Welcome to the Model Zoo for MindSpore

In order to facilitate developers to enjoy the benefits of MindSpore framework, we will continue to add typical networks and some of the related pre-trained models. If you have needs for the model zoo, you can file an issue on gitee or MindSpore, We will consider it in time.

• SOTA models using the latest MindSpore APIs

• The best benefits from MindSpore

• Officially maintained and supported

Table of Contents

Official

Domain	Sub Domain	Network	Ascend	GPU	CPU
Audio	Speaker Recognition	ecapa_tdnn	✅
Audio	Speech Synthesis	lpcnet	✅	✅
Audio	Speech Synthesis	melgan	✅	✅
Audio	Speech Synthesis	tacotron2	✅
Graph Neural Network	Text Classification	bgcf	✅	✅
Graph Neural Network	Text Classification	gat	✅	✅
Graph Neural Network	Text Classification	gcn	✅	✅
Recommendation	Recommender System	naml	✅	✅
Recommendation	Recommender System	ncf	✅	✅
Recommendation	Recommender System	tbnet	✅	✅
Image	Image Classification	alexnet	✅	✅
Image	Image Denoise	brdnet	✅
Image	Object Detection	centerface	✅	✅	✅
Image	Image Classification	cnn_direction_model	✅	✅
Image	Scene Text Recognition	cnnctc	✅	✅	✅
Image	Scene Text Recognition	crnn	✅	✅	✅
Image	Scene Text Recognition	crnn_seq2seq_ocr	✅
Image	Image Classification	cspdarknet53	✅
Image	Object Detection	ctpn	✅	✅
Image	Object Detection	darknet53		✅
Image	Semantic Segmentation	deeplabv3	✅	✅	✅
Image	Text Detection	deeptext	✅	✅
Image	Image Classification	densenet100	✅	✅
Image	Image Classification	densenet121	✅	✅
Image	Depth Estimation	depthnet	✅
Image	Image Denoise	dncnn		✅
Image	Image Classification	dpn	✅	✅
Image	Scene Text Detection	east	✅	✅
Image	Image Classification	efficientnet		✅	✅
Image	Image Classification	erfnet	✅	✅
Image	Scene Text Recognition	essay-recogination		✅
Image	Object Detection	FasterRCNN_Inception_Resnetv2	✅	✅
Image	Object Detection	FasterRCNN_ResNetV1.5_50	✅	✅
Image	Object Detection	FasterRCNN_ResNetV1_101	✅	✅
Image	Object Detection	FasterRCNN_ResNetV1_152	✅	✅
Image	Object Detection	FasterRCNN_ResNetV1_50	✅	✅
Image	Semantic Segmentation	fastscnn	✅
Image	Semantic Segmentation	FCN8s	✅	✅
Image	Image Classification	googlenet	✅	✅
Image	Image Classification	inceptionv3	✅	✅	✅
Image	Image Classification	inceptionv4	✅	✅	✅
Image	Image Denoise	LearningToSeeInTheDark	✅
Image	Image Classification	lenet	✅	✅	✅
Image	Object Detection	maskrcnn_resnet50	✅	✅
Image	Object Detection	maskrcnn_mobilenetv1	✅	✅	✅
Image	Crowd Counting	MCNN	✅	✅
Image	Image Classification	mobilenetv1	✅	✅
Image	Image Classification	mobilenetv2	✅	✅	✅
Image	Image Classification	mobilenetv3	✅	✅	✅
Image	Image Classification	nasnet	✅	✅
Image	Image Quality Assessment	nima	✅	✅
Image	Point Cloud Model	octsqueeze	✅	✅
Image	Keypoint Detection	openpose	✅
Image	Defect Detection	patchcore	✅	✅
Image	Camera Relocalization	posenet	✅	✅
Image	Video Predictive Learning	predrnn++	✅
Image	Scene Text Detection	psenet	✅	✅
Image	Pose Estimation	pvnet	✅
Image	Optical Flow Estimation	pwcnet	✅	✅
