Torch compile and Lightning CLI
|
|
2
|
1345
|
October 30, 2023
|
Converting PyTorch implementation to PyTorch Lightning for Graph Neural Networks
|
|
2
|
95
|
October 29, 2023
|
Re-train the fine tune model for new class
|
|
2
|
356
|
October 29, 2023
|
The time proportion of each module in pre-training process
|
|
0
|
66
|
October 27, 2023
|
Size mismatch for model
|
|
1
|
286
|
October 26, 2023
|
Yielding batches from training dataloaders at different frequencies
|
|
0
|
59
|
October 26, 2023
|
Dose batch norm need to convert to SyncBatchNorm
|
|
2
|
80
|
October 26, 2023
|
Ignore log in one of the GPUs as it does not have a specific loss
|
|
2
|
76
|
October 24, 2023
|
Training slowing down
|
|
1
|
81
|
October 24, 2023
|
How to use seed everything in version 2.1.0 for pytorch 2.0.1
|
|
1
|
98
|
October 24, 2023
|
PyTorch Lightning CLI with Optuna Hyperparameter search - Hot to set PruningCallback?
|
|
1
|
125
|
October 24, 2023
|
How to set some special layers to float32 when training use mix-precision float16
|
|
2
|
116
|
October 24, 2023
|
Metrics not logged properly in PyTorch Lightning
|
|
1
|
156
|
October 22, 2023
|
Resume training by loading only the optimizer states in deepspeed enabled training
|
|
0
|
85
|
October 20, 2023
|
Are on_fit_end and on_train_end the same?
|
|
4
|
3882
|
October 19, 2023
|
Best way to wrap a LightningModule to report generic metrics
|
|
0
|
79
|
October 18, 2023
|
ValueError: too many values to unpack (expected 3)
|
|
3
|
156
|
October 18, 2023
|
Question about recover nested model from checkpoint
|
|
0
|
113
|
October 17, 2023
|
How to fix: RuntimeError: mat1 and mat2 shapes cannot be multiplied (256x4096 and 1024x4)?
|
|
0
|
181
|
October 17, 2023
|
How to not load complete in-memory dataset for every process in DDP training
|
|
2
|
3087
|
October 17, 2023
|
How to use IPU accelerator in paperspace
|
|
1
|
147
|
October 16, 2023
|
Accumulate_grad_batches and learning rate
|
|
1
|
139
|
October 14, 2023
|
Import lightning fails in pop os 22.04(nvidia)
|
|
1
|
210
|
October 14, 2023
|
Data not loading when num_workers>0
|
|
1
|
138
|
October 10, 2023
|
Initialize model with data before training
|
|
1
|
532
|
October 9, 2023
|
How to install gpu version of pyroch-lightning?
|
|
3
|
702
|
October 9, 2023
|
Training/predicting takes forever before predict_step is even called
|
|
2
|
125
|
October 7, 2023
|
Error with ddp when updating from pytorch-lightning 1.6.5 to version2.0.9
|
|
0
|
281
|
October 4, 2023
|
Multi-task model in version 2.0.9 with DDP error
|
|
0
|
215
|
October 4, 2023
|
Logging one value per epoch?
|
|
0
|
107
|
October 4, 2023
|