830a0e7bef1e06b5e932b1be138a6106"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp39-cp39-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "958f6c1c4f1dc6b188befce5d045d5c1",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8",
"size": 413507,
"upload_time": "2024-10-23T10:18:14",
"upload_time_iso_8601": "2024-10-23T10:18:14.612184Z",
"url": "https://files.pythonhosted.org/packages/d8/8a/e1eb0cb4edfdcb798596fefed596f882ea16d6ff1277de06ec770a06df0a/constriction-0.4.1-cp39-cp39-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "29bb486ab03562d2894f68bcb5e0db86ef6393858a699ecc6738c49f0bf9943f",
"md5": "0243d4b984f09112e79f3f6e3b8c92e9",
"sha256": "14f4c3b0213d8caefd38f1526e8d5cbf07979640583c4c2404fcb78f6bbcb917"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "0243d4b984f09112e79f3f6e3b8c92e9",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8",
"size": 461844,
"upload_time": "2024-10-23T10:14:02",
"upload_time_iso_8601": "2024-10-23T10:14:02.393605Z",
"url": "https://files.pythonhosted.org/packages/29/bb/486ab03562d2894f68bcb5e0db86ef6393858a699ecc6738c49f0bf9943f/constriction-0.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "d12095ea1c1482f234a511ab0c58802ec1c21a150b7a095ec1cac7807891230a",
"md5": "42f41c82835b72590b3336410961997b",
"sha256": "693e74cf9ed7bdb4e36596281fb492c01fbe7c1d5b2c036d69597c97c7b393dc"
},
"downloads": -1,
"filename": "constriction-0.4.1-cp39-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "42f41c82835b72590b3336410961997b",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8",
"size": 302310,
"upload_time": "2024-10-23T10:16:53",
"upload_time_iso_8601": "2024-10-23T10:16:53.282525Z",
"url": "https://files.pythonhosted.org/packages/d1/20/95ea1c1482f234a511ab0c58802ec1c21a150b7a095ec1cac7807891230a/constriction-0.4.1-cp39-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-10-23 10:17:58",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "bamler-lab",
"github_project": "constriction",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "constriction"
}
JSON
