Collection of Docker images for ML/DL and video processing projects

Supported NVIDIA architectures: Pascal (sm_60, sm_61), Volta (sm_70), Turing (sm_75), Ampere (sm_80, sm_86), Ada Lovelace (sm_89).

CUDA (11.8.0), cuDNN (8.6.0) FFmpeg (release/4.4), nv-codec-headers (sdk/11.0) Python (3.10.6) CMake (3.22.1)

pip==22.3.1 setuptools==65.5.1 packaging==21.3 numpy==1.23.4 opencv-python==4.6.0.66 scipy==1.9.3 matplotlib==3.6.2 pandas==1.5.1 scikit-learn==1.1.3 scikit-image==0.19.3 Pillow==9.3.0 librosa==0.9.2 albumentations==1.3.0 pyzmq==24.0.1 Cython==0.29.32 numba==0.56.4 requests==2.28.1 psutil==5.9.4 pydantic==1.10.2 PyYAML==6.0 notebook==6.5.2 ipywidgets==8.0.2 tqdm==4.64.1 pytest==7.2.0 pytest-cov==4.0.0 mypy==0.991 flake8==5.0.4 pre-commit==2.20.0

additionally to dokai:22.11-base:

TensorRT (8.5.1) MAGMA (2.6.2)

torch==1.14.0a0+git71fe069 (source, close to v1.13.0 after commit "ada lovelace (arch 8.9) support #87436") torchvision==0.14.0 (source, v0.14.0 tag) torchaudio==0.13.0 (source, v0.13.0 tag) pytorch-ignite==0.4.10 pytorch-argus==1.0.0 pretrainedmodels==0.7.4 efficientnet-pytorch==0.7.1 pytorch-toolbelt==0.5.2 kornia==0.6.8 timm==0.6.11 segmentation-models-pytorch==0.3.0

additionally to dokai:22.11-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.4.2), cuDNN (8.2.4)
FFmpeg (release/4.4), nv-codec-headers (sdk/11.0)
Python (3.8.10)
CMake (3.21.4)

pip==21.3.1
setuptools==58.5.3
packaging==21.2
numpy==1.21.4
opencv-python==4.5.4.58
scipy==1.7.2
matplotlib==3.4.3
pandas==1.3.4
scikit-learn==1.0.1
scikit-image==0.18.3
Pillow==8.4.0
librosa==0.8.1
albumentations==1.1.0
pyzmq==22.3.0
Cython==0.29.24
numba==0.53.1
requests==2.26.0
psutil==5.8.0
pydantic==1.8.2
PyYAML==6.0
notebook==6.4.5
ipywidgets==7.6.5
tqdm==4.62.3
pytest==6.2.5
mypy==0.910
flake8==4.0.1

additionally to dokai:21.11-base:

MAGMA (2.6.1)

torch==1.10.0 (source, v1.10.0 tag)
torchvision==0.11.1 (source, v0.11.1 tag)
torchaudio==0.10.0 (source, v0.10.0 tag)
pytorch-ignite==0.4.7
pytorch-argus==1.0.0
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.1
timm==0.4.12
segmentation-models-pytorch==0.2.0
kornia==0.6.1

additionally to dokai:21.11-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.4.2), cuDNN (8.2.4)
FFmpeg (release/4.4), nv-codec-headers (sdk/11.0)
Python (3.8.10)

pip==21.2.4
setuptools==58.1.0
packaging==21.0
numpy==1.21.2
opencv-python==4.5.3.56
scipy==1.7.1
matplotlib==3.4.3
pandas==1.3.3
scikit-learn==1.0
scikit-image==0.18.3
Pillow==8.3.2
librosa==0.8.1
albumentations==1.0.3
pyzmq==22.3.0
Cython==0.29.24
numba==0.53.1
requests==2.26.0
psutil==5.8.0
pydantic==1.8.2
PyYAML==5.4.1
notebook==6.4.4
ipywidgets==7.6.5
tqdm==4.62.3
pytest==6.2.5
mypy==0.910
flake8==3.9.2

additionally to dokai:21.09-base:

MAGMA (2.6.1)

torch==1.10.0-rc1 (source, v1.10.0-rc1 tag)
torchvision==0.10.1 (source, v0.10.1 tag)
torchaudio==0.9.1 (source, v0.9.1 tag)
pytorch-ignite==0.4.6
pytorch-argus==0.2.1
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.1
timm==0.4.12
segmentation-models-pytorch==0.2.0
kornia==0.5.11
apex (source, master branch)

additionally to dokai:21.09-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.4.1), cuDNN (8.2.2)
FFmpeg (release/4.4), nv-codec-headers (sdk/11.0)
Python (3.8.10)

pip==21.2.3
setuptools==57.4.0
packaging==21.0
numpy==1.21.1
opencv-python==4.5.3.56
scipy==1.7.1
matplotlib==3.4.2
pandas==1.3.1
scikit-learn==0.24.2
scikit-image==0.18.2
Pillow==8.3.1
librosa==0.8.1
albumentations==1.0.3
pyzmq==22.2.1
Cython==0.29.24
numba==0.53.1
requests==2.26.0
psutil==5.8.0
pydantic==1.8.2
PyYAML==5.4.1
notebook==6.4.3
ipywidgets==7.6.3
tqdm==4.62.0
pytest==6.2.4
mypy==0.910
flake8==3.9.2

additionally to dokai:21.08-base:

