User Tools

Site Tools


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
Next revision Both sides next revision
auto_2d_class_selection [2019/06/06 12:26]
twagner [Cinderella: Automatic 2D class selection]
auto_2d_class_selection [2019/07/29 14:46]
twagner [Cinderella]
Line 10: Line 10:
  
 Our automatic 2d class selection tool (**Cinderella**) is based on a deep learning network to seperate 2D classes from .hdf / .mrcs files into good and bad classes. It uses the same deep neural network as crYOLO and was pretrained on a set good / bad classes. Cinderella was written to automate cryo-em data processing. It's open source and easy to use ([[auto2d_tutorial|see tutorial]]). You can easily train it with your own set of classes. Our automatic 2d class selection tool (**Cinderella**) is based on a deep learning network to seperate 2D classes from .hdf / .mrcs files into good and bad classes. It uses the same deep neural network as crYOLO and was pretrained on a set good / bad classes. Cinderella was written to automate cryo-em data processing. It's open source and easy to use ([[auto2d_tutorial|see tutorial]]). You can easily train it with your own set of classes.
 +<note>
   * **License**: MIT   * **License**: MIT
   * **GitHub repository**: https://github.com/MPI-Dortmund/sphire_classes_autoselect   * **GitHub repository**: https://github.com/MPI-Dortmund/sphire_classes_autoselect
 +</note>
 Here are a couple of examples for good / bad classes in Cinderella:  Here are a couple of examples for good / bad classes in Cinderella: 
  
Line 19: Line 19:
  
 ====== The Model ====== ====== The Model ======
-Our model was trained on a set of 2D classes from ISAC. The training dataset does **not contain any Relion classes**, so it might be that Cinderella will not work with them. However, you can easily [[auto_2d_class_selection#contribute|contribute]]!  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.+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!  
  
-Right now our model is trained on **17 datasets**. But we will increase the number often!+Right now our model is trained on **19 datasets**. But we will increase the number often!
 ====== Download ====== ====== Download ======
 ====Cinderella==== ====Cinderella====
-Version: 0.2.0+Version: 0.3.1
  
-Uploaded: 28May 2019+Uploaded: 29July 2019
  
-[[ftp://ftp.gwdg.de/pub/misc/sphire/cinderella_v0.2.0/cinderella-0.2.0.tar.gz|DOWNLOAD]]+[[https://pypi.org/project/cinderella/#files|DOWNLOAD]]
  
 ====Pretrained model==== ====Pretrained model====
-Uploaded: 28May 2019+Uploaded: 11July 2019, Datasets: 19
  
-[[ftp://ftp.gwdg.de/pub/misc/sphire/auto2d_models/model_cinderella_20190528.h5|DOWNLOAD]]+[[ftp://ftp.gwdg.de/pub/misc/sphire/auto2d_models/model_cinderella_20190708.h5|DOWNLOAD]]
  
 [[auto2d_tutorial#classify|Valid configuration file]] [[auto2d_tutorial#classify|Valid configuration file]]
  
 +====Archive====
 +Old version of cinderella and the pretrained model can be found in the [[cinderella_archive|archive]]
 +
 +====Changelog====
 +=== Version 0.3.0 ===
 +  * More data augmentation (add rotation)
 +  * Better sampling of validation data. It is now ensured that each file contributes some validation data.
 +  * Updated tensorflow to 1.12.3  to make it compatible to the crYOLO environment
 ====== Contribute ====== ====== Contribute ======
 Here is the repository of our training data: Here is the repository of our training data:
Line 65: Line 77:
 </code> </code>
  
-Install Cinderella:+Install fast numpy:
 <code> <code>
-conda install numpy==1.14.5 +conda install numpy==1.15.4 
-pip install cinderella-X.Y.Z.tar.gz +</code>
  
 +Install Cinderella for **GPU**:
 +<code>
 +pip install cinderella[gpu]
 +</code>
 +
 +**... or CPU**:
 +<code>
 +pip install cinderella[cpu]
 </code> </code>
  
 ====== Tutorial ====== ====== Tutorial ======
 [[auto2d_tutorial|We created a tutorial how to use Cinderella!]] [[auto2d_tutorial|We created a tutorial how to use Cinderella!]]
auto_2d_class_selection.txt · Last modified: 2020/08/27 15:11 by twagner