descriptron
Image annotation, phenomics and evidence-tiered species descriptions.
pip install torch==2.10.0 torchvision==0.25.0 --index-url https://download.pytorch.org/whl/cu128 # NVIDIA; Mac/CPU: drop --index-url
SAM2_BUILD_CUDA=0 pip install git+https://github.com/facebookresearch/sam2.git # SAM2 library (Windows: set SAM2_BUILD_CUDA=0 first)
pip install descriptron # analysis programs, detectors and the GUI
descriptron-gui # first start downloads the SAM2 model file (898 MB) once
descriptron --list # the 62 analysis programs
SAM2 is not on PyPI, so it cannot be installed with this package: install it from Meta's GitHub as above, into
the same Python (3.10 or newer; 3.12 recommended). The GUI downloads the SAM2 model file sam2_hiera_large.pt once
into ~/.cache/descriptron/sam2/; if you already have it, copy it there or set DESCRIPTRON_SAM2_CHECKPOINT.
Step by step for Windows, macOS and Linux:
https://github.com/alexrvandam/Descriptron/blob/main/docs/INSTALL_PIP.md
This is a convenience package with no code of its own. It installs:
descriptron-core |
the analysis half — pure pip, no torch, no GPU, no compiler |
descriptron-vision |
torchvision detectors, SAM2-PAL propagation, DINOv3 landmark transfer |
descriptron-gui |
the annotation GUI |
If you only want to reproduce an analysis from deposited COCO files, install
descriptron-core alone: it needs no model of any kind and installs in about a
minute on Linux, macOS and Windows.
Detectron2 is an optional advanced backend and is not installed here — it has no PyPI wheel and needs a compiler. See the repository for the Docker image that carries it ready to run.
1 maintainer
Metadata
1 maintainer
Release files for descriptron 2.8.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| descriptron-2.8.1.tar.gz | 7.2 kB | Without provenance | Details for descriptron-2.8.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| descriptron-2.8.1-py3-none-any.whl | Python 3 | none | any | Details for descriptron-2.8.1-py3-none-any.whl |
Total release size: 15.0 kB
Release files / descriptron-2.8.1.tar.gz
| Download URL | descriptron-2.8.1.tar.gz |
|---|---|
| Size | 7.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a7074cc31519978e20b9a1eaaf5dceaac50e29a9afb6a1744537b4de6e7a7e06
|
|
BLAKE2b-256 checksum How to use checksums |
2bcea7ad6b11ac2bcfb49195ab3c55183b787a5d97098be83b1dbccad938872a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.4
|
Release files / descriptron-2.8.1-py3-none-any.whl
| Download URL | descriptron-2.8.1-py3-none-any.whl |
|---|---|
| Size | 7.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
312f1d6654d412f1edda5ef0eee69826f3e3802e58fb8b4f3307aa1b121958c9
|
|
BLAKE2b-256 checksum How to use checksums |
76972d055526952eddd6b52eb6b7343ca47548365e508cd8507de2c4237d8406
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.4
|
Security for descriptron 2.8.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
