auto_2d_class_selection

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
auto_2d_class_selection [2019/12/16 10:45]
twagner [Cinderella: Deep learning based binary classification tool]
auto_2d_class_selection [2020/08/27 15:11]
twagner [Changelog]
Line 8: Line 8:
 ---- ----
  
-Our binary classification ​ tool (**Cinderella**) is based on a deep learning network to classify class averages, micrographs or subtomograms into good and bad categories. +Our binary classification ​ tool (//Cinderella//) is based on a deep learning network to classify class averages, micrographs or subtomograms into good and bad categories. 
-For class averages, it supports .hdf/.mrcsfor micrographs ​.mrc format ​and for subtomograms ​it expect that they are saved in a .hdf file+Cinderella ​supports ​''​.hdf/.mrcs''​ ** files for class averages**, ''​.mrc''​ **files for micrographs**, ​and ''​.hdf''​ **files ​for subtomograms**
-Cinderella was written to automate cryo-em data processing. ​ It's open source and easy to use. +//Cinderella// was written to automate cryo-em data processing. ​ It's open source and easy to use. 
-We provide a pretrained general model for classifying class averages.([[auto2d_tutorial|see tutorial]]). But you can easily train it with your own set of classes/micrographs/​subtomograms.+We provide a pretrained general model for classifying class averages ([[auto2d_tutorial|see tutorial]]). But you can easily train it with your own set of classesmicrographs, and/or subtomograms.
  
 <​note>​ <​note>​
-  * **License**:​ MIT +  * **License**: ​[[https://​github.com/​MPI-Dortmund/​sphire_classes_autoselect/​blob/​master/​LICENSE|MIT]] 
-  * **GitHub repository**: https://​github.com/​MPI-Dortmund/​sphire_classes_autoselect+  * **Repository**: [[https://​github.com/​MPI-Dortmund/​sphire_classes_autoselect|GitHub]]
 </​note>​ </​note>​
  
-Here are a couple of examples for good / bad classes in Cinderella: ​+ 
 +====== 2D class selection model ====== 
 +Our model was trained on a set of 2D classes from both [[https://​sphire.mpg.de/​wiki/​doku.php?​id=pipeline:​isac:​sxisac2|ISAC]] and Relion. During the creation of the training data set, we tried to answer the question, "Which class would I select If I would not know the particle?"​ when deciding what is a "​good"​ class. ​Here are a couple of examples for good/bad classes in //Cinderella//
  
 {{ ::​cinderellea.png?​450 |}} {{ ::​cinderellea.png?​450 |}}
  
-====== 2D class selection model ====== +
-Our model was trained on a set of 2D classes from ISAC. During the creation of the training dataset, I tried to ask myself "Which class would I select If I would not know the particle?"​ to decide which is a good class. +
-<note important>​ +
-The training dataset does **not contain any Relion classes**, so it might be that Cinderella will not work with well them. +
-</​note>​+
 You can easily [[auto_2d_class_selection#​contribute|contribute]] your own classes!  ​ You can easily [[auto_2d_class_selection#​contribute|contribute]] your own classes!  ​
  
-Right now our model is trained on **19 datasets**. But we will increase the number often!+Right now our model is trained on **4773 good classes and 5390 bad classes**.
 ====== Download ====== ====== Download ======
 ====Cinderella==== ====Cinderella====
-Version: 0.5.0+Version: 0.7.0
  
-Uploaded: ​16December 2019+Uploaded: ​27August 2020
  
 [[https://​pypi.org/​project/​cinderella/#​files|DOWNLOAD]] [[https://​pypi.org/​project/​cinderella/#​files|DOWNLOAD]]
  
 ====Pretrained model (2D classes)==== ====Pretrained model (2D classes)====
-Uploaded: ​10December 2019Datasets20 +Uploaded: ​27August 2020Dataset4773 good classes and 5390 bad classes.
- +
-[[ftp://​ftp.gwdg.de/​pub/​misc/​sphire/​auto2d_models/​gmodel_cinderella_201912_N20.h5|DOWNLOAD]] +
- +
-[[auto2d_tutorial#​classify|Valid configuration file]]+
  
 +[[ftp://​ftp.gwdg.de/​pub/​misc/​sphire/​auto2d_models/​gmodel_cinderella07_202008_N10163.h5|DOWNLOAD]]
 ====Archive==== ====Archive====
-Old versions of cinderella and the pretrained model can be found in the [[cinderella_archive|archive]]+Old versions of cinderella and the pretrained model can be found in the [[cinderella_archive|archive]].
  
 ====Changelog==== ====Changelog====
 +
 +=== Version 0.7 ===
 +  * Now uses a **circular masks by default**. This allows to use full rotation during data augmentation. Can be deactivated by setting the field ''​mask_radius''​ in the configuration file to -1. In case you want to use an model trained with Cinderella < 0.7 please set the radius to -1. Otherwise you case specify any radius you want. **By default** (no ''​mask_radius''​ provided) it will use 0.4*''​input_size''​.
 +  * The general models now includes **300 new good Relion classes and 2000 new bad Relion classes** (//Thanks to Takanori Nakane and Grigory Sharov//).
 +  * Fixed numerical instability that occurs when you have classes filled with a constant value (//Thanks to Grigory Sharov//).
 +  * Fixed a problem with classes that contain NaN values. NaN values are now replaced with 0. (//Thanks to Grigory Sharov//).
 +  * Fixed an issue when filenames contain more than one point.
 +
 +=== Version 0.6 ===
 +  * Fix an issue for classes in mrcs format
 +  * Minor changes
  
 === Version 0.5 === === Version 0.5 ===
Line 106: Line 112:
  
 ====== Tutorial ====== ====== Tutorial ======
-We created ​two tutorials:+We created ​three tutorials:
  
   * [[auto2d_tutorial|How to use Cinderella for 2D class selection]]   * [[auto2d_tutorial|How to use Cinderella for 2D class selection]]
  • auto_2d_class_selection.txt
  • Last modified: 2020/08/27 15:11
  • by twagner