-
Notifications
You must be signed in to change notification settings - Fork 3.3k
Issues: Lightning-AI/pytorch-lightning
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
torch.cuda.OutOfMemoryError after running tuner.scale_batch_size() in "binsearch" mode
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20231
opened Aug 26, 2024 by
rittik9
KeyError: 'Trying to restore optimizer state but checkpoint contains only the model. This is probably due to Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.2.x
ModelCheckpoint.save_weights_only being set to True.' But optim_cfg is in model
bug
#20230
opened Aug 26, 2024 by
CSteinhardt153
RuntimeError: each element in list of batch should be of equal size
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
#20229
opened Aug 26, 2024 by
loretoparisi
Dashboard
feature
Is an improvement or enhancement
needs triage
Waiting to be triaged by maintainers
#20227
opened Aug 25, 2024 by
qbilius
"FileExistsError: [Errno 17] File exists: '/000000_epoch_shape'" using the ddp_notebook strategy with data stored in MDS (mosaic streaming) format
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20226
opened Aug 25, 2024 by
elbamos
metric.compute() hangs when using DDP with multiple GPUs
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20223
opened Aug 23, 2024 by
manavkulshrestha
Can no longer install versions 1.5.10-1.6.5
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
#20220
opened Aug 21, 2024 by
JonathanBhimani-Burrows
NCCL error: Invalid rank requested
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
#20219
opened Aug 20, 2024 by
loretoparisi
using deepspeed in pytorch lightning, a bug occurred : RuntimeError: Function ConvolutionBackward0 returned an invalid gradient at index 1
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
#20218
opened Aug 20, 2024 by
hongsixin
Questions about loading a pre-trained model using lightnining CLI for continue training
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20217
opened Aug 20, 2024 by
HelloWorldLTY
Switching into training mode in training_step
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20216
opened Aug 19, 2024 by
heth27
Model does not update its weights
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20215
opened Aug 19, 2024 by
kopalja
You really should make the access to optimizers and schedulers more comprehensible and more detailed.
docs
Documentation related
needs triage
Waiting to be triaged by maintainers
#20214
opened Aug 19, 2024 by
onbigion13
KeyError:pytorch_lightning.utilities.argparse_utils
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
#20212
opened Aug 19, 2024 by
Aaron1993
ImportError: cannot import name '_TORCHMETRICS_GREATER_EQUAL_1_0_0' from 'pytorch_lightning.utilities.imports'
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.2.x
ver: 2.3.x
ver: 2.4.x
#20209
opened Aug 17, 2024 by
Horizon-369
Unexpected Behavior: Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.3.x
Fabric.load operates out-of-place on nested states
bug
#20208
opened Aug 17, 2024 by
Markus28
Training crash when using XLA profiler on XLA accelerator and manual optimization
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20206
opened Aug 16, 2024 by
sdsuster
Allow passing custom reader/writer in _distributed_checkpoint_save and _distributed_checkpoint_load.
feature
Is an improvement or enhancement
needs triage
Waiting to be triaged by maintainers
#20205
opened Aug 15, 2024 by
Yash9060
Loading a model changes pytorch random state
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20204
opened Aug 15, 2024 by
heth27
7x slower training speed when switching from lightning 1.0 to 2.0
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.1.x
ver: 2.2.x
ver: 2.3.x
ver: 2.4.x
#20201
opened Aug 14, 2024 by
MaiBe-ctrl
ModelCheckpoint Callback not working/saving unless Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.1.x
ver: 2.2.x
ver: 2.3.x
ver: 2.4.x
save_on_train_epoch_end is enabled True which considerably slows down training
bug
#20200
opened Aug 14, 2024 by
snknitin
LightningCLI: --help argument given after the subcommand fails
bug
Something isn't working
needs triage
Waiting to be triaged by maintainers
ver: 2.4.x
#20199
opened Aug 14, 2024 by
nisar2
Add param_group name for BaseFinetuningCallback
feature
Is an improvement or enhancement
needs triage
Waiting to be triaged by maintainers
#20194
opened Aug 13, 2024 by
Jserax
shortcuts for logging weights and biases norms
feature
Is an improvement or enhancement
needs triage
Waiting to be triaged by maintainers
#20190
opened Aug 11, 2024 by
heth27
Support IO Type Checkpoints for trainer.fit() in ckpt_path Parameter
feature
Is an improvement or enhancement
needs triage
Waiting to be triaged by maintainers
#20189
opened Aug 10, 2024 by
kimjw0623
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.

