Jiang, Beibei 781e07249c ModelArts 2021430 (GA) UMN 25072022 provided by R&D (third review)

Reviewed-by: gtema <artem.goncharov@gmail.com>
Co-authored-by: Jiang, Beibei <beibei.jiang@t-systems.com>
Co-committed-by: Jiang, Beibei <beibei.jiang@t-systems.com>

2022-09-06 10:45:57 +00:00

29 KiB

Raw Permalink Blame History

Which AI Frameworks Does ModelArts Support?

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 work environments (that is, different Python versions). After creating a notebook instance in the corresponding work environment, create a file based on the corresponding version in Table AI engines. 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
Work Environment	Built-in AI Engine and Version	Supported Chip
Multi-Engine 1.0 (Python 3, Recommended)	MXNet-1.2.1	CPU/GPU
	PySpark-2.3.2	CPU
	Pytorch-1.0.0	GPU
	TensorFlow-1.13.1	CPU/GPU
	TensorFlow-1.8	CPU/GPU
	XGBoost-Sklearn	CPU
Multi-Engine 1.0 (Python2)	Caffe-1.0.0	CPU/GPU
	MXNet-1.2.1	CPU/GPU
	PySpark-2.3.2	CPU
	PyTorch1.0.0	GPU
	TensorFlow-1.13.1	CPU/GPU
	TensorFlow-1.8	CPU/GPU
	XGBoost-Sklearn	CPU
Multi-Engine 2.0 (Python3)	Pytorch-1.4.0	GPU
	R-3.6.1	CPU/GPU
	TensorFlow-2.1.0	CPU/GPU

Training Jobs

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

**Table 2** AI engines supported by training jobs
Environment	Supported Chip	System Architecture	System Version	AI Engine and Version	Supported CUDA Version
TensorFlow	CPU and GPU	x86_64	Ubuntu 16.04	TF-1.13.1-python3.6	CUDA 10.0
				TF-1.8.0-python3.6	CUDA 9.0
				TF-2.1.0-python3.6	CUDA 10.1
Caffe	CPU and GPU	x86_64	Ubuntu 16.04	Caffe-1.0.0-python2.7	CUDA 8.0
Spark_MLlib	CPU	x86_64	Ubuntu 16.04	Spark-2.3.2-python3.6	N/A
XGBoost-Sklearn	CPU	x86_64	Ubuntu 16.04	Scikit_Learn-0.18.1-python3.6	N/A
PyTorch	CPU and GPU	x86_64	Ubuntu 16.04	PyTorch-1.3.0-python3.6	CUDA 10.0
PyTorch	CPU and GPU	x86_64	Ubuntu 16.04	PyTorch-1.0.0-python3.6	CUDA 9.0
MXNet	CPU/GPU	x86_64	Ubuntu16.04	MXNet-1.2.1-python3.6	CUDA 9.0

Model Inference

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

**Table 3** Supported AI engines and their runtime
Engine	Runtime	Precautions
TensorFlow	python3.6 python2.7	TensorFlow 1.8.0 is used in python2.7 and python3.6. python3.6 and python2.7 indicate that the model can run on both CPUs and GPUs. The default runtime is python3.6.
MXNet	python3.7 python2.7	MXNet 1.2.1 is used in python3.7 and python2.7. python3.6 and python2.7 indicate that the model can run on both CPUs and GPUs. The default runtime is python3.7.
Caffe	python2.7-gpu python3.7-gpu python2.7-cpu python3.7-cpu	Caffe 1.0.0 is used in python2.7-gpu, python3.7-gpu, python2.7-cpu, and python3.7-cpu. For runtime values, if the suffix contains cpu or gpu, the model can run only on CPUs or GPUs. Use the runtime of python3.6-gpu, python3.7-gpu, python3.6-cpu, or python3.7-cpu. The default runtime is python3.7.
PyTorch	python2.7 python3.7	PyTorch 1.0 is used in python2.7 and python3.7. python2.7 and python3.7 indicate that the model can run on both CPUs and GPUs. The default runtime is python3.7.

Parent topic: General Issues

29 KiB Raw Permalink Blame History

Which AI Frameworks Does ModelArts Support?

Development Environment

Training Jobs

Model Inference

29 KiB

Raw Permalink Blame History