User Tools

Site Tools

pipeline:window:cryolo

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
Next revision Both sides next revision
pipeline:window:cryolo [2019/08/29 14:04]
twagner 		pipeline:window:cryolo [2019/08/29 16:28]
twagner [Training]
Line 184: Line 184:
 Navigate to the folder with ''config.json'' file, ''train_image'' folder, etc. Navigate to the folder with ''config.json'' file, ''train_image'' folder, etc.
  
-**1. Warm up your network**+**Train your network with 3 warmup epochs:**
  
 <code> <code>
 cryolo_train.py -c config.json -w 3 -g 0 cryolo_train.py -c config.json -w 3 -g 0
-</code> 
- 
-**2. Train your network** 
- 
-<code> 
-cryolo_train.py -c config.json -w 0 -g 0 
 </code> </code>
  
Line 199: Line 193:
  
 The training stops when the "loss" metric on the validation data does not improve 10 times in a row. This is typically enough. In case want to give the training more time to find the best model. You might increase the "not changed in a row" parameter to, for example, 15 by adding the flag //-e 15//: The training stops when the "loss" metric on the validation data does not improve 10 times in a row. This is typically enough. In case want to give the training more time to find the best model. You might increase the "not changed in a row" parameter to, for example, 15 by adding the flag //-e 15//:
 +
 <code> <code>
-cryolo_train.py -c config.json -w 0 -g 0 -e 15+cryolo_train.py -c config.json -w 3 -g 0 -e 15
 </code> </code>
 +
 to the training command. to the training command.
 ==== Picking ==== ==== Picking ====
Line 424: Line 420:
 In principle, there is not much difference in training crYOLO for filament picking and particle picking. For project with roughly 20 filaments per image we successfully trained on 40 images (=> 800 filaments). However, in our experience the warm-up phase and training need a little bit more time: In principle, there is not much difference in training crYOLO for filament picking and particle picking. For project with roughly 20 filaments per image we successfully trained on 40 images (=> 800 filaments). However, in our experience the warm-up phase and training need a little bit more time:
  
-**1. Warm up your network** +**Train your network with 10 warm up epochs:**
- +
-<code> +
-cryolo_train.py -c config.json -w 10 -g 0 +
-</code> +
- +
-**2. Train your network**+
  
 <code> <code>
-cryolo_train.py -c config.json -w 0 -g 0 -e 10+cryolo_train.py -c config.json -w 10 -g 0 -e 10
 </code> </code>
  
pipeline/window/cryolo.txt · Last modified: 2021/02/19 10:00 by twagner

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: GNU Free Documentation License 1.3
GNU Free Documentation License 1.3 Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki