Modev

Main changes with respect to previous release:
* Validation functions now receive data, instead of indexes, which gives more freedom to generate train/test indexes.

For more details please visit:
https://pypi.org/project/modev/

Main changes with respect to previous release:
* Generalize execution.execute_model, so that:
* It receives as inputs model, data, train indexes for current fold, test indexes for current fold, and target. By doing this, we can include a preprocessing method inside the function, that applies to entire data, not train/test separately.
* It returns a dictionary of execution results (e.g. {'truth': np.array([...]), 'prediction': np.array([...])}). This allows to include other metrics, like timing ones.

For more details please visit:
https://pypi.org/project/modev/

Main changes with respect to previous release:
* Fix issue with setup.py.

For more details please visit:
https://pypi.org/project/modev/

Main changes with respect to previous release:
* Parameter test_mode belongs now to validation, not execution.
* Execution now gives more freedom to the approach (for example to do preprocessing).

For more details please visit:
https://pypi.org/project/modev/

Main changes with respect to previous release:
* Results (temporary or final) can be saved to and loaded from a local file.
* Pass train_x and test_x to approach as pandas dataframes instead of numpy arrays; train_y and test_y are still numpy arrays, for convenience.

Please visit:
https://pypi.org/project/modev/