Discussions

• 1 You must be logged in to vote

  💬

  Introducing TorchMetrics- Deep learning metrics for scalable PyTorch models

  edenlightning started Mar 12, 2021 in General

  

  

  0
• 3 You must be logged in to vote

  💬

  How to apply KFold in pytorch_lightning

  AliceNEET started Mar 12, 2021 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  Why optimizer_zero_grad() is behind backward()

  AliceNEET started Mar 11, 2021 in General

  

  

  2
• 1 You must be logged in to vote

  ⚡

  How do I push a whole dataloader through a LightningModule?

  wittenator asked Mar 12, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  It seems like the rng_state is not changing after a call of a random method.

  dhorka started Mar 11, 2021 in General

  

  

  1
• 3 You must be logged in to vote

  💡

  IMPORTANT Add a robots.txt to stop Google from indexing Deprecated Versions of the Docs

  aribornstein started Mar 12, 2021 in Idea pool

  

  

  2
• 1 You must be logged in to vote

  💬

  Does anyone know how computation requirement scales with an increase in dataset size?

  seunggs started Mar 10, 2021 in General

  

  

  3
• 1 You must be logged in to vote

  ⚡

  Custom training loop for LightningModule

  imirzadeh asked Mar 10, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  2
• 2 You must be logged in to vote

  ⚡

  Does truncated_bptt_steps have any effect on the eval/test phases?

  colllin asked Mar 11, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  How can be the rng_state stored using the checkpoint_callbak.

  dhorka asked Mar 9, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  5
• 1 You must be logged in to vote

  💭

  HuggingFace Transformers

  chris-boson asked Mar 10, 2021 in code help: NLP / ASR / TTS · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🤖

  How to use DDP + FP16 without fit?

  turian asked Mar 10, 2021 in DDP / multi-GPU / multi-node · Unanswered

  

  

  2
• 1 You must be logged in to vote

  ⚡

  How to get global step from checkpoint?

  timothybrooks asked Mar 10, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 1 You must be logged in to vote

  ⚡

  Pretrain some sections of a model, then initialize those parts when training a full model

  stanleyshly asked Mar 10, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 1 You must be logged in to vote

  ⚡

  Scheduler.step() only called on the end of validation

  richarddwang asked Mar 10, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  2
• 2 You must be logged in to vote

  ⚡

  import pytorch_lightning as pl does not work on colab

  Bajo1994 asked Mar 9, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  

  

  2
• 1 You must be logged in to vote

  ⚡

  how to print loss every n steps without progress bar

  sun-peach asked Mar 10, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  1
• 1 You must be logged in to vote

  ⚡

  How set number of epochs

  Vichoko asked Nov 8, 2019 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  

  

  4
• 1 You must be logged in to vote

  ⚡

  changing val_check_interval from 0.01 to 0.005 changes number of steps in an epoch

  AyushP123 asked Mar 9, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 1 You must be logged in to vote

  🤖

  Predict on GPU and calculate metrics in CPU, DDP Mode

  frapercan asked Mar 9, 2021 in DDP / multi-GPU / multi-node · Unanswered

  

  

  0
• 2 You must be logged in to vote

  🤖

  DDP number of training iterations independent of #GPUs?

  timothybrooks asked Mar 3, 2021 in DDP / multi-GPU / multi-node · Unanswered

  

  

  5
• 1 You must be logged in to vote

  ⚡

  Should Lightning have separate setup and teardown hooks for each stage?

  ananthsub asked Mar 7, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  

  

  4
• 1 You must be logged in to vote

  ⚡

  Access datamodule in custom callbacks

  vishal-wiai asked Mar 8, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 1 You must be logged in to vote

  ⚡

  How to correctly apply metrics API in binary use case

  kapsner asked Mar 5, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  7
• 1 You must be logged in to vote

  🤖

  How to sequentially call fit() and test() in DDP

  kapsner asked Mar 5, 2021 in DDP / multi-GPU / multi-node · Answered

  

  

  2