Image	Image Super Resolution	RDN	✅	✅
Image	Image Classification	resnet101	✅	✅	✅
Image	Image Classification	resnet152	✅	✅	✅
Image	Image Classification	resnet18	✅	✅	✅
Image	Image Classification	resnet34	✅	✅	✅
Image	Image Classification	resnet50	✅	✅	✅
Image	Image Classification	resnet50_thor	✅	✅
Image	Image Classification	resnext101	✅	✅
Image	Image Classification	resnext50	✅	✅
Image	Object Detection	retinaface_resnet50		✅
Image	Object Detection	retinanet	✅	✅
Image	Image Classification	se_resnext50	✅
Image	Image Matting	semantic_human_matting	✅
Image	Image Classification	se-resnet50	✅	✅	✅
Image	Image Classification	shufflenetv1	✅	✅	✅
Image	Image Classification	shufflenetv2	✅	✅	✅
Image	Image Classification	simclr	✅	✅
Image	Keypoint Detection	simple_pose	✅	✅
Image	Object Detection	sphereface	✅	✅
Image	Image Classification	squeezenet	✅	✅
Image	Image Classification	SqueezeNet_Residual	✅	✅
Image	Image Super Resolution	srcnn	✅	✅
Image	Object Detection	ssd_mobilenet-v1-fpn	✅	✅	✅
Image	Object Detection	ssd-mobilenet-v2	✅	✅	✅
Image	Object Detection	ssd-resnet50-fpn	✅	✅	✅
Image	Object Detection	ssd-vgg16	✅	✅	✅
Image	Defect Detection	ssim-ae	✅
Image	Image Classification	tinydarknet	✅	✅	✅
Image	Semantic Segmentation	UNet_nested	✅	✅
Image	Semantic Segmentation	unet2d	✅	✅
Image	Semantic Segmentation	unet3d	✅	✅
Image	Image Classification	vgg16	✅	✅	✅
Image	Image Classification	vit	✅	✅
Image	Scene Text Recognition	warpctc	✅	✅
Image	Image Classification	xception	✅	✅
Image	Object Detection	yolov3_darknet53	✅	✅
Image	Object Detection	yolov3_resnet18	✅
Image	Object Detection	yolov4	✅
Image	Object Detection	yolov5s	✅	✅
Recommendation	Click-Through Rate Prediction	deep_and_cross		✅
Recommendation	Click-Through Rate Prediction	deepfm	✅	✅
Recommendation	Click-Through Rate Prediction	fibinet		✅
Recommendation	Click-Through Rate Prediction	wide_and_deep	✅	✅
Recommendation	Click-Through Rate Prediction	wide_and_deep_multitable	✅	✅
Text	Natural Language Understanding	bert_base	✅	✅
Text	Natural Language Understanding	bert_bilstm_crf	✅	✅
Text	Natural Language Understanding	bert_finetuning	✅	✅
Text	Natural Language Understanding	bert_large	✅
Text	Natural Language Understanding	bert_nezha	✅	✅
Text	Natural Language Understanding	cpm	✅	✅
Text	Dialogue	dgu	✅	✅
Text	Dialogue	duconv	✅	✅
Text	Emotion Classification	emotect	✅	✅
Text	Natural Language Understanding	ernie	✅	✅
Text	Natural Language Understanding	fasttext	✅	✅
Text	Natural Language Understanding	gnmt_v2	✅	✅
Text	Natural Language Understanding	gpt3	✅
Text	Natural Language Understanding	gru	✅	✅
Text	Emotion Classification	lstm	✅	✅
Text	Natural Language Understanding	mass	✅	✅
Text	Pre Training	pangu_alpha	✅	✅
Text	Natural Language Understanding	textcnn	✅	✅
Text	Natural Language Understanding	tinybert	✅	✅
Text	Natural Language Understanding	transformer	✅	✅
Video	Object Tracking	ADNet	✅
Video	Video Classification	c3d	✅	✅
Video	Object Tracking	Deepsort	✅	✅

