Installation

With pixi (recommended)

pixi installs cellmap-flow with everything it needs, including what PyPI cannot provide (the MinIO server and client that annotation painting needs), and the separate environments some model families run in. It needs pixi 0.68 or later, which reads this lockfile’s format: check with pixi --version, and update with pixi self-update, or brew upgrade pixi if Homebrew installed it. pixi install sets up the default environment (the dashboard, the catalog’s models and finetuning), and pixi run runs a command in it:

git clone https://github.com/janelia-cellmap/cellmap-flow.git
cd cellmap-flow
pixi install
pixi run cellmap_flow view -d /path/to/dataset.zarr

It installs on Linux (x86-64) and on Apple-silicon Macs. On a Mac:

Compiler: install Apple’s command-line tools first (xcode-select --install). The neuroglancer fork used for painting annotations is compiled during pixi install.
Model servers: a Mac has no cluster to submit jobs to, so the dashboard runs each model’s server on the Mac itself. Models serve and finetune on Apple’s GPU, much more slowly than on the cluster: Cellpose-SAM takes about 35 s for a chunk of 8 slices, so open one x–y panel rather than four, and for Cellpose consider slices_per_chunk: 1.
CUDA: it is Linux-only.

Model families whose dependencies clash with the default environment (fly_organelles checkpoints, BioImage Model Zoo models, Cellpose 4) run in a pixi environment of their own, chosen by the model’s type; nobody picks one by hand. Each is installed from the lockfile the first time a model needs it, before that model’s job is submitted (several minutes, once; the dashboard’s Job Logs shows pixi’s output meanwhile), so a new user, Fileglancer’s included, needs nothing beforehand. pixi run cellmap_flow envs install <name> installs one ahead of time, and cellmap_flow envs lists them and what is installed. See Running a model in its own environment.

runnables.yaml registers the same commands as Fileglancer apps. Inside a pixi environment, inference servers submitted to the cluster start with pixi run cellmap_flow serve (CELLMAP_FLOW_SERVER_COMMAND), so they run from the same lockfile. Command line lists the commands.

With pip

pip install cellmap-flow

The basic installation does not include DaCapo’s or BioImage.io’s dependencies; the extras add them, one or both:

pip install "cellmap-flow[dacapo]"
pip install "cellmap-flow[bioimageio]"
pip install "cellmap-flow[dacapo,bioimageio]"

To get an unreleased branch without cloning:

pip install "cellmap-flow @ git+https://github.com/janelia-cellmap/cellmap-flow.git"

Append @<branch> to the URL for a specific branch, and add extras in brackets after the name, e.g. "cellmap-flow[dacapo] @ git+https://github.com/janelia-cellmap/cellmap-flow.git@my-branch". For development, clone and install in editable mode instead (pip install -e . in the checkout).

Finetuning

Interactive finetuning needs two dependencies that cannot come from PyPI, a MinIO server (conda-forge only) and a Neuroglancer fork with voxel-annotation support, so pip install "cellmap-flow[finetune]" on its own is not enough; pixi installs both. See docs/finetuning.md in the repository for the full setup.

Any model whose network is plain PyTorch can be finetuned from the dashboard, with LoRA or in full: cellmap and Hugging Face exports, fly checkpoints, DaCapo runs, scripts, BioImage Model Zoo models with PyTorch weights, and Cellpose (on its flows, from sparsely painted instances). A compiled (TorchScript) zoo model is finetuned in full only, and ONNX or TensorFlow ones not at all; the Finetune tab offers only what the selected model allows.