SimpleDet Docs

Installation

Install the lightweight base package, add only the extras needed for CPU, TIMM, geospatial, plotting, docs, or development workflows, then verify the runtime entry points.

Overview Quickstart Core Concepts API Reference

Use this page first if commands from the other docs are failing and you are not yet sure the environment is valid.

On this page: Requirements Install paths Editable development install Verify Troubleshooting

Requirements

  • Python >=3.10
  • Base installs do not pull in PyTorch, TorchVision, TIMM, geospatial, or plotting libraries.
  • PyTorch runtime compatible with your CPU or CUDA environment when using simpledet[cpu].
  • simpledet[timm] if you want suite-driven TIMM backbone swapping

Install paths

Install the base package first for lightweight imports, then add extras for the workflows that need heavier dependencies.

Workflow Command Purpose
Base package python -m pip install simpledet Package metadata, CLI discovery, and lightweight imports without runtime libraries.
CPU runtime python -m pip install 'simpledet[cpu]' Native PyTorch and Lightning runtime validation for training, inference, and evaluation.
TIMM encoders python -m pip install 'simpledet[timm]' TIMM-backed backbone discovery and feature metadata. Combine with cpu for runtime smoke tests.
Documentation checks python -m pip install 'simpledet[docs]' Documentation workflow setup. The current docs checks are lightweight static audits.
Geospatial helpers python -m pip install 'simpledet[geo]' Geospatial data dependencies such as raster and vector IO libraries.
Plotting utilities python -m pip install 'simpledet[plots]' Matplotlib and SciencePlots support for visualization workflows.
Development tooling python -m pip install 'simpledet[dev]' Build, release, coverage, and lint tooling for maintainers. 
python -m pip install simpledet
python -m pip install 'simpledet[cpu]'
python -m pip install 'simpledet[cpu,timm]'
python -m pip install 'simpledet[docs]'
python -m pip install 'simpledet[geo,plots]'

Editable development install

git clone https://github.com/sirbastiano/SimpleDet.git
cd SimpleDet
python -m pip install -e '.[cpu,timm,dev]'

This editable development install includes the CPU runtime, TIMM encoders, and development dependencies for local TIMM validation without requiring MMDetection, MMEngine, or MMCV.

Verify the runtime

python -m simpledet --version
python -m simpledet --check-runtime
python - <<'PY'
from simpledet.detectors.data import list_formats
from simpledet.suite import build_detector, compile_native_detector_plan

print(list_formats())
spec = build_detector("retinanet", encoder="resnet18.a1_in1k", num_classes=2, in_channels=3)
print(compile_native_detector_plan(spec).encoder.type)
PY

Expected output shape:

0.1.0
SimpleDet native runtime stack is available.
['coco', 'csv', 'json', 'voc', 'yolo', ...]
TimmEncoder

Notes

  • The documented training, inference, and evaluation flow is native-only.
  • Use simpledet[cpu] before importing simpledet.native or running CPU runtime commands.
  • Use simpledet[timm] before building TIMM-backed encoders.
  • device="auto" in the lightweight inference helper currently resolves to CPU.

Troubleshooting

  • If python -m simpledet --check-runtime fails, reinstall with python -m pip install 'simpledet[cpu]' and verify that PyTorch and TorchVision wheels match your Python version.
  • If a TIMM-backed detector build cannot find TIMM, install python -m pip install 'simpledet[cpu,timm]' or use the local editable command above.
  • If geospatial or plotting imports fail, install those stacks separately with simpledet[geo], simpledet[plots], or simpledet[geo,plots].
  • If dependency resolution mixes incompatible wheels, create a fresh virtual environment and install one combined extras set instead of layering many partial installs.

SimpleDet uses the native PyTorch and Lightning runtime path. MMDetection, MMEngine, and MMCV are not required runtime dependencies.