This package implements a framework for streaming 1D convolutions in PyTorch without padding or pseudo-streaming/cross-fading.

pip install torch-streamer

Docs are available at torch-streamer.readthedocs.io.

The following script creates a virtual environment using pyenv for the project and installs dependencies with uv.

pyenv install 3.10
pyenv virtualenv 3.10 torch-streamer
bin/deps

You can also use pre-commit with the project to run tests, etc. at commit time.

pre-commit install

Testing, formatting, and static checking can all be done with pre-commit at any time.

pre-commit run --all-files

There is also a watcher script that can be used to run these whenever a file changes.

bin/watch

The project uses MkDocs with mkdocstrings for documentation, and you can start a mkdocs web server to test/edit documentation as follows.

bin/docserve

Documentation is hosted by Read the Docs and will automatically update when the main branch is merged.

The library can be updated on the main PyPI repository as follows.

bin/release pypi

If needed, you can release to the test PyPI repository with this command.

bin/release pypi-test

Copyright © 2024 Brent M. Spell

Licensed under the MIT License (the "License"). You may not use this package except in compliance with the License. You may obtain a copy of the License at

https://opensource.org/licenses/MIT

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.