Add a template generic example script as a quick start with Ignite

🚀 Feature

Idea is to simplify user's learning curve and provide a template example that can be copied by the user and modified to his/her needs. Something like that but an executable working example:

def training():

    train_loader, val_loader = get_dataflow(fold_index=fold_index, **kwargs)
    model, optimizer, criterion, lr_scheduler = initialize_model(**kwargs)
    metrics = {
         "accuracy": Accuracy(),
    }

    trainer = create_supervised_trainer(model, optimizer, criterion)

    @trainer.on(Events.EPOCH_COMPLETED)
    def update_lr_scheduler(_):
        lr_scheduler.step()

    evaluator = create_supervised_evaluator(model, metrics=metrics)

    @trainer.on(Events.ITERATION_COMPLETED(every=validate_every))
    def validate(trainer):
        evaluator.run(val_loader)
        metrics = evaluator.state.metrics
        print("After {} iterations, binary accuracy = {:.2f}"
              .format(trainer.state.iteration, metrics['accuracy']))

    trainer.run(train_loader, max_epochs=kwargs.get(max_epochs, 100))
    

Start from the proposed Python training snippet and inspect Ignite's existing training examples and APIs for data loading, model initialization, trainers, evaluators, metrics, and scheduler updates. The work is done when users have a copyable, executable generic quick-start example that follows this training flow.