ModelArts provides deep learning-powered base images such as TensorFlow, PyTorch, and MindSpore images. In these images, the software mandatory for running training jobs has been installed. If the software in the base images cannot meet your service requirements, create new images based on the base images and use the new images to create training jobs.
Available Training Base Images
The following table lists the preset training base images of ModelArts.
Table 1 ModelArts training base images
|
Engine |
Version |
|
PyTorch |
pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64 |
|
TensorFlow |
tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64 |
|
Horovod |
horovod_0.20.0-tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64 |
|
horovod_0.22.1-pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64 |
|
MPI |
mindspore_1.3.0-cuda_10.1-py_3.7-ubuntu_1804-x86_64 |
Supported AI engines vary depending on regions.
Engine Version: pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64
- Image address: swr.{Region}.myhuaweicloud.com/aip/pytorch_1_8:train-pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64-roma-20220309171256-40adcc1
- Image creation time: 20220309171256 (yyyy-mm-dd-hh-mm-ss)
- Image system version: Ubuntu 18.04.4 LTS
- CUDA: 10.2.89
- cuDNN: 7.6.5.32
- Path and version of the Python interpreter: /home/ma-user/anaconda3/envs/PyTorch-1.8/bin/python, Python 3.7.10
- Third-party package installation path: /home/ma-user/anaconda3/envs/PyTorch-1.8/lib/python3.7/site-packages
TensorFlow Base Images
This section describes preset TensorFlow images.
Engine Version: tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64
- Image address: swr.{Region}.myhuaweicloud.com/aip/tensorflow_2_1:train-tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64-20210912152543-1e0838d
- Image creation time: 20210912152543(yyyy-mm-dd-hh-mm-ss)
- Image system version: Ubuntu 18.04.4 LTS
- CUDA: 10.1.243
- cuDNN: 7.6.5.32
- Path and version of the Python interpreter: /home/ma-user/anaconda3/envs/TensorFlow-2.1/bin/python, Python 3.7.10
- Third-party package installation path: /home/ma-user/anaconda3/envs/TensorFlow-2.1/lib/python3.7/site-packages
Horovod Base Images
This section describes preset Horovod images.
Engine Version 1: horovod_0.20.0-tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64
- Image address: swr.{Region}.myhuaweicloud.com/aip/horovod_tensorflow:train-horovod_0.20.0-tensorflow_2.1.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64-20210912152543-1e0838d
- Image creation time: 20210912152543 (yyyy-mm-dd-hh-mm-ss)
- Image system version: Ubuntu 18.04.4 LTS
- CUDA: 10.1.243
- cuDNN: 7.6.5.32
- Path and version of the Python interpreter: /home/ma-user/anaconda3/envs/horovod_0.20.0-tensorflow_2.1.0/bin/python, Python 3.7.10
- Third-party package installation path: /home/ma-user/anaconda3/envs/horovod_0.20.0-tensorflow_2.1.0/lib/python3.7/site-packages
Engine Version 2: horovod_0.22.1-pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64
- Image address: swr.{Region}.myhuaweicloud.com/aip/horovod_pytorch:train-horovod_0.22.1-pytorch_1.8.0-cuda_10.2-py_3.7-ubuntu_18.04-x86_64-20210912152543-1e0838d
- Image creation time: 20210912152543 (yyyy-mm-dd-hh-mm-ss)
- Image system version: Ubuntu 18.04.4 LTS
- CUDA: 11.1.1
- cuDNN: 8.0.5.39
- Path and version of the Python interpreter: /home/ma-user/anaconda3/envs/horovod-0.22.1-pytorch-1.8.0/bin/python, Python 3.7.10
- Third-party package installation path: /home/ma-user/anaconda3/envs/horovod-0.22.1-pytorch-1.8.0/lib/python3.7/site-packages
Engine Version: mindspore_1.3.0-cuda_10.1-py_3.7-ubuntu_1804-x86_64
- Image address: swr.{Region}.myhuaweicloud.com/aip/mindspore_1_3_0:train-mindspore_1.3.0-cuda_10.1-py_3.7-ubuntu_18.04-x86_64-roma-20211104202338-f258e59
- Image creation time: 20211104202338 (yyyy-mm-dd-hh-mm-ss)
- Image system version: Ubuntu 18.04.4 LTS
- CUDA: 10.1.243
- cuDNN: 7.6.5.32
- Path and version of the Python interpreter: /home/ma-user/anaconda3/envs/MindSpore-1.3.0-gpu/bin/python, Python 3.7.10
- Third-party package installation path: /home/ma-user/anaconda3/envs/MindSpore-1.3.0-gpu/lib/python3.7/site-packages
