wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-09 11:32:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

working on trainer docs

Browse Source
This commit is contained in:
William Falcon
2020-01-16 10:51:15 -05:00
parent d05f805b25
commit f353b021b5
1 changed files with 2 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pytorch_lightning/trainer/trainer.py
+2 -4
View File
@@ -396,10 +396,8 @@ class Trainer(TrainerIOMixin,
truncated_bptt_steps (int): Truncated back prop breaks performs backprop every k steps of a much longer sequence
If this is enabled, your batches will automatically get truncated
and the trainer will apply Truncated Backprop to it. Make sure your batches have a sequence dimension.
`Williams, Ronald J., and Jing Peng. "An efficient gradient-based algorithm for on-line training of recurrent network trajectories."
<http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.56.7941&rep=rep1&type=pdf>`_
(`Williams et al. "An efficient gradient-based algorithm for on-line training of recurrent network trajectories."
<http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.56.7941&rep=rep1&type=pdf>`_)
Example::
# default used by the Trainer (ie: disabled)
trainer = Trainer(truncated_bptt_steps=None)