Commit 1e6510f2 authored by Nicolas Elie's avatar Nicolas Elie
Browse files

Mettre à jour README.md

parent dafedded

README.md

+34 −19
Changes for README.md: 34 added lines, 19 removed lines.
Original line number Diff line number Diff line
@@ -27,31 +27,39 @@ The proposed installation uses Conda environment: https://docs.conda.io/en/lates 


1. Require java on your system

Install the JRE on Debian (java)



> sudo apt install default-jre



```

sudo apt install default-jre

```



2. CONDA: Installation on Linux

https://conda.io/projects/conda/en/latest/user-guide/install/linux.html



> wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh



> chmod +x Miniconda3-latest-Linux-x86_64.sh

```

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh



> bash Miniconda3-latest-Linux-x86_64.sh -p $HOME/miniconda3

chmod +x Miniconda3-latest-Linux-x86_64.sh



bash Miniconda3-latest-Linux-x86_64.sh -p $HOME/miniconda3

```



$HOME you can indicate your directory



Close your terminal  



3. Download project : 

> git clone https://git.unicaen.fr/nicolas.elie/redpol-open.git (~ 2.5 Go)

4. cd redpol-open/

5. You can install an environment via the provided environment file(s): requirements.yml.



> conda env create -f requirements.yml



```

git clone https://git.unicaen.fr/nicolas.elie/redpol-open.git

```

_Size project : (~ 2.5 Go)_



4.  Access to folder project

```

cd redpol-open/

```

5. You can install an conda environment via the provided environment file(s): requirements.yml.

```

conda env create -f requirements.yml

```



# Procedure
@@ -66,8 +74,9 @@ In this project, a example file is given : 'Examples/Process_13.vsi' as well as 
To test program :



Run _AnalyzeSequence.ipnyb_ program on conda environment

> jupyter nbconvert --execute --clear-output _AnalyzeSequence.ipynb



```

jupyter nbconvert --execute --clear-output _AnalyzeSequence.ipynb

```

#### To build a new model



1. Annotation of the images with Qupath : https://qupath.github.io/
@@ -83,18 +92,24 @@ On your images, annotate the objects that interest you. If you have categories, 
2. Once your annotations are done, use the script "_Export Annotations gson.groovy_" to export the annotations as a json file in Qupath.

3. Then use the python program "_JSON_Qupath_to_ImageLabel.ipynb_" which will allow to format the images and annotations so that they can be used by the program "_Model_Build.ipynb_." This program is directly from an example provided on the github Stardist repository.



> jupyter nbconvert --execute --clear-output JSON_Qupath_to_ImageLabel.ipynb

```

jupyter nbconvert --execute --clear-output JSON_Qupath_to_ImageLabel.ipynb

```



4. Configure and execute _Model_Build.ipynb_



> jupyter nbconvert --execute --clear-output Model_Build.ipynb

```

jupyter nbconvert --execute --clear-output Model_Build.ipynb

```



It is possible to test code with a small series with images examples : _Examples/0001.tif_ and _Examples/0001b.tif_ and JSON files on _Examples/qupathProject/ExportJSON_



> jupyter nbconvert --execute --clear-output Model_Build_short_example.ipynb

```

jupyter nbconvert --execute --clear-output Model_Build_short_example.ipynb

```



It is a simple test, the given model is not efficiency



> **_Be careful, for the program to work, you need a compatible GPU graphics card._**

> **_Be careful, for that Model_Build works, you need a compatible GPU graphics card._**



5. The program "_AnalyzeSequence.ipnyb_" can then be used by designating your own model in the program code.