Yin Li
|
85efb9e3a3
|
Add autograd detect anomaly flag, and test cudnn.deterministic
cudnn.deterministic is only 10% slower
|
2020-09-12 18:26:44 -04:00 |
|
Yin Li
|
d8c6be797d
|
Move figures from .data to .utils
|
2020-09-12 18:25:58 -04:00 |
|
Yin Li
|
154376d95a
|
Add tgt_pad, rename pad to in_pad
tgt_pad can be useful for scale_factor > 1
|
2020-09-12 18:24:36 -04:00 |
|
Yin Li
|
f3216f4cbd
|
Add multiple pkgs to import_attr
|
2020-09-09 14:55:36 -04:00 |
|
Yin Li
|
670364e54c
|
Change Lag2Eul to lag2eul as a function
|
2020-08-22 23:25:08 -04:00 |
|
Yin Li
|
3eb1b0bccc
|
Add power spectrum tracking
|
2020-08-22 23:24:25 -04:00 |
|
Yin Li
|
5d22594ede
|
Add shape report after narrow_cast in training
|
2020-08-22 12:28:08 -07:00 |
|
Yin Li
|
9c8b331bf5
|
Add loglxe loss as log of product of Lagrangian and Eulerian losses
|
2020-08-09 12:59:24 -07:00 |
|
Yin Li
|
265587922d
|
Add srsgan model and scale_factor to model arguments
|
2020-08-04 21:13:10 -07:00 |
|
Yin Li
|
b54fc4ba3a
|
Add a line about socket hopefully will avoid port conflicts
|
2020-07-28 07:38:07 -07:00 |
|
Yin Li
|
a886e53c54
|
Always use training loss for lr scheduler
|
2020-07-27 17:47:58 -07:00 |
|
Yin Li
|
2336d83f2d
|
Add optimizer and scheduler kwargs via json.loads
|
2020-07-27 12:40:19 -07:00 |
|
Yin Li
|
8ce13e67f6
|
Fix training hang due to constrained layout of matplotlib
|
2020-07-27 12:39:41 -07:00 |
|
Yin Li
|
ca39b2028f
|
Replace cuda.device with env var to assign GPU to processes
|
2020-07-22 18:48:12 -07:00 |
|
Yin Li
|
98cdd4795c
|
Add DistFieldSampler to benefit from page cache
|
2020-07-21 23:37:59 -07:00 |
|
Yin Li
|
ec46f41ba5
|
Add lag x eul loss tracking to tensorboard
|
2020-07-21 21:10:34 -07:00 |
|
Yin Li
|
56f8fa932b
|
Revert "Deprecate GroupedRandomSampler for use of memmap in place of cache"
This reverts commit bbf77c9f91.
|
2020-07-16 15:22:52 -07:00 |
|
Yin Li
|
4160ed25f7
|
Add lagrangian and eulerian alternate training
|
2020-07-15 15:57:10 -04:00 |
|
Yin Li
|
e8744d6c7b
|
Remove adversary
|
2020-07-14 21:07:05 -04:00 |
|
Yin Li
|
6fa85f8285
|
Merge branch 'master' into lag2eul
|
2020-07-14 18:47:54 -04:00 |
|
Yin Li
|
bbf77c9f91
|
Deprecate GroupedRandomSampler for use of memmap in place of cache
|
2020-07-14 18:47:26 -04:00 |
|
Yin Li
|
3437b20ed8
|
Merge branch 'master' into lag2eul
|
2020-07-14 18:28:10 -04:00 |
|
Yin Li
|
818ed6923d
|
Add memmap to numpy data loading
|
2020-07-14 17:42:10 -04:00 |
|
Yin Li
|
d0ab596e11
|
Remove min-reduction; Split grad plots; Disable cgan w/o adv
|
2020-07-13 09:40:15 -04:00 |
|
Yin Li
|
29ab550032
|
Change instance noise
Simpler linear annealing as in the blog post;
different noise to output and target
|
2020-07-11 01:56:46 -04:00 |
|
Yin Li
|
873258d8a7
|
Add linear resamplers
|
2020-07-11 01:37:23 -04:00 |
|
Yin Li
|
c956dca795
|
Add narrow_cast and narrow_by
|
2020-07-11 01:24:20 -04:00 |
|
Yin Li
|
28ec245a7a
|
Add pixel-shifting data augmentation
|
2020-07-11 00:46:19 -04:00 |
|
Yin Li
|
8a95d69818
|
Add cropping anchors controlled by start, stop, step
|
2020-07-11 00:39:19 -04:00 |
|
Yin Li
|
819e77cd86
|
Add slice plotting for field >2d
|
2020-07-10 14:58:08 -04:00 |
|
Yin Li
|
bab3f08aa5
|
Add Lag2Eul to training
|
2020-07-10 12:50:07 -04:00 |
|
Yin Li
|
c3d7456d0c
|
Change .pth to .pt following torch convention
|
2020-06-20 18:31:09 -04:00 |
|
Yin Li
|
90a0d6e0f8
|
Add callback loading mechanism
|
2020-06-14 17:59:31 -04:00 |
|
Yin Li
|
5bb2a19933
|
Add grouped cache for data bigger than CPU RAM
|
2020-05-28 23:01:56 -04:00 |
|
Yin Li
|
2e687da905
|
Add LRU cache to replace existing cache
|
2020-05-16 23:26:53 -04:00 |
|
Yin Li
|
4cc2fd51eb
|
Refactor runtime default argument setting
|
2020-05-16 17:28:42 -04:00 |
|
Yin Li
|
897c3563db
|
Add additive and multiplicative augmentation
|
2020-05-07 15:57:05 -04:00 |
|
Yin Li
|
67e5ed9eb6
|
Remove smooth transition to adversarial loss
|
2020-05-07 10:48:55 -04:00 |
|
Yin Li
|
f442dd59ba
|
Remove noise channels
|
2020-05-06 15:33:21 -04:00 |
|
Yin Li
|
a88f27a3a1
|
Add better looking tensorboard images
|
2020-04-22 00:56:29 -04:00 |
|
Yin Li
|
996c0d3aed
|
Add auto loading of checkpoint
|
2020-04-21 18:34:18 -04:00 |
|
Yin Li
|
c9f468c568
|
Add explicit barrier after init dist groups
|
2020-04-21 13:38:02 -04:00 |
|
Yin Li
|
5adf6cbac6
|
Add exit when GPU not found
|
2020-04-20 19:27:57 -04:00 |
|
Yin Li
|
01a60cc0c7
|
Add checkpoint symlink to state file
|
2020-04-16 17:50:08 -04:00 |
|
Yin Li
|
d01d0cee83
|
Add noise annealing duration
|
2020-03-09 12:07:55 -04:00 |
|
Yin Li
|
93d973b5c8
|
Change state saving to every epoch
|
2020-03-07 18:03:19 -05:00 |
|
Yin Li
|
7a0f0bd979
|
Add new pytorch features with fallback to old
|
2020-03-07 14:49:18 -05:00 |
|
Yin Li
|
5fbf447e2d
|
Fix spectral norm interference with gradient visualization
spectral norm introduces weight_orig, so that not all weight parameters
ends with ".weight"
|
2020-03-04 16:21:00 -05:00 |
|
Yin Li
|
33fd13e4ac
|
Fix bug from 1963530 to show last instead of 2nd layer gradients
|
2020-03-03 20:53:34 -05:00 |
|
Yin Li
|
ef993d83e9
|
Add instance noise
|
2020-03-03 20:22:31 -05:00 |
|