Compilation of Quick Features

[ejbrannonwu]

Running thru all of the TODOs that we've been marking as we completed other tasks...

• Figure out progression bar (SCRAPPED)
• Fix proper blockchain update command
• Add db types to connect_to_db()
• If training process is in-progress, do not allow another call to /start-training
• If training process is in-progress, do not allow another call to /continue-training
• Add another endpoint to manually stop training process
• Fix "rare" bug (line 333 in aggregator_server.py - "when training two different models and both are in-progress, one of them may stop when update_minParams endpoint is called...or if a node is added mid-way...not sure" - only occurred once)
• Update direct_inference() in node_server.py
• Make pathing for init-commands be easier for the user (?)
• new line after curl response
• allow MNIST direct-inference input to accept raw image data
• Add a message at the end of a training process (via logger) to indicate the training has finished
• check that module_path exists (if not, don't initialize)
• node checked to see if actually initialized (shows success even if off)

[ejbrannonwu]

For Figure out Progression bar:

• Tried to add index within Progression Bar; bar comes from Keras model in data_handlers, could not add additional content within the bar itself (or on the same line, would be overwritten)
• Tried to add message ("index: ===========Model Training Progress======"; would fail if there was a delay in training when running multiple models, left the work commented

tldr; decided to scrap and move on

[ejbrannonwu]

edit 05/15: completed features that were attached to recent PR 

• Add a message at the end of a training process (via logger) to indicate the training has finished
• check that module_path exists (if not, don't initialize)
• node checked to see if actually initialized (shows success even if off)