MAGMA (2.6.1)

torch==1.10.0a0+git5b8389e (source, master branch)
torchvision==0.10.0 (source, v0.10.0 tag)
torchaudio==0.9.0 (source, v0.9.0 tag)
pytorch-ignite==0.4.6
pytorch-argus==0.2.1
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.1
timm==0.4.12
segmentation-models-pytorch==0.2.0
kornia==0.5.8
apex (source, master branch)

additionally to dokai:21.08-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.3.1), cuDNN (8.2.0)
FFmpeg (release/4.4), nv-codec-headers (sdk/10.0)
Python (3.8.10)

pip==21.1.3
setuptools==57.0.0
packaging==20.9
numpy==1.21.0
opencv-python==4.5.2.54
scipy==1.7.0
matplotlib==3.4.2
pandas==1.2.5
scikit-learn==0.24.2
scikit-image==0.18.2
Pillow==8.2.0
librosa==0.8.1
albumentations==1.0.0
pyzmq==22.1.0
Cython==0.29.23
numba==0.53.1
requests==2.25.1
psutil==5.8.0
trafaret-config==2.0.2
pydantic==1.8.2
PyYAML==5.4.1
notebook==6.4.0
ipywidgets==7.6.3
tqdm==4.61.1
pytest==6.2.4
mypy==0.910
flake8==3.9.2

additionally to dokai:21.07-base:

torch==1.9.0 (source, v1.9.0 tag)
torchvision==0.10.0 (source, v0.10.0 tag)
torchaudio==0.9.0 (source, v0.9.0 tag)
pytorch-argus==0.2.1
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.1
timm==0.4.12
segmentation-models-pytorch==0.1.3
kornia==0.5.5
apex (source, master branch)

additionally to dokai:21.07-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.3), cuDNN (8.2.0)
FFmpeg (release/4.4), nv-codec-headers (sdk/10.0)
Python (3.8.5)

pip==21.1.1
setuptools==56.2.0
packaging==20.9
numpy==1.20.3
opencv-python==4.5.2.52
scipy==1.6.3
matplotlib==3.4.2
pandas==1.2.4
scikit-learn==0.24.2
scikit-image==0.18.1
Pillow==8.2.0
librosa==0.8.0
albumentations==0.5.2
pyzmq==22.0.3
Cython==0.29.23
numba==0.53.1
requests==2.25.1
psutil==5.8.0
trafaret-config==2.0.2
pydantic==1.8.1
PyYAML==5.4.1
notebook==6.3.0
ipywidgets==7.6.3
tqdm==4.60.0
pytest==6.2.4
mypy==0.812
flake8==3.9.2

additionally to dokai:21.05-base:

torch==1.8.1 (source, v1.8.1 tag)
torchvision==0.9.1 (source, v0.9.1 tag)
torchaudio==0.8.1 (source, v0.8.1 tag)
pytorch-argus==0.2.1
timm==0.4.8 (source, master branch)
kornia==0.5.1
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.1
segmentation-models-pytorch==0.1.3
apex (source, master branch)

additionally to dokai:21.05-pytorch:

tensor-stream==0.4.6 (source, dev branch)

ghcr.io/osai-ai/dokai:21.03-base

CUDA (11.2.2), cuDNN (8.1.1)
FFmpeg (release/4.4), nv-codec-headers (sdk/10.0)
Python (3.8.5)

pip==21.0.1
setuptools==54.2.0
packaging==20.9
numpy==1.20.1
opencv-python==4.5.1.48
scipy==1.6.1
matplotlib==3.3.4
pandas==1.2.3
scikit-learn==0.24.1
scikit-image==0.18.1
Pillow==8.1.2
librosa==0.8.0
albumentations==0.5.2
pyzmq==22.0.3
Cython==0.29.22
numba==0.53.0
requests==2.25.1
psutil==5.8.0
trafaret-config==2.0.2
pydantic==1.8.1
PyYAML==5.4.1
notebook==6.3.0
ipywidgets==7.6.3
tqdm==4.59.0
pytest==6.2.2
mypy==0.812
flake8==3.9.0

additionally to dokai:21.03-base:

torch==1.8.0 (source, v1.8.0 tag)
torchvision==0.9.0 (source, v0.9.0 tag)
torchaudio==0.8.0 (source, v0.8.0 tag)
pytorch-argus==0.2.1
timm==0.4.5
kornia==0.5.0
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.0
segmentation-models-pytorch==0.1.3
apex (source, master branch)

additionally to dokai:21.03-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.2.1), cuDNN (8.1.0)
FFmpeg (release/4.3), nv-codec-headers (sdk/10.0)
Python (3.8.5)

