Glenn Jocher
|
1c13e67b33
|
evolution bug fix #346
|
4 years ago |
Laughing
|
51a161d9e1
|
Update train.py
|
4 years ago |
Glenn Jocher
|
e16e9e43e1
|
new nc=len(names) check
|
4 years ago |
Glenn Jocher
|
cb527d3af9
|
new nc=len(names) check
|
4 years ago |
Glenn Jocher
|
603ea0bfdc
|
update log_dir to runs/exp #107
|
4 years ago |
Glenn Jocher
|
72d5b58b9a
|
disable LR plot to suppress warning message
|
4 years ago |
Glenn Jocher
|
24c5a941f0
|
--resume EMA fix #292
|
4 years ago |
Alex Stoken
|
9d631408a2
|
Move hyp and opt yaml save to top of train()
Fixes bug where scaled values were saved in hyp.yaml, which would cause continuity issues with --resume
|
4 years ago |
Glenn Jocher
|
bf6f41567a
|
hyperparameter printout update
|
4 years ago |
Glenn Jocher
|
6b134d93c5
|
Update train.py
|
4 years ago |
Glenn Jocher
|
16f6834486
|
update train.py and experimental.py
|
4 years ago |
Alex Stoken
|
52bac22f09
|
Add in --resume functionality with option to specify path or to get most recent run
|
4 years ago |
Laughing
|
956511dafd
|
fix LR bug
|
4 years ago |
Glenn Jocher
|
bb3c346916
|
model.yaml nc inherited from dataset.yaml
|
4 years ago |
Glenn Jocher
|
df224a0d8f
|
EMA bug fix #279
|
4 years ago |
Glenn Jocher
|
3bdea3f697
|
strip_optimizer() bug fix #253
|
4 years ago |
Glenn Jocher
|
597ed4ce63
|
Update train.py
|
4 years ago |
Glenn Jocher
|
9705c15da1
|
Update train.py
|
4 years ago |
ChristopherSTAN
|
a8d17ce712
|
Update train.py
|
4 years ago |
Glenn Jocher
|
13f69777a6
|
typo fix
|
4 years ago |
yxNONG
|
1aa2b67933
|
Update train.py
|
4 years ago |
Glenn Jocher
|
86784cfdbf
|
--resume bug fix #252
|
4 years ago |
Glenn Jocher
|
ad4c22cbfe
|
--resume bug fix #187
|
4 years ago |
Glenn Jocher
|
3b16c865f0
|
assert --epochs
|
4 years ago |
yxNONG
|
cdb9bde181
|
Unify the check point of single and multi GPU
save the model.hyp etc to checkpoint when use multi GPU training
|
4 years ago |
Glenn Jocher
|
b203c9b7ff
|
update train.py incompatible model message fix #222
|
4 years ago |
Glenn Jocher
|
37e13f8846
|
update mosaic border
|
4 years ago |
Glenn Jocher
|
22fb2b0c25
|
refactor dataloader
|
4 years ago |
Glenn Jocher
|
256a3e89d2
|
small dataset bug fix #140
|
4 years ago |
Glenn Jocher
|
b50fdf16af
|
model.names multi-GPU bug fix #94
|
4 years ago |
Alex Stoken
|
de191655e4
|
Fix yaml saving (don't sort keys), reorder --opt keys, bug fix hyp dict accessor
|
4 years ago |
Alex Stoken
|
2d396bea00
|
Fix bug in --help from percent sign in help string
|
4 years ago |
Glenn Jocher
|
b8557f87e3
|
add stride to datasets.py
|
4 years ago |
Alex Stoken
|
611aacf1bf
|
Turn opt into dictionary before sending it to yaml
|
4 years ago |
Alex Stoken
|
bc4ef4861b
|
Default optimizer SGD
|
4 years ago |
Alex Stoken
|
7abf202cad
|
Mode all optimizer settings to 'hyp.yaml', integrate proper momentum with Adam optimizer
|
4 years ago |
Alex Stoken
|
7edbf6570e
|
Fix help message for cfg files
|
4 years ago |
Alex Stoken
|
d64ad0fbf3
|
Remove --resume functionality and related checks/logic.
|
4 years ago |
Glenn Jocher
|
6c1b87a42e
|
update google_utils import
|
4 years ago |
Glenn Jocher
|
1f1917ef56
|
remove fast, add merge
|
4 years ago |
Alex Stoken
|
e572bb0803
|
Add plot_results save location to log_dir
|
4 years ago |
Lornatang
|
899f1d4bde
|
Fix DDP bug in single process multiple device use cases
|
4 years ago |
Glenn Jocher
|
cce95e744d
|
backbone as FP16, save default to FP32
|
4 years ago |
Glenn Jocher
|
d9b64c27c2
|
save ckpt in FP16 #119
|
4 years ago |
Glenn Jocher
|
9fdb0fbacf
|
AutoAnchor bug fix # 117
|
4 years ago |
Alex Stoken
|
c8152c81a6
|
Syntax fixes
|
4 years ago |
Alex Stoken
|
9b7386f603
|
Add save_dir arg to test.test, use arg as location for saving batch jpgs
|
4 years ago |
Alex Stoken
|
945307beba
|
Add save_dir to plot_lr_scheduler and plot_labels
Set save_dir = log_dir in train.py
|
4 years ago |
Alex Stoken
|
3b2b330872
|
Move results.txt from weights/ to log_dir
|
4 years ago |
Alex Stoken
|
ade023cff2
|
Fix hyp file read in and dict update.
Add example of hyp yaml
|
4 years ago |