Skip to content

cvnpy

Getting Started

Charestlab/cvnpy

Getting Started¶

Requirements¶

Python >=3.10
FreeSurfer-compatible surface files (surf/)
macOS/Linux supported (tested on macOS ARM)

Installation¶

cd /Users/iancharest/Documents/codex/cvncode/cvnpy
python3 -m venv .venv
.venv/bin/python -m pip install --upgrade pip setuptools wheel
.venv/bin/python -m pip install -e '.[dev]'

Data Discovery Modes¶

You can provide surface data in one of two ways.

1) `SUBJECTS_DIR` + `--subject`¶

export SUBJECTS_DIR=/Applications/freesurfer/8.2.0/subjects
.venv/bin/python examples/inflated_demo.py --subject fsaverage --hemi lh --out out/lh_inflated.png

2) `--subject-dir`¶

.venv/bin/python examples/inflated_demo.py \
  --subject-dir /Users/iancharest/data/nsd_fsaverage \
  --hemi lh \
  --out out/lh_inflated.png

--subject-dir should point at a folder that contains surf/.

Cache Behavior¶

Lookup generation is cached by default.

default cache location:
SUBJECTS_DIR/<subject>/surf/imglookup
or fallback temp cache if path is read-only
override explicitly:
--cache-dir /tmp/cvnpy_cache
disable cache writes/reads for a run:
--no-cache

Smoke Test¶

.venv/bin/python examples/flatmap_demo.py \
  --subject-dir /Users/iancharest/data/nsd_fsaverage \
  --hemi lh \
  --out out/smoke_flat_lh.png

Success criteria: - command exits without traceback - output PNG exists - output is non-empty and visibly cortical (not blank or all-white)