This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
cryolo_release_note_110 [2018/08/23 07:56] twagner [Changes] |
cryolo_release_note_110 [2018/08/23 11:45] twagner |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== crYOLO 1.1.0 release notes ====== | ====== crYOLO 1.1.0 release notes ====== | ||
- | ===== Changes ===== | + | ===== Changes |
- | * crYOLO | + | * crYOLO now supports |
* New evaluation tool (more here) | * New evaluation tool (more here) | ||
- | * Support | + | * Supports |
- | * Extended data augmentation: | + | * Extended data augmentation: |
- | * Experimental support of periodic restarts during training (with --warm_restarts). | + | * Experimental support of periodic restarts during training (with --warm_restarts) |
+ | |||
+ | ==== Changes crYOLO Boxmanager ==== | ||
+ | * Support of visualization of EMAN2 helical coordinates (particle coordinates) | ||
+ | * New boxes could be loaded with a new color while keeping the old. | ||
+ | * Several bug fixes | ||
Line 11: | Line 17: | ||
==== Data preparation ==== | ==== Data preparation ==== | ||
- | As described [[pipeline: | + | As described [[pipeline: |
After this is done, you have to prepare training data for your model. | After this is done, you have to prepare training data for your model. | ||
Line 19: | Line 25: | ||
</ | </ | ||
- | After tracing your training data in e2helixboxer, | + | After tracing your training data in e2helixboxer, |
==== Configuration ==== | ==== Configuration ==== | ||
- | You can configure it the same way as for a " | + | You can configure it the same way as for a " |
==== Training ==== | ==== Training ==== | ||
- | In principle, there is not much difference in training crYOLO for filament picking and particle picking. However, | + | In principle, there is not much difference in training crYOLO for filament picking and particle picking. However, |
**1. Warm up your network** | **1. Warm up your network** | ||
Line 45: | Line 52: | ||
* //- -filament//: | * //- -filament//: | ||
- | * //-fw//: Filament width (pixel) | + | * //-fw//: Filament width (pixels) |
- | * //-bd//: Inter-Box distance (pixel). | + | * //-bd//: Inter-Box distance (pixels). |
Let's assume you want to pick a filament with a width of 100 pixels. The box size is 200x200 and you want a 90% overlap, than the picking command would be: | Let's assume you want to pick a filament with a width of 100 pixels. The box size is 200x200 and you want a 90% overlap, than the picking command would be: | ||
Line 54: | Line 61: | ||
The directory boxes will be created and all results are saved there. The format is the eman2 helix format with particle coordinates. | The directory boxes will be created and all results are saved there. The format is the eman2 helix format with particle coordinates. | ||
+ | |||
==== Visualize the results ==== | ==== Visualize the results ==== | ||
- | You can use the boxmanager as described [[pipeline: | + | You can use the boxmanager as described [[pipeline: |
- | ===== New evaluation tool ===== | + | ===== Evaluation your results |
The new evaluation tool allows you, based on your validation data, to get statistics about your training. | The new evaluation tool allows you, based on your validation data, to get statistics about your training. | ||
- | If you followed the tutorial, the validation data is selected randomly. With crYOLO 1.1.0 a run file for each training is created and saved into the folder runfiles/ in your project directory. This run file contains, which files were selected for validation and you can run your evaluation as follows: | + | If you followed the tutorial, the validation data are selected randomly. With crYOLO 1.1.0 a run file for each training is created and saved into the folder runfiles/ in your project directory. This run file contains which files were selected for validation, and you can run your evaluation as follows: |
< | < | ||
cryolo_evaluation.py -c config.json -w model.h5 -r runfiles/ | cryolo_evaluation.py -c config.json -w model.h5 -r runfiles/ | ||
Line 68: | Line 76: | ||
{{ : | {{ : | ||
- | The table contains several | + | The table contains several |
* AUC: Area under curve of the precision-recall curve. Overall summary statistics. Perfect classifier = 1, Worst classifier = 0 | * AUC: Area under curve of the precision-recall curve. Overall summary statistics. Perfect classifier = 1, Worst classifier = 0 | ||
- | * Topt: Optimal confidence threshold with respect to the F1 score. It might not be ideal for your picking, as the F1 score weights | + | * Topt: Optimal confidence threshold with respect to the F1 score. It might not be ideal for your picking, as the F1 score weighs |
* R (Topt): Recall using the optimal confidence threshold. | * R (Topt): Recall using the optimal confidence threshold. | ||
* R (0.3): Recall using a confidence threshold of 0.3. | * R (0.3): Recall using a confidence threshold of 0.3. | ||
* R (0.2): Recall using a confidence threshold of 0.2. | * R (0.2): Recall using a confidence threshold of 0.2. | ||
+ | * P (Topt): Precision using the optimal confidence threshold. | ||
* P (0.3): Precision using a confidence threshold of 0.3. | * P (0.3): Precision using a confidence threshold of 0.3. | ||
* P (0.2): Precision using a confidence threshold of 0.2. | * P (0.2): Precision using a confidence threshold of 0.2. | ||
Line 79: | Line 88: | ||
* F1 (0.3): Harmonic mean of precision and recall using a confidence threshold of 0.3. | * F1 (0.3): Harmonic mean of precision and recall using a confidence threshold of 0.3. | ||
* F1 (0.2): Harmonic mean of precision and recall using a confidence threshold of 0.2. | * F1 (0.2): Harmonic mean of precision and recall using a confidence threshold of 0.2. | ||
- | * IOU (Topt): Intersection over union of the auto-picked | + | * IOU (Topt): Intersection over union of the auto-picked |
- | * IOU (0.3): Intersection over union of the auto-picked | + | * IOU (0.3): Intersection over union of the auto-picked |
- | * IOU (0.2): Intersection over union of the auto-picked | + | * IOU (0.2): Intersection over union of the auto-picked |
+ | |||
+ | If the training data consists of multiple folders, then evaluation will be done for each folder separately. | ||
===== General model ===== | ===== General model ===== | ||
Increase the number of hand picked datasets to 25 by adding: | Increase the number of hand picked datasets to 25 by adding: |