tissue-tag 0.1.1

Tissue Tag: jupyter image annotator

Tissue Tag jupyter image annotator

General instructions until replaced with something proper 

1. You basically only need the tissue_tag.py file, which you will use to import the relevant functions like  import tissue_tag as tt  You also need the tutorial notebook https://github.com/nadavyayon/TissueTag/blob/main/Tutorials/image_annotation_tutorial.ipynb
2. You need to setup an env that has bokeh (https://github.com/bokeh/bokeh) and jupyter-bokeh (https://pypi.org/project/jupyter-bokeh/) installed. This is the trickiest bit sometimes 
3. When running on farm, you need to set the port (e.g., 5011) you are using to get it plot properly in the correct place  Python package to interactively annotate histological images within a Jupyter notebook

installation (partial - only tissue-tag)

installation (complete - imagespot)

Instructions for imagespot env

1. Create a new conda environment: conda create -n imagespot python=3.9 say 'y' when asked

2. activate new environment conda activate imagespot

3. install scanpy conda install -c conda-forge scanpy python-igraph leidenalg

4. install jupyter-lab conda install -c conda-forge jupyterlab

5. install open-cv pip install opencv-python

6_5) optional - install cellpose pip install cellpose pip uninstall torch conda install pytorch cudatoolkit=11.3 -c pytorch

7. install bokeh pip install jupyter_bokeh pip install bokeh

8. install scikit-image pip install scikit-image

9. add the new enviroment to jupyter lab path
   $ ipython kernel install --name imagespot --user