Search results

overfit_batches replaces custom sampler with Sequential sampler

bugSomething isn't workingSomething isn't working

needs triageWaiting to be triaged by maintainersWaiting to be triaged by maintainers

ver: 2.5.x

Status: Open.

#21282 In Lightning-AI/pytorch-lightning;

· nilsleh opened on Oct 12, 2025

LightingModel.save_checkpoint()

checkpointingRelated to checkpointingRelated to checkpointing

featureIs an improvement or enhancementIs an improvement or enhancement

Status: Open.

#21280 In Lightning-AI/pytorch-lightning;

· Karesto opened on Oct 10, 2025

SaveConfigCallback does not work as expected and raises an error under specific settings (DDPStrategy start_method="spawn")

bugSomething isn't workingSomething isn't working

lightningclipl.cli.LightningCLIpl.cli.LightningCLI

strategy: ddpDistributedDataParallelDistributedDataParallel

ver: 2.5.x

Status: Open.

#21277 In Lightning-AI/pytorch-lightning;

· lengmo1996 opened on Oct 9, 2025

TQDMProgressBar bug (?) causes DDP to hang

progress bar: tqdm

questionFurther information is requestedFurther information is requested

strategy: ddpDistributedDataParallelDistributedDataParallel

Status: Open.

#21264 In Lightning-AI/pytorch-lightning;

· tchittesh opened on Oct 4, 2025

Test fails on MPS due to unsupported float64 precision

accelerator: mpsApple Silicon GPUApple Silicon GPU

bugSomething isn't workingSomething isn't working

tests

ver: 2.5.x

Status: Open.

#21261 In Lightning-AI/pytorch-lightning;

· caiopetruccirosa opened on Oct 3, 2025

Unable to set <code>val_check_interval</code> to an integer with the Lightning CLI

bugSomething isn't workingSomething isn't working

lightningclipl.cli.LightningCLIpl.cli.LightningCLI

ver: 2.5.x

Status: Open.

#21259 In Lightning-AI/pytorch-lightning;

· fschlatt opened on Oct 2, 2025

Bug in ThroughputMonitor.on_validation_end: using sum() instead of last value can corrupt

bugSomething isn't workingSomething isn't working

callback: throughput

ver: 2.5.x

Status: Open.

#21257 In Lightning-AI/pytorch-lightning;

· itzhakstern opened on Sep 30, 2025

<code>torch.set_float32_matmul_precision</code> deprecation

refactor

Status: Open.

#21256 In Lightning-AI/pytorch-lightning;

· cw-tan opened on Sep 29, 2025

Allow weight reuse in a different lightning module

Status: Open.

#21255 In Lightning-AI/pytorch-lightning;

· ziw-liu opened on Sep 29, 2025

[CLI] predict(ckpt_path="best") raises ValueError even though ModelCheckpoint is configured under trainer.callbacks

bugSomething isn't workingSomething isn't working

checkpointingRelated to checkpointingRelated to checkpointing

lightningclipl.cli.LightningCLIpl.cli.LightningCLI

trainer: predict

ver: 2.5.x

Status: Open.

#21254 In Lightning-AI/pytorch-lightning;

· oooooio opened on Sep 29, 2025

Directly save checkpoint to specified disk location

checkpointingRelated to checkpointingRelated to checkpointing

questionFurther information is requestedFurther information is requested

Status: Open.

#21253 In Lightning-AI/pytorch-lightning;

· HarveyYan opened on Sep 29, 2025

Redundant Host &amp; Device Synchronizations

accelerator: cudaCompute Unified Device Architecture GPUCompute Unified Device Architecture GPU

refactor

Status: Open.

#21232 In Lightning-AI/pytorch-lightning;

· alpha0422 opened on Sep 21, 2025