Discussions

• 1 You must be logged in to vote

  🤖

  WandB Logger in Single Machine + Multi-GPU DDP setting

  angadkalra asked Mar 29, 2023 in DDP / multi-GPU / multi-node · Closed · Unanswered

  

  

  1
• 1 You must be logged in to vote

  😎

  RAM usage increases quickly over the training step

  celsofranssa asked Mar 29, 2023 in code help: CV · Closed · Unanswered

  

  

  0
• 2 You must be logged in to vote

  ⚡

  torch.compile - exclude certain pl.Trainer functions

  mishooax asked Mar 28, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Closed · Unanswered

  

  

  1
• 1 You must be logged in to vote

  💬

  Is there a markdown-badges for PyTorch Lighting?

  Keiku started Mar 30, 2022 in General

  

  

  

  

  3
• 1 You must be logged in to vote

  ⚡

  Resampling a Dataset every epoch through the LightningModule on_train_epoch_start hook

  Kominaru asked Mar 25, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  AttributeError: 'Trainer' object has no attribute 'lr_find'

  ozgugoksu asked Feb 22, 2022 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  

  

  3
• 1 You must be logged in to vote

  🤖

  Single-Node multi-GPU Deepspeed training fails with cuda OOM on Azure

  gabriead asked Mar 23, 2023 in DDP / multi-GPU / multi-node · Unanswered

  

  

  0
• 6 You must be logged in to vote

  ⚡

  How to reset train data loader every epoch end??

  elephantoid asked Mar 21, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  1
• 2 You must be logged in to vote

  🤖

  Why DeepSpeed executes all the code several times?

  igonro asked Mar 23, 2023 in DDP / multi-GPU / multi-node · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  gradient optimizer for PyTorch Temporal Fusion Transformer (TFT)

  grosestq asked Mar 11, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 3 You must be logged in to vote

  🤖

  Using log() and log_dict() in pl 1.9.4.

  vgthengane asked Mar 22, 2023 in DDP / multi-GPU / multi-node · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  Compiled model, self.log() returns inconsistent values ?

  w2ex asked Mar 22, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  on_training_epoch_end is never called

  bavShehata asked Mar 20, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Closed · Answered

  

  

  1
• 1 You must be logged in to vote

  💬

  How to get the best epoch number?

  Struggle-Forever started Mar 19, 2023 in General

  

  

  1
• 1 You must be logged in to vote

  ⚡

  Trained weights stored on CPU despite the model being trained on GPU

  bavShehata asked Mar 20, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🤖

  FSDP with named_parameters for disabling weight_decay on some parameters.

  fmocking asked Mar 19, 2023 in DDP / multi-GPU / multi-node · Unanswered

  

  

  0
• 1 You must be logged in to vote

  ⚡

  Efficient Prediction Writing for 2D/3D Medical Image Segmentation Tasks

  ndahiya3 asked Mar 15, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 2 You must be logged in to vote

  ⚡

  Cannot import the accuracy, f1 score and accuracy from the pytorch lightning metric library

  shubheshswain91 asked Oct 29, 2021 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  

  

  7
• 1 You must be logged in to vote

  💭

  Constructing KNN index in the end of each epoch.

  gfx73 asked Mar 1, 2023 in code help: NLP / ASR / TTS · Unanswered

  

  

  20
• 2 You must be logged in to vote

  ⚡

  Getting ValueError Error when using ModelCheckpoint with auto_insert_metric_name=False

  quy-ng asked Jan 16, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Answered

  

  

  1
• 3 You must be logged in to vote

  🤖

  Disabling find_unused_parameters

  strategy: ddp DistributedDataParallel

  rahulvigneswaran asked Jan 30, 2022 in DDP / multi-GPU / multi-node · Closed · Answered

  

  

  

  

  3
• 2 You must be logged in to vote

  ⚡

  Lightning Flash: How to configure optimizer with filter

  brianlow asked Mar 12, 2023 in Lightning Trainer API: Trainer, LightningModule, LightningDataModule · Unanswered

  

  

  0
• 3 You must be logged in to vote

  🤖

  How does validation loop work ?

  Honzys asked Oct 14, 2020 in DDP / multi-GPU / multi-node · Unanswered

  

  

  

  

  2
• 1 You must be logged in to vote

  🤖

  Current best practices to initialize massive (50B parameter+) models

  azton asked Mar 3, 2023 in DDP / multi-GPU / multi-node · Answered

  

  

  

  

  4
• 1 You must be logged in to vote

  💬

  ProcessExitedException: process 0 terminated with signal SIGABRT

  Bk073 started Mar 8, 2023 in General

  

  

  0