Alex Stoken
|
bc4ef4861b
|
Default optimizer SGD
|
5 years ago |
|
Alex Stoken
|
7abf202cad
|
Mode all optimizer settings to 'hyp.yaml', integrate proper momentum with Adam optimizer
|
5 years ago |
|
Alex Stoken
|
7edbf6570e
|
Fix help message for cfg files
|
5 years ago |
|
Alex Stoken
|
d64ad0fbf3
|
Remove --resume functionality and related checks/logic.
|
5 years ago |
|
Alex Stoken
|
e572bb0803
|
Add plot_results save location to log_dir
|
5 years ago |
|
Alex Stoken
|
c8152c81a6
|
Syntax fixes
|
5 years ago |
|
Alex Stoken
|
9b7386f603
|
Add save_dir arg to test.test, use arg as location for saving batch jpgs
|
5 years ago |
|
Alex Stoken
|
945307beba
|
Add save_dir to plot_lr_scheduler and plot_labels
Set save_dir = log_dir in train.py
|
5 years ago |
|
Alex Stoken
|
3b2b330872
|
Move results.txt from weights/ to log_dir
|
5 years ago |
|
Alex Stoken
|
ade023cff2
|
Fix hyp file read in and dict update.
Add example of hyp yaml
|
5 years ago |
|
Alex Stoken
|
5f2eeba233
|
remove old print statements
|
5 years ago |
|
Alex Stoken
|
333f678b37
|
add update default hyp dict with provided yaml
|
5 years ago |
|
Alex Stoken
|
a448c3bcd7
|
add logic for resuming and getting hyp for resume run
|
5 years ago |
|
Alex Stoken
|
25e51bcec7
|
add util function to get most recent last.pt file
added logic in train.py __main__ to handle resuming from a run
|
5 years ago |
|
Alex Stoken
|
490f1e7b9c
|
add save_dir arg to plot_lr_scheduler, default to current dir.
Uncomment plot_lr_scheduler in train() and pass log_dir as save location
|
5 years ago |
|
Alex Stoken
|
4418809cf5
|
change weights dir (wdir) to be unique to each run, under log_dir
|
5 years ago |
|
Alex Stoken
|
d9f446cd81
|
add save yaml of opt and hyp to tensorboard log_dir in train()
|
5 years ago |
|
Alex Stoken
|
a85e6d0fc0
|
add parser arg for hyp yaml file
|
5 years ago |
Glenn Jocher
|
8b26e89006
|
AutoAnchor bug fix #72
|
5 years ago |
|
Glenn Jocher
|
bdd9fee841
|
update fast mode
|
5 years ago |
|
Glenn Jocher
|
1c0b6236e3
|
update fast mode
|
5 years ago |
|
Glenn Jocher
|
915b1481fc
|
default check_git_status() to True
|
5 years ago |
|
Glenn Jocher
|
14523bb030
|
FP16 to FP32 ckpt load
|
5 years ago |
|
Glenn Jocher
|
c5966abba8
|
glob search bug fix #77
|
5 years ago |
|
Glenn Jocher
|
31f3310029
|
assert best possible recall > 0.9 before training
|
5 years ago |
Lauritzen Kasper Primdal
|
c3d4d321d3
|
Ensures weights/ dir exists
Allows train.py to be run outside of yolov5/ directory.
|
5 years ago |
|
Glenn Jocher
|
099e6f5ebd
|
--img-size stride-multiple verification
|
5 years ago |
|
Glenn Jocher
|
22d6088205
|
speed-reproducibility fix #17
|
5 years ago |
|
Glenn Jocher
|
55ca5c74d2
|
multi-scale fix #16
|
5 years ago |
|
Glenn Jocher
|
7c2832cd49
|
assert equal model and dataset classes
|
5 years ago |
|
Glenn Jocher
|
11121e39ed
|
updates
|
5 years ago |
|
Glenn Jocher
|
eb97b2e413
|
NMS fast mode
|
5 years ago |
|
Glenn Jocher
|
ce36905358
|
updates
|
5 years ago |
|
Glenn Jocher
|
1e84a23f38
|
initial commit
|
5 years ago |
