Skip to content

Actions: Lightning-AI/pytorch-lightning

Actions

Label Issues

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
labeler-issue.yml
788 workflow runs
788 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[CLI] predict(ckpt_path="best") raises ValueError even though ModelCheckpoint is configured under trainer.callbacks Label Issues #3169: Issue #21254 opened by oooooio
16s
16s
Directly save checkpoint to specified disk location Label Issues #3168: Issue #21253 opened by HarveyYan
12s
12s
[WandbLogger] resume="never" without id still tries to resume existing run Label Issues #3167: Issue #21248 opened by oooooio
12s
12s
Lost hp_metric when call self.trainer.test on fit end Label Issues #3166: Issue #21247 edited by razgzy
8s
8s
Lost hp_metric when call self.trainer.test on fit end Label Issues #3165: Issue #21247 edited by razgzy
11s
11s
Lost hp_metric when call self.trainer.test on fit end Label Issues #3164: Issue #21247 opened by razgzy
10s
10s
CLI parsing error: fit -c run.yaml works but -c run.yaml fit fails Label Issues #3163: Issue #21245 opened by oooooio
11s
11s
Redundant Host & Device Synchronizations Label Issues #3162: Issue #21232 edited by alpha0422
9s
9s
Redundant Host & Device Synchronizations Label Issues #3161: Issue #21232 opened by alpha0422
8s
8s
flop_counts missing for deepspeed in 2.5.5 Label Issues #3160: Issue #21231 opened by bnestor
10s
10s
Support wandb_logger.watch() when using LightningCLI Label Issues #3159: Issue #19791 edited by Borda
10s
10s
LightningCLI trainer_defaults with callbacks raises "Expected a parent" Label Issues #3158: Issue #21189 edited by ANKer661
13s
13s
LightningCLI trainer_defaults with callbacks raises "Expected a parent" Label Issues #3157: Issue #21189 opened by ANKer661
14s
14s
Feature request: Support splitting model weights and training states into separate checkpoint files Label Issues #3156: Issue #21170 opened by yilin404
11s
11s
Subtract n from batch_size found by BatchSizeFinder with mode='binsearch' Label Issues #3155: Issue #21154 opened by sergey-protserov-uhn
8s
8s
[BUG] Learning Rate finder does not find optimal lr anymore since version 2.5.3 Label Issues #3154: Issue #21141 opened by dennisbader
9s
9s
Lightning+DDP: use_distributed_sampler=True always shuffles data in DDP despite using custom sampler Label Issues #3153: Issue #21131 opened by dmholtz
11s
11s
FSDP with custom state_dict() Label Issues #3152: Issue #21124 opened by LiuTaowen-Tony
10s
10s
batch sampler, ddp, and sampler wrapper Label Issues #3151: Issue #21122 opened by Jserax
8s
8s
Lightning log always log in working dir, together with the dir I specify Label Issues #3150: Issue #21118 opened by Mo-Junyang
10s
10s
Comet logger does not support "after_save_checkpoint" method? Label Issues #3149: Issue #21111 opened by hieubnt235
9s
9s
ModelCheckpoint options save_last='link' and save_top_k=-1 cause recursive symlink creation Label Issues #3148: Issue #21110 opened by Hippskill
13s
13s
Mis-aligned information in RichModelSummary with `DeepSpeedStrategy Label Issues #3147: Issue #21099 edited by GdoongMathew
11s
11s
Mis-aligned information in RichModelSummary with `DeepSpeedStrategy Label Issues #3146: Issue #21099 opened by GdoongMathew
39s
39s
Checkpoints not saving: global_step and current_epoch not updating despite correct progress bar Label Issues #3145: Issue #21096 edited by homjay
11s
11s