Get started

This guide will help you install Celldetective and launch your first session.

Step 1: Installation

Celldetective is a Python package typically installed via pip in a conda environment.

$ pip install celldetective[all]

For detailed instructions, troubleshooting, and GPU setup, see the Installation Reference.

Step 2: Launching the GUI

Once installed, open a terminal and run:

$ python -m celldetective

static_launch_gui — Launching the software from a terminal.

Step 3: Next Steps

Upon launch, you can create a new experiment or load an existing one.

To understand the experiment structure, see First Experiment.
To create your first project, see How to create an experiment.

Getting help inside the software

Several panels carry a button, next to the steps where a choice has to be made: how to structure an experiment, preprocess, segment, track, propagate a classification or compute a neighborhood. It opens a helper that asks a few yes/no questions about your data and ends on a suggestion.

the help panels — **Helpers.** The *SEGMENT* step offers two helpers, one to choose between a threshold pipeline and deep learning, one to choose a deep learning strategy. The *TRACK* step asks whether your cells can be tracked at all.

The help button of the SEGMENT step (1) opens the list of its helpers (2), while the one of the TRACK step (3) opens straight on its first question. A helper asks one yes/no question at a time (4) and keeps the answers given so far in view (5). Once it ends on a suggestion, Back changes the last answer (6) and Read the tutorial opens the matching page of this documentation (7).