segmentation_models.pytorch

(★ 11,702)

Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.

File Explorer

Download Latest Version (.zip)

- devcontainer.json
- - pypi.yml
  - stale.yaml
  - tests.yml
- codecov.yml
- dependabot.yml
- FUNDING.yml
- conf.py
- encoders.rst
- encoders_dpt.rst
- encoders_timm.rst
- index.rst
- insights.rst
- install.rst
- logo.png
- losses.rst
- make.bat
- Makefile
- metrics.rst
- models.rst
- quickstart.rst
- save_load.rst
- binary_segmentation_buildings.py
- binary_segmentation_intro.ipynb
- camvid_segmentation_multiclass.ipynb
- cars segmentation (camvid).ipynb
- convert_to_onnx.ipynb
- dpt_inference_pretrained.ipynb
- save_load_model_and_share_with_hf_hub.ipynb
- segformer_inference_pretrained.ipynb
- upernet_inference_pretrained.ipynb
- LICENSE_apache.md
- LICENSE_apple.md
- LICENSE_nvidia.md
- LICENSES.md
- generate_table.py
- generate_table_timm.py
- generate_test_models.py
- logo-small-h300.png
- logo-small-w300.png
- docs.txt
- minimum.old
- required.txt
- test.txt
- - dpt-original-to-smp.py
  - segformer-original-decoder-to-smp.py
  - upernet-hf-to-smp.py
- - __init__.py
  - heads.py
  - hub_mixin.py
  - initialization.py
  - model.py
  - modules.py
  - utils.py
- - __init__.py
  - oxford_pet.py
- - __init__.py
- - __init__.py
  - _base.py
  - _dpn.py
  - _efficientnet.py
  - _inceptionresnetv2.py
  - _inceptionv4.py
  - _preprocessing.py
  - _senet.py
  - _utils.py
  - _xception.py
  - densenet.py
  - dpn.py
  - efficientnet.py
  - inceptionresnetv2.py
  - inceptionv4.py
  - mix_transformer.py
  - mobilenet.py
  - mobileone.py
  - resnet.py
  - senet.py
  - timm_efficientnet.py
  - timm_sknet.py
  - timm_universal.py
  - timm_vit.py
  - vgg.py
  - xception.py
- - __init__.py
  - _functional.py
  - constants.py
  - dice.py
  - focal.py
  - jaccard.py
  - lovasz.py
  - mcc.py
  - soft_bce.py
  - soft_ce.py
  - tversky.py
- - __init__.py
  - functional.py
- - __init__.py
  - base.py
  - functional.py
  - losses.py
  - meter.py
  - metrics.py
  - train.py
- __init__.py
- __version__.py
- - test_freeze_encoder.py
  - test_modules.py
- - __init__.py
  - base.py
  - test_batchnorm_deprecation.py
  - test_common.py
  - test_pretrainedmodels_encoders.py
  - test_smp_encoders.py
  - test_timm_ported_encoders.py
  - test_timm_universal.py
  - test_timm_vit_encoders.py
  - test_torchvision_encoders.py
- - __init__.py
  - base.py
  - test_deeplab.py
  - test_dpt.py
  - test_fpn.py
  - test_linknet.py
  - test_manet.py
  - test_pan.py
  - test_psp.py
  - test_segformer.py
  - test_unet.py
  - test_unetplusplus.py
  - test_upernet.py
- __init__.py
- conftest.py
- test_base.py
- test_losses.py
- test_preprocessing.py
- utils.py
.gitignore
.readthedocs.yaml
HALLOFFAME.md
LICENSE
Makefile
pyproject.toml
README.md

# Installation Guide

1. Get the code

git clone https://github.com/qubvel-org/segmentation_models.pytorch

Downloads the entire project code from GitHub to your computer.

cd segmentation_models.pytorch

Moves into the project folder you just downloaded.

2. Python

Easy Recommended

Prerequisites

Git Needed to download the project code from GitHub.
Python 3 On Windows, be sure to check 'Add Python to PATH' during installation.

$ pip install segmentation-models-pytorch

Installs the Python libraries listed in requirements.txt (or similar).

$ pip install git+https://github.com/qubvel/segmentation_models.pytorch

Installs the Python libraries listed in requirements.txt (or similar).

✅ If it runs without errors and prints output in the terminal, it worked.

Pulled directly from this repo's README.

3. Make

Medium

Prerequisites

Git Needed to download the project code from GitHub.
Make Usually pre-installed on Linux/macOS. On Windows, install separately (e.g. via MSYS2 or WSL).

make install_dev # Create .venv, install SMP in dev mode

Compiles the code based on the generated build configuration to produce an executable.

make test # Run tests suite with pytest

Compiles the code based on the generated build configuration to produce an executable.

make fixup # Ruff for formatting and lint checks

Compiles the code based on the generated build configuration to produce an executable.

make table # Generates a table with encoders and print to stdout

Compiles the code based on the generated build configuration to produce an executable.

✅ If it finishes without errors, it worked. Try running the generated executable directly.

Pulled directly from this repo's README.