cryolo_reference_example

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
cryolo_reference_example [2019/04/16 16:21]
twagner [Download reference data and getting started]
cryolo_reference_example [2019/09/23 09:28]
twagner [Reference setup]
Line 7: Line 7:
   * Titan V   * Titan V
   * Intel Core i9 7920X @ 2.90 Ghz   * Intel Core i9 7920X @ 2.90 Ghz
 +  * SSD Harddrive
 +  * crYOLO 1.5.0
  
  
Line 35: Line 37:
 ===== Training ===== ===== Training =====
  
-First, you have to warm up the network with "-g 0" I'm selection GPU 0. It takes 2 minutes 30 seconds on our reference setup: + 
-<​code>​ +The training is done with this command:
-cryolo_train.py -c config_phosnet.json -w 5 -g 0 +
-</​code>​ +
-The actual ​training is done with this command:+
 <​code>​ <​code>​
-cryolo_train.py -c config_phosnet.json -w -e 10 -g 0+cryolo_train.py -c config_phosnet.json -w -e -g 0
 </​code>​ </​code>​
-It needed 6 minutes ​45 seconds to converge (18 epochs). The best validation loss was 0.03316. These numbers might be a little bit different on your case.+It needes 5 minutes ​50 seconds to converge (5 warmup + 10 "​normal" ​epochs). The best validation loss was 0.03042. These numbers might be a little bit different on your case.
  
 ===== Prediction ===== ===== Prediction =====
Line 50: Line 49:
 </​code>​ </​code>​
  
-It picked 1617 particles on 12 micrographs in seconds. Including filtering the image and loading the model the command needed ​46 seconds.+It picked 1617 particles on 12 micrographs in seconds. Including filtering the image and loading the model the command needed ​38 seconds.
  
 ===== Visualize results ===== ===== Visualize results =====
  • cryolo_reference_example.txt
  • Last modified: 2019/09/23 09:28
  • by twagner