issues
search
LarsKue
/
lightning-trainable
A default trainable module for pytorch lightning.
MIT License
10
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Reworked hypersphere dataset to include cube sampling
#29
thelostscout
closed
4 months ago
2
Added Gaussian Mixture Models as a toy distribution
#28
thelostscout
closed
4 months ago
12
Make toy models deterministic
#27
thelostscout
closed
4 months ago
1
omegaconf
#26
ThisIsForReview
closed
9 months ago
1
No last.ckpt
#25
ThisIsForReview
closed
9 months ago
1
Cannot run lightning-trainable example
#24
ThisIsForReview
closed
9 months ago
6
Allow single-layer (linear) `FullyConnectedNetwork`
#23
LarsKue
closed
10 months ago
1
No support for cpu only
#22
thelostscout
closed
10 months ago
2
HParams with dicts (i.e. almost all) store dicts as special AttributeDict type in `hparams.yaml`
#21
fdraxler
closed
10 months ago
3
HParams validation batch has `self.training = False`
#20
fdraxler
closed
10 months ago
1
Test case for workaround for hparams loading issues
#19
fdraxler
closed
11 months ago
2
load_from_checkpoint loads hparams as dict instead of class HParams
#18
thelostscout
closed
11 months ago
6
Cannot assign new keys to hparams of type dict
#17
thelostscout
closed
11 months ago
0
on_train_start does not transfer batch to self.device
#16
LarsKue
closed
11 months ago
0
Extract PyTorch Lightning LR scheduler kwargs from scheduler kwargs
#15
fdraxler
closed
11 months ago
1
More Modules, Metrics, Bugfixes, and more
#14
LarsKue
closed
11 months ago
0
Copying HParams returns dict
#13
thelostscout
closed
11 months ago
1
How to validate without logger?
#12
fdraxler
closed
11 months ago
5
Use default engine for resolving class specs for loggers as well
#11
fdraxler
closed
11 months ago
3
Continued training `model.fit()` creates spurious validation entries in external log
#10
fdraxler
closed
11 months ago
5
Consistent class loading by name, support other loggers
#9
fdraxler
closed
1 year ago
0
Progressbar confusing when continuing training of a loaded model.
#8
HeiseMax
closed
1 year ago
1
pip doesn't install the dependency tomli
#7
Fjallripa
closed
1 year ago
1
Don't modify originally passed hparams
#6
fdraxler
closed
1 year ago
0
Start from checkpoint
#5
fdraxler
closed
1 year ago
0
Collection of launcher scripts
#4
fdraxler
closed
1 year ago
0
Better Optimizer and LR Scheduler kwargs specification + DenseTrainable
#3
RussellALA
closed
1 year ago
0
Automatically convert dicts to nested HParams
#2
fdraxler
closed
1 year ago
7
Suggestion: Use epoch progress bar
#1
fdraxler
closed
1 year ago
1