Help Center> ModelArts> Service Overview> Supported AI Frameworks
View PDF

Supported AI Frameworks

Supported AI frameworks and versions of ModelArts vary slightly based on the development environment, training jobs, and model inference (model management and deployment). The following describes the AI frameworks supported by each module.

Development Environment

Notebook instances in the development environment support different AI engines and versions based on specific working environments (that is, different Python versions). After creating a notebook instance in the corresponding working environment, create a file based on the corresponding version in Table 1. ModelArts notebook instances support multiple engines. That is, a notebook instance can use all supported engines. Different engines can be switched quickly and conveniently.

Table 1 AI engines

Framework

Version

Python Version

Chip

TensorFlow
  • 1.8
  • 1.13.1
  • 2.7
  • 3.6
  • CPU
  • GPU

TensorFlow

2.1.0

3.6
  • CPU
  • GPU

TensorFlow

1.15.0

3.7

Ascend 910

MindSpore

0.5

3.7

Ascend 910

MXNet

1.2.1
  • 2.7
  • 3.6
  • CPU
  • GPU

Caffe

1.0.0

2.7
  • CPU
  • GPU

PySpark

2.3.2
  • 2.7
  • 3.6

CPU

Scikit-learn & XGBoost

XGBoost 0.80, Sklearn 0.20.0
  • 2.7
  • 3.6

CPU

Conda

4.3.30

2.7
  • CPU
  • GPU

Conda

4.4.10

3.6
  • CPU
  • GPU

PyTorch

1.0.0
  • 2.7
  • 3.6

GPU

PyTorch

1.4.0

3.6

GPU

Training Jobs

Supported AI engines and versions when creating training jobs are as follows:

  • TensorFlow: TF-1.8.0-python3.6, TF-1.8.0-python2.7, TF-1.13.1-python3.6, TF-1.13.1-python2.7, TF-2.1.0-python3.6
  • MXNet: MXNet-1.2.1-python3.6, MXNet-1.2.1-python2.7
  • Caffe: Caffe-1.0.0-python2.7
  • Spark_MLlib: Spark-2.3.2-python2.7, Spark-2.3.2-python3.6
  • Ray: RAY-0.7.4-python3.6
  • XGBoost-Sklearn: XGBoost-0.80-Sklearn-0.18.1-python2.7, XGBoost-0.80-Sklearn-0.18.1-python3.6
  • PyTorch: PyTorch-1.0.0-python2.7, PyTorch-1.0.0-python3.6, PyTorch-1.3.0-python2.7, PyTorch-1.3.0-python3.6, PyTorch-1.4.0-python3.6
  • Ascend-Powered-Engine: MindSpore-0.5-python3.7-aarch64 and TF-1.15-python3.7-aarch64
  • Ascend-Powered-Engine is only available in the CN North-Beijing4 region.

Model Inference

For imported models and model inference is completed on ModelArts, supported engines and their runtime are as follows:

Table 2 Supported engines and their runtime

Engine

Runtime

Notes

TensorFlow

python3.6

python2.7

tf1.13-python2.7-gpu

tf1.13-python2.7-cpu

tf1.13-python3.6-gpu

tf1.13-python3.6-cpu

tf1.13-python3.7-cpu

tf1.13-python3.7-gpu

tf2.1-python3.7
  • TensorFlow 1.8.0 is used in python2.7, python3.7, and python3.6.
  • python2.7, tf2.1-python3.7, and python3.6 indicate that the model can run on the CPU or GPU at the same time. For other runtime values, if the suffix contains cpu or gpu, the model can run only in the CPU or GPU.
  • The default runtime is python2.7.

MXNet

python3.7

python3.6

python2.7
  • MXNet 1.2.1 is used in python2.7, python3.7, and python3.6.
  • python2.7, python3.7, and python3.6 indicate that the model can run on the CPU or GPU at the same time.
  • The default runtime is python2.7.

Caffe

python2.7

python3.6

python3.7

python2.7-gpu

python3.6-gpu

python3.7-gpu

python2.7-cpu

python3.6-cpu

python3.7-cpu
  • Caffe 1.0.0 is used in python2.7, python3.6, python3.7, python2.7-gpu, python3.6-gpu, python3.7-gpu, python2.7-cpu, python3.7-cpu, and python3.6-cpu.
  • python 2.7, python3.7, and python 3.6 can only be used to run models applicable to CPU. For other runtime values, if the suffix contains cpu or gpu, the model can run only in the CPU or GPU. You are advised to use the runtime of python2.7-gpu, python3.6-gpu, python3.7-gpu, python2.7-cpu, python3.7-cpu, and python3.6-cpu.
  • The default runtime is python2.7.

Spark_MLlib

python2.7

python3.6
  • Spark_MLlib 2.3.2 is used in python2.7 and python3.6.
  • The default runtime is python2.7.
  • python 2.7 and python 3.6 can only be used to run models applicable to CPU.

Scikit_Learn

python2.7

python3.6
  • Scikit_Learn 0.18.1 is used in python2.7 and python3.6.
  • The default runtime is python2.7.
  • python 2.7 and python 3.6 can only be used to run models applicable to CPU.

XGBoost

python2.7

python3.6
  • XGBoost 0.80 is used in python2.7 and python3.6.
  • The default runtime is python2.7.
  • python 2.7 and python 3.6 can only be used to run models applicable to CPU.

PyTorch

python2.7

python3.6

python3.7

pytorch1.4-python3.7
  • PyTorch 1.0 is used in python2.7, python3.7, and python3.6.
  • python2.7, python3.6, python3.7, and pytorch1.4-python3.7 indicate that the model can run on the CPU or GPU at the same time.
  • The default runtime is python2.7.

MindSpore

python3.7

MindSpore 0.5 is used in python3.7.