Trustii

During the competition, the organizers gave access to a JupyterHub instance. Here are the steps to run the project on the platform. The commands are different because it was a Debian VM.

To run on the Trustii platform, follow these steps:

Open a terminal
Clone the repo
```
git clone ...
cd digilut
```

Create a Conda env to force the python version

conda create -n env-digitut python=3.12
conda init
source ~/.bashrc
conda activate env_digilut

Install poetry
```
pip install poetry -q
```

Install FAST deps:

Install deps (enter yes if asked):

sudo apt install libgl1 libopengl0 libopenslide0 libusb-1.0-0 libxcb-xinerama0

If errors encountered, run the following commands (enter yes if asked), then try again the previous command.

sudo apt update
sudo apt --fix-broken install
sudo apt install libpocl2

The last dependency of FAST is OpenCL. It depends on your CPU. Check your CPU specs:

cat /proc/cpuinfo

Then download the relevant driver. More doc here (for me it was Intel on Trustii).

# Intel driver opencl. Other option is to go manual: https://github.com/intel/compute-runtime/releases (not sure if sudo apt-get install intel-gmmlib is needed, I ran it)
sudo apt-get install intel-opencl-icd

Finally, install package deps
```
poetry install --with dev,docs
```

Test your installation:

source $(poetry env info --path)/bin/activate 
digilut --help

You should be prompted something like:

Usage: digilut [OPTIONS] COMMAND [ARGS]...                                                                                                        

Entrypoint of Digilut's main CLI.                                                                                                                 

╭─ Options ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ --install-completion          Install completion for the current shell.                                                                         │
│ --show-completion             Show completion for the current shell, to copy it or customize the installation.                                  │
│ --help                        Show this message and exit.                                                                                       │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ Commands ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ clean-bbox    Remove obvious labelling mistakes from the input bounding box csv and save a new cleaned dataset csv.                             │
│ credits       Print credits with style.                                                                                                         │
│ pyfast        Pyfast processing commands to patchify .tif WSI images.                                                                           │
│ tiles         Commands for tiles.                                                                                                               │
│ undersample   Undersample patches to solve class imbalance                                                                                      │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