• Search by Module
• Search by Words
• Search Projects

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

Project: PSSR (GitHub Link)

• PSSR-master
  • tile_from_info.py
  • utils
    • tile.py
    • synth.py
    • __pycache__
      • czi.cpython-37.pyc
      • unet.cpython-37.pyc
      • crappifiers.cpython-37.pyc
      • tile.cpython-37.pyc
      • synth.cpython-37.pyc
      • rrdb.cpython-37.pyc
      • metrics.cpython-37.pyc
      • __init__.cpython-37.pyc
      • multi.cpython-37.pyc
      • utils.cpython-37.pyc
      • resnet.cpython-37.pyc
      • models.cpython-37.pyc
    • metrics.py
    • rrdb.py
    • .gitkeep
    • resnet.py
    • models.py
    • crappifiers.py
    • multi.py
    • unet.py
    • __init__.py
    • utils.py
    • czi.py
  • metric_gen.py
  • LICENSE
  • models
    • .gitkeep
  • Inference_PSSR_for_EM.ipynb
  • image_gen.py
  • colab_notebooks
    • PSSR 03 - Train your data.ipynb
    • PSSR 02 - Generate training datasets.ipynb
    • PSSR 01 - Get to know your data.ipynb
    • PSSR 04 - Inference.ipynb
  • train.py
  • gen_sample_info.py
  • README.md
  • stats
    • LR-PSSR
      • .gitkeep
    • models
      • .gitkeep
    • HR
      • .gitkeep
    • LR
      • .gitkeep
  • env.yml

Point-Scanning Super-Resolution (PSSR)

Good stuff

BioRxiv Preprint: Deep Learning-Based Point-Scanning Super-Resolution Imaging

All data is hosted in 3DEM Dataverse: https://doi.org/10.18738/T8/YLCK5A, which includes:

• Training, testing, and validation data
• Pretrained PSSR models for each image type described in the PSSR manuscript
• Environment set-up env.yml file that lists all prerequisite libraries.

There is also a beautifully written PSSR Tweetorial. Enjoy!

Instruction of use

Run PSSR from Google Colaboratory (Colab)

Google Colaboratory (Colab) version of PSSR is now ready. (PSSR - Colab for programmers)

Another PSSR Colab version that orients to non-programmers is also going to be released soon. (PSSR - Colab for non-programmers (I'M WORKING ON IT))

Very few libraries need to be installed manually for PSSR Colab verion - most dependencies are preinstalled in the Colab environment. This makes the environment set-up step painless and you will be able to quickly get straight to the real fun. However, it also means some of the libraries you will be using are more recent than the ones used for the manuscript, which can be accessed from the instruction below.

Run PSSR from the command line

Environment set-up

• Install Anaconda (Learn more)
• Create a new conda environment for PSSR: conda env create --file=env.yaml (Learn more)

EM inference

Please refer to the handy Inference_PSSR_for_EM.ipynb. You need to modify the path for the test images accordingly. Note the input pixel size needs to be 8 nm.

TO BE CONTINUED...