ML4PD - an open-source libray for building Aspen-like process models via machine learning.
Project description
ml4pd
Process design with Machine Learning.
How to Set Up for Development
via Conda
- Clone repository with
git clone
. - Create new env. & install dependencies:
mamba env create -f ml4pd/environment.yml -n [env-name]
- Add repo to path with
conda develop ml4pd
. - Optional: register conda environment with jupyter notebook
python -m ipykernel install --user --name=ml4pd
Additional GitHub repositories for docs, tests and training
ml4pd_utils
: code base for generating & preparing data for training.ml4pd_models
: to store model files.autoaspen
: database for data obtained by aspen & python.
To debug docs
- Use
mkdocs serve
withinml4pd
directory, then go tolocalhost:8000
. generate_site.py
gets docstrings (written in makrdown) from classes and put them in the right directory.- To make changes to notebooks, add
ml4pd
to path withsys.path.append()
orconda develop
.
Relationship with ml4pd_models
To minimize manual work, ml4pd dependends on a specific ml4pd_models github branch. When changes are made to either ml4pd or ml4pd_models that will break compatibility, create new branch in ml4pd_models, and link requirements.txt with the new branch.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
ml4pd-2022.6.22.tar.gz
(23.8 kB
view hashes)
Built Distribution
ml4pd-2022.6.22-py3-none-any.whl
(30.1 kB
view hashes)
Close
Hashes for ml4pd-2022.6.22-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | db50f7f7bb3d8764c907c42cbd4f86ffecfff3780900b2ee532a659b1d00d466 |
|
MD5 | 2ae71fc39ee7b326bf3a3da5d873e237 |
|
BLAKE2b-256 | d6560cedf49ae0ca2616e9fe6f9266c98831904465aaa719e37d300344ccd144 |