Bug fixes, tool additions

• Fixed bug where Server.train() did not obtain shuffled sets from its DataFrame if the user is shuffling sets for each candidate neural network
• Server._sets is now public instead of private
• Removed unnecessary import in ecnet\model.py
• Added a suite of tools for creating an ECNet-formatted database:
  • tools\create_ecnet_db.py (aggregates SMILES, QSPR descriptors for database creation)
  • tools\name_to_smiles.py (uses PubChemPy to obtain SMILES strings for supplied molecule names)
  • tools\smiles_to_qspr.py (generates QSPR descriptors from SMILES strings using Open Babel and PaDEL-Descriptor)
  • added version of PaDEL-Descriptor
  • added how-to-use documentation (tools\README.md)

Multiprocessing and More

• Rework of Server method arguments, functionality
  • This includes an API change (check your scripts!)
• Rework of Server training/selection algorithms
• Improvements to logging
• Rework of model saving/loading
• When input dimensionality is reduced, selected inputs are applied to current data immediately
• Updates to hyperparameter tuning
  • Can now shuffle data sets for each bee
  • Supports ECabc 2.1.0

Logging, DocStrings and Documentation

• Added support for logging with the ColorLogging package
• Updated docstrings, uses Google's documentation format
• Modules now only import functions/classes they use instead of whole packages
• Added unit testing for Server methods
• Updated genetic limiting per PyGenetics 0.5.0 update
• Made project variables accessible
• Updated documentation
  • Moved Server method documentation to ecnet/README.md
  • Updated README.md to include more code block examples
  • Updated terminology, project scope, clarity

Multiprocessing Support for Input Parameter Dimensionality Reduction

• Added support for PyGenetics multiprocessing for limiting input parameter dimensionality

Bug fixes for saving/opening projects from other directories

• Full paths for project_file supplied to Server will now be utilized
• Model configuration file will not be extracted to the current working directory

Bug fixes for input parameter limiting

• Fixed issue with data shuffling for genetic algorithm population members
• Fixed case where feedback printing variable was not defined when calling limiting function

1.5 update - pep8 styling, rework of server, model classes/methods

• Changes to Server methods:
  • open_project -> __open_project, now called with optional argument on Server initialization
  • Updated default arguments for create_project, import_data, train_model, select_best
  • Configuration .yml file now only contains model architecture/learning variables
    • Variable names renamed for clarity
  • tune_hyperparameters now scales to predefined number of hidden layers
• Changes to ecnet/model.py methods:
  • Activation functions are now stored in ACTIVATION_FUNCTIONS, a dictionary of callable functions
  • corresponding to supplied activation functions
  • Layer objects house activation functions corresponding to ACTIVATION_FUNCTIONS
  • __feed_forward uses the activation function stored in Layer for computation
• Updates to documentation/examples reflecting above changes
• Code style == pep8

Fixed project opening (sets project_name before unzipping)

1.4.6

Fixed project importing for real this time

Update to project saving/using

Upon saving a project, the current Server configuration is now included in the .project file; when opening the .project file for use, the configuration found in the .project file is imported as the Server's configuration.

Update to docstrings and code styling

• General cleanup for easy-reading and documentation