pip==21.0.1
setuptools==53.0.0
packaging==20.9
numpy==1.20.1
opencv-python==4.5.1.48
scipy==1.6.1
matplotlib==3.3.4
pandas==1.2.2
scikit-learn==0.24.1
scikit-image==0.18.1
Pillow==8.1.0
librosa==0.8.0
albumentations==0.5.2
pyzmq==22.0.3
Cython==0.29.22
numba==0.52.0
requests==2.25.1
psutil==5.8.0
trafaret-config==2.0.2
pydantic==1.7.3
PyYAML==5.4.1
notebook==6.2.0
ipywidgets==7.6.3
tqdm==4.57.0
pytest==6.2.2
mypy==0.812
flake8==3.8.4

additionally to dokai:21.02-base:

torch==1.9.0a0+c2b9283 (source, master branch)
torchvision==0.8.2 (source, v0.8.2 tag)
pytorch-argus==0.2.0
timm==0.4.4 (source, master branch)
kornia==0.4.1
pretrainedmodels==0.7.4
efficientnet-pytorch==0.7.0
segmentation-models-pytorch==0.1.3
apex (source, master branch)

additionally to dokai:21.02-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.1.1), cuDNN (8.0.5)
FFmpeg (release/4.3), nv-codec-headers (sdk/10.0)
Python (3.8.5)

pip==20.3.3
setuptools==51.3.3
packaging==20.8
numpy==1.19.5
opencv-python==4.5.1.48
scipy==1.6.0
matplotlib==3.3.3
pandas==1.2.0
notebook==6.2.0
scikit-learn==0.24.1
scikit-image==0.18.1
albumentations==0.5.2
Cython==0.29.21
Pillow==8.1.0
trafaret-config==2.0.2
pyzmq==21.0.1
librosa==0.8.0
psutil==5.8.0
pydantic==1.7.3
requests==2.25.1

additionally to dokai:21.01-base:

torch==1.8.0a0+4aea007 (source, master branch)
torchvision==0.8.2 (source, v0.8.2 tag)
pytorch-argus==0.2.0
timm==0.3.4
kornia==0.4.1
apex (source, master branch)

additionally to dokai:21.01-pytorch:

tensor-stream==0.4.6 (source, dev branch)

CUDA (11.1), cuDNN (8.0.5) FFmpeg (release/4.3), nv-codec-headers (sdk/9.1)
Python (3.8.5)

pip==20.3.3
setuptools==51.0.0
packaging==20.8
numpy==1.19.4
opencv-python==4.4.0.46
scipy==1.5.4
matplotlib==3.3.3
pandas==1.1.5
notebook==6.1.5
scikit-learn==0.23.2
scikit-image==0.18.0
albumentations==0.5.2
Cython==0.29.21
Pillow==8.0.1
trafaret-config==2.0.2
pyzmq==20.0.0
librosa==0.8.0
psutil==5.8.0
pydantic==1.7.3
requests==2.25.1

additionally to dokai:20.12-base:

torch==1.7.1 (source, v1.7.1 tag)
torchvision==0.8.2 (source, v0.8.2 tag)
pytorch-argus==0.2.0
timm==0.3.2
kornia==0.4.1
apex (source, master branch)

additionally to dokai:20.12-pytorch:

tensor-stream==0.4.6 (source, dev branch)

FFmpeg (release/4.3), nv-codec-headers (sdk/9.1)
Python (3.6.9)

pip==20.2.4
setuptools==50.3.2
packaging==20.4
numpy==1.19.2
opencv-python==4.4.0.44
scipy==1.5.3
matplotlib==3.3.2
pandas==1.1.3
notebook==6.1.4
scikit-learn==0.23.2
scikit-image==0.17.2
albumentations==0.5.0
Cython==0.29.21
Pillow==8.0.0
trafaret-config==2.0.2
pyzmq==19.0.2
librosa==0.8.0
psutil==5.7.2
dataclasses==0.7
pydantic==1.6.1
requests==2.24.0

additionally to dokai:20.10-base:

torch==1.6.0
torchvision==0.7.0
pytorch-argus==0.1.2
timm==0.2.1
apex (master)

additionally to dokai:20.10-pytorch:

tensor-stream==0.4.6 (dev)

FFmpeg (release/4.3), nv-codec-headers (sdk/9.1)
Python (3.6.9)

pip==20.2.3
setuptools==50.3.0
packaging==20.4
numpy==1.19.2
opencv-python==4.4.0.42
scipy==1.5.2
matplotlib==3.3.2
pandas==1.1.2
notebook==6.1.4
scikit-learn==0.23.2
scikit-image==0.17.2
albumentations==0.4.6
Cython==0.29.21
Pillow==7.2.0
trafaret-config==2.0.2
pyzmq==19.0.2
librosa==0.8.0
psutil==5.7.2
dataclasses==0.7

additionally to dokai:20.09-base:

torch==1.6.0
torchvision==0.7.0
pytorch-argus==0.1.2
timm==0.2.1
apex (master)

additionally to dokai:20.09-pytorch:

tensor-stream==0.4.6 (dev)