Various callbacks to customize training behavior

class ShortEpochCallback[source]

ShortEpochCallback(pct=0.01, short_valid=True) :: Callback

Fit just pct of an epoch, then stop

learn = synth_learner(), cbs=ShortEpochCallback())
epoch train_loss valid_loss time
0 00:00
learn = synth_learner(), cbs=ShortEpochCallback(short_valid=False))
epoch train_loss valid_loss time
0 14.867975 00:00

class GradientAccumulation[source]

GradientAccumulation(n_acc=32) :: Callback

Accumulate gradients before updating weights

When the number of steps per accumulation is higher than the number of batches, the parameters (and therefore validation loss) don't change at all:

learn = synth_learner(), lr=0.01, cbs=GradientAccumulation(n_acc=1000))
# ensure valid_loss didn't change
assert learn.recorder.values[-1][1] == learn.recorder.values[0][1]
epoch train_loss valid_loss time
0 10.941168 10.280428 00:00

class GradientClip[source]

GradientClip(max_norm:float=1.0, norm_type:float=2.0) :: Callback

Clip norm of gradients

Normally if we use a learning rate that is too high, our training will diverge. This even happens if we use mixed precision training, which avoid infinities by using dynamic loss scaling, but still diverges:

fp16 = MixedPrecision()
learn = synth_learner(lr=1.1, cuda=True), cbs=fp16)
epoch train_loss valid_loss time
0 38.214169 25.269012 00:00
1 377.146088 890.011780 00:00
2 839.391907 9965.712891 00:00

By adding the GradientClip callback, the gradient norm_type (default:2) norm is clipped to at most max_norm (default:1) using nn.utils.clip_grad_norm_, which can avoid loss divergence:

learn = synth_learner(lr=1.1, cuda=True), cbs=[GradientClip,fp16])
epoch train_loss valid_loss time
0 2.039427 2.372183 00:00
1 1.402424 0.300724 00:00
2 1.013551 0.332668 00:00



set_bn_eval(m:Module, use_eval=True)

Set bn layers in eval mode for all recursive children of m.

class BnFreeze[source]

BnFreeze(after_create=None, before_fit=None, before_epoch=None, before_train=None, before_batch=None, after_pred=None, after_loss=None, before_backward=None, before_step=None, after_cancel_step=None, after_step=None, after_cancel_batch=None, after_batch=None, after_cancel_train=None, after_train=None, before_validate=None, after_cancel_validate=None, after_validate=None, after_cancel_epoch=None, after_epoch=None, after_cancel_fit=None, after_fit=None) :: Callback

Basic class handling tweaks of the training loop by changing a Learner in various events

BnFreeze is useful when you'd like to train two separate models that have a common feature extractor / body. The only part of the model that's different is the head that you attach for transfer learning.

Learner.freeze()) doesn't suffice here as the BatchNorm layers are trainable by default, and running mean and std of batches are tracked. For feature extractors to fully match, you need to set train_bn=False and these stats need to be frozen as well, which is precisely the function of BnFreeze.

path = untar_data(URLs.MNIST_TINY)
dls  = ImageDataLoaders.from_folder(path, valid_pct=0.2)

We first demonstrate the mismatch of the running stats when using only train_bn=False, by creating a Learner...:

learn1 = cnn_learner(deepcopy(dls), resnet18, pretrained=True, train_bn=False)

...and grab the first BatchNorm layer, and store its running mean:

m = learn1.model[0][1].running_mean.clone()

You can see that now that running mean has changed:, lr=0.02)
test_ne(to_detach(learn1.model[0][1].running_mean), m)
epoch train_loss valid_loss time
0 1.152701 0.468892 00:02

When we use the BnFreeze callback, the running statistics will not be changed during training. This is often important for getting good results from transfer learning.

learn1 = cnn_learner(deepcopy(dls), resnet18, pretrained=True, train_bn=False, cbs=BnFreeze)
m = learn1.model[0][1].running_mean.detach().clone(), lr=0.02)
test_eq(to_detach(learn1.model[0][1].running_mean), m)
epoch train_loss valid_loss time
0 0.488634 0.277683 00:02