Fine-Tune Training

[BryonLewis]

Add in a new command line argument --init-weights when running training and provide an input file that can be used to continue training on. This file can be a .zip, .pth, .pt, .py, .weights or .wt

• Add capability in Web version to have an optional input argument to training --init-weights which is an already trained model. This model could be a built-in model or user trained model.
• Add checkbox to training to do Fine-Tune Training
• Need endpoint to return all of the available models for fine-tune training.
• User interface to search the built-in models and user models to get a list of possible inputs. Should be able to use the existing pipelines file to to this.

Desktop