twodlearn 0.5.0

Easy development of machine learning models

A library to develop machine learning models.

A. Installation

• 1. Install the desired version of tensorflow (CPU or GPU)

  pip install tensorflow        # for CPU
  pip install tensorflow-gpu    # for GPU

• 2. Clone the project

  git clone [e-mail]:danmar3/twodlearn.git twodlearn
  cd twodlearn

• 3. Install the project

  pip install -e .

• 4. Install extras (optional)

  pip install -e .[reinforce]
  pip install -e .[development]

B. Run the tests using pytest

install pytest pip install -U pytest

run the unit-tests using pytest:

cd twodlearn/tests/
pytest -ra                # print a short test summary info at the end of the session
pytest -x --pdb           # drop to PDB on first failure, then end test session
pytest --pdb --maxfail=3  # drop to PDB for first three failures
pytest --durations=10     # get the test execution time
pytest --lf               # to only re-run the failures.
pytest --cache-clear      # clear the cache of failed tests

Roadmap for v0.6

• [x] migrate to TF 1.14

• [ ] add documentation

• [ ] add project to pypi

• [ ] create LayerNamespace

• [x] add a shortcut for required and optional input arguments

• [x] add check_arguments method to Layer and TdlModel

• [x] get_parameters now supports nested structures and nested SimpleNamespace

• [ ] deprecate tuple initialization

• [ ] deprecate optim

• [ ] move feedforward to dense

• [ ] cleanup common: clean deprecated descriptors and put them in separate file

• [ ] remove redundant base classes, such as TdlObject

• [ ] deprecate templates and design a format for estimators

• [ ] deprecate options value

• [ ] deprecate pyfmi and jmodelica