get_imaggenette[source]

get_imaggenette()

class CustomLearner[source]

CustomLearner(model, loss_func, optimizer, db)

get_learner[source]

get_learner()

Helper function to get learner

get_runner[source]

get_runner(callbacks=None)

class ProgressCallback[source]

ProgressCallback() :: Callback

Callback to make a nice progress bar with metrics for training. Slightly modified version of: https://github.com/fastai/course-v3/blob/master/nbs/dl2/09c_add_progress_bar.ipynb

class TrainEvalCallback[source]

TrainEvalCallback() :: Callback

Keeps track of training/eval mode of model and progress through training

accuracy[source]

accuracy(preds, targ, *args)

Compute accuracy of preds with respect to targ

class CudaCallback[source]

CudaCallback() :: Callback

Base class for callbacks, defines order of execution and allows abstraction of self to runner class

class Runner[source]

Runner(learner, cbs=None)

All encompossing class to train a model with specific callbacks

!python notebook2script.py ScratchToPytorch.ipynb

Converted ScratchToPytorch.ipynb to ModernArchitecturesFromPyTorch/nb_ScratchToPytorch.py