Research

Domain	Sub Domain	Network	Ascend	GPU	CPU
3D	3D Reconstruction	cmr		✅
3D	3D Reconstruction	DecoMR		✅
3D	3D Reconstruction	DeepLM		✅
3D	3D Reconstruction	eppmvsnet		✅
3D	3D Object Detection	pointpillars	✅	✅
Audio	Speech Recognition	ctcmodel	✅
Audio	Speech Recognition	deepspeech2		✅
Audio	Keyword Spotting	dscnn	✅	✅
Audio	Speech Synthesis	FastSpeech		✅
Audio	Audio Tagging	fcn-4	✅	✅
Audio	Speech Recognition	jasper	✅	✅
Audio	Speech Synthesis	wavenet	✅	✅
Graph Neural Network	Graph Classification	dgcn	✅
Graph Neural Network	Text Classification	hypertext	✅	✅
Graph Neural Network	Graph Classification	sdne	✅
Graph Neural Network	Social and Information Networks	sgcn	✅	✅
Graph Neural Network	Text Classification	textrcnn	✅	✅
High Performance Computing	High Performance Computing	deepbsde		✅
High Performance Computing	High Performance Computing	molecular_dynamics	✅
High Performance Computing	High Performance Computing	ocean_model		✅
High Performance Computing	High Performance Computing	pafnucy	✅	✅
High Performance Computing	High Performance Computing	pfnn		✅
High Performance Computing	High Performance Computing	pinns		✅
Image	Image Classification	3D_DenseNet	✅	✅
Image	Semantic Segmentation	3dcnn	✅	✅
Image	Semantic Segmentation	adelaide_ea	✅
Image	Scene Text Detection	advanced_east	✅	✅
Image	Style Transfer	aecrnet	✅	✅
Image	Re-Identification	AlignedReID		✅
Image	Re-Identification	AlignedReID++	✅	✅
Image	Pose Estimation	AlphaPose	✅
Image	Style Transfer	APDrawingGAN	✅	✅
Image	Style Transfer	ArbitraryStyleTransfer	✅	✅
Image	Object Detection	arcface	✅	✅
Image	Keypoint Detection	ArtTrack		✅
Image	Style Transfer	AttGAN	✅	✅
Image	Image Classification	augvit		✅
Image	Image Classification	autoaugment	✅	✅
Image	Semantic Segmentation	Auto-DeepLab	✅
Image	Neural Architecture Search	AutoSlim	✅	✅
Image	Image Classification	AVA_cifar	✅	✅
Image	Image Classification	AVA_hpa	✅	✅
Image	Image Classification	cait	✅	✅
Image	Object Detection	CascadeRCNN	✅	✅
Image	Image Classification	CBAM	✅
Image	Image Classification	cct	✅	✅
Image	Keypoint Detection	centernet	✅		✅
Image	Keypoint Detection	centernet_det	✅
Image	Keypoint Detection	centernet_resnet101	✅	✅
Image	Keypoint Detection	centernet_resnet50_v1	✅
Image	Image Generation	CGAN	✅	✅
Image	Image Classification	convnext	✅	✅
Image	Image Super Resolution	csd	✅	✅
Image	Image Generation	CTSDG		✅
Image	Style Transfer	CycleGAN	✅	✅
Image	Image Super Resolution	DBPN	✅
Image	Image Super Resolution	DBPN_GAN	✅
Image	Image Generation	dcgan	✅	✅
Image	Re-Identification	DDAG	✅	✅
Image	Semantic Segmentation	DDM	✅
Image	Semantic Segmentation	DDRNet	✅	✅
Image	Object Detection	DeepID	✅	✅
Image	Semantic Segmentation	deeplabv3plus	✅	✅
Image	Image Retrieval	delf	✅
Image	Zero-Shot Learning	dem	✅	✅
Image	Object Detection	detr	✅	✅
Image	Semantic Segmentation	dgcnet_res101		✅
Image	Instance Segmentation	dlinknet	✅
Image	Image Denoise	DnCNN	✅
Image	Image Classification	dnet_nas	✅
Image	Image Classification	DRNet	✅	✅
Image	Image Super Resolution	EDSR	✅
Image	Object Detection	EfficientDet_d0	✅
Image	Image Classification	efficientnet-b0	✅
Image	Image Classification	efficientnet-b1	✅
Image	Image Classification	efficientnet-b2	✅	✅
Image	Image Classification	efficientnet-b3	✅	✅
Image	Image Classification	efficientnetv2	✅
Image	Salient Object Detection	EGnet	✅	✅
Image	Semantic Segmentation	E-NET	✅	✅
Image	Image Super Resolution	esr_ea	✅	✅
Image	Image Super Resolution	ESRGAN	✅	✅
Image	Image Classification	FaceAttribute	✅	✅
Image	Object Detection	faceboxes	✅
Image	Object Detection	FaceDetection	✅	✅
Image	Face Recognition	FaceNet	✅	✅
Image	Image Classification	FaceQualityAssessment	✅	✅	✅
Image	Object Detection	FaceRecognition	✅	✅
Image	Object Detection	FaceRecognitionForTracking	✅	✅	✅
Image	Object Detection	faster_rcnn_dcn	✅	✅
Image	Image Matting	FCANet	✅
