Testing BMI with torch tensors

Installation

From Github

# install `bmi_pytorch` package
pip install "git+https://github.com/hellkite500/bmi_pytorch@main#egg=bmi_pytorch&subdirectory=bmi_pytorch"

# install `bmi_sdk` package
pip install "git+https://github.com/hellkite500/bmi_pytorch@main#egg=bmi_sdk&subdirectory=bmi_sdk"

From Source

# clone repo
git clone https://github.com/hellkite500/bmi_pytorch && cd bmi_pytorch

# create python virtual environment (python >= 3.8 required)
python -m venv venv
source ./venv/bin/activate

# install `bmi_pytorch` package
pip install ./bmi_pytorch

# install `bmi_sdk` package
pip install ./bmi_sdk/

Local Development

pre-commit hooks

We use git pre-commit hooks to run tasks like code formatting and linting. After cloning the repo and setting up a python virtual environment, install pre-commit with:

pip install pre-commit

Next, run pre-commit install (you must run from a directory in your repo clone). That's all! Now, whenever you make a commit, pre-commit will run hooks to ensure that your code is properly formatted and linted!

Running Tests

We use pytest to write and run our tests. Do the following to build and run tests:

# clone repo
git clone https://github.com/hellkite500/bmi_pytorch && cd bmi_pytorch

# create python virtual environment (python >= 3.8 required)
python -m venv venv
source ./venv/bin/activate

# install `bmi_pytorch` package
pip install "./bmi_pytorch[test]"
# install `bmi_sdk` package
pip install "./bmi_sdk/[test]"

# run tests
pytest

model.py

Contains an intitial multi-layer torch neural network for testing.

Note

TODO add a model_bmi.py module

Note

TODO add a Model.update() and connect to BMI update function