Image	Image Classification	FDA-BNN	✅	✅
Image	Image Classification	fishnet99	✅	✅
Image	Optical Flow Estimation	flownet2	✅
Image	Image Generation	gan	✅	✅
Image	Image Classification	GENet_Res50	✅
Image	Image Classification	ghostnet	✅
Image	Image Classification	ghostnet_d	✅	✅
Image	Image Classification	glore_res200	✅	✅
Image	Image Classification	glore_res50	✅	✅
Image	Image Classification	hardnet	✅	✅
Image	Edge Detection	hed	✅	✅
Image	Image Generation	HiFaceGAN		✅
Image	Image Classification	HourNAS		✅
Image	Image Classification	HRNetW48_cls	✅	✅
Image	Semantic Segmentation	HRNetW48_seg	✅
Image	Image Classification	ibnnet	✅	✅
Image	Semantic Segmentation	ICNet	✅
Image	Image Classification	inception_resnet_v2	✅	✅
Image	Image Classification	Inceptionv2	✅	✅
Image	Image Matting	IndexNet		✅
Image	Image Generation	IPT	✅
Image	Image Super Resolution	IRN	✅	✅
Image	Image Classification	ISyNet	✅	✅
Image	Image Classification	ivpf		✅
Image	Image Denoise	LearningToSeeInTheDark	✅
Image	Meta Learning	LEO	✅	✅
Image	Object Detection	LightCNN	✅	✅	✅
Image	Image Super Resolution	lite-hrnet		✅
Image	Image Classification	lresnet100e_ir		✅
Image	Object Detection	m2det		✅
Image	Autoencoder	mae	✅	✅
Image	Meta Learning	MAML	✅	✅
Image	Scene Text Recognition	ManiDP		✅
Image	Face Recognition	MaskedFaceRecognition	✅
Image	Meta Learning	meta-baseline	✅	✅
Image	Re-Identification	MGN	✅	✅
Image	Depth Estimation	midas	✅	✅
Image	Image Denoise	MIMO-UNet		✅
Image	Image Classification	mnasnet	✅	✅
Image	Image Classification	mobilenetv3_large	✅		✅
Image	Image Classification	mobilenetV3_small_x1_0	✅	✅	✅
Image	Image Classification	MultiTaskNet	✅	✅
Image	Re-Identification	MVD	✅	✅
Image	Object Detection	nas-fpn	✅
Image	Image Denoise	Neighbor2Neighbor	✅	✅
Image	Image Classification	NFNet	✅	✅
Image	Image Quality Assessment	nima_vgg16		✅
Image	Semantic Segmentation	nnUNet	✅	✅
Image	Image Classification	ntsnet	✅	✅
Image	Semantic Segmentation	OCRNet	✅	✅
Image	Re-Identification	osnet	✅	✅
Image	Salient Object Detection	PAGENet	✅	✅
Image	Image Retrieval	pcb		✅
Image	Image Retrieval	pcb		✅
Image	Image Retrieval	pcb_rpp		✅
Image	Image Classification	PDarts	✅	✅
Image	Image Generation	PGAN	✅	✅
Image	Image Generation	Pix2Pix	✅	✅
Image	Image Super Resolution	Pix2PixHD	✅
Image	Image Classification	pnasnet	✅	✅
Image	Point Cloud Model	pointnet	✅	✅
Image	Point Cloud Model	pointnet2	✅	✅
Image	Image Classification	PoseEstNet	✅	✅
Image	Image Classification	ProtoNet	✅	✅
Image	Image Classification	proxylessnas	✅	✅
Image	Semantic Segmentation	PSPNet	✅
Image	Salient Object Detection	ras	✅	✅
Image	Image Super Resolution	RCAN	✅
Image	Object Detection	rcnn	✅	✅
Image	Image Super Resolution	REDNet30	✅	✅
Image	Object Detection	RefineDet	✅	✅
Image	Semantic Segmentation	RefineNet	✅	✅
Image	Re-Identification	ReIDStrongBaseline	✅	✅
Image	Image Classification	relationnet	✅	✅
Image	Image Classification	renas	✅	✅	✅
Image	Semantic Segmentation	repvgg	✅	✅
Image	Semantic Segmentation	res2net_deeplabv3	✅		✅
Image	Object Detection	res2net_faster_rcnn	✅	✅
Image	Object Detection	res2net_yolov3	✅	✅
Image	Image Classification	res2net101	✅	✅
Image	Image Classification	res2net152	✅	✅
Image	Image Classification	res2net50	✅	✅
Image	Image Classification	ResNeSt50	✅	✅
Image	Image Classification	resnet50_adv_pruning	✅	✅
Image	Image Classification	resnet50_bam	✅	✅
Image	Image Classification	ResNet50-Quadruplet	✅	✅
Image	Image Classification	ResNet50-Triplet	✅	✅
Image	Image Classification	ResnetV2_101	✅	✅
Image	Image Classification	ResnetV2_152	✅	✅
Image	Image Classification	ResnetV2_50	✅	✅
Image	Image Classification	resnetv2_50_frn	✅	✅
Image	Image Classification	resnext152_64x4d	✅	✅
Image	Object Detection	retinaface_mobilenet0.25	✅	✅
Image	Object Detection	retinanet_resnet101	✅	✅
Image	Object Detection	retinanet_resnet152	✅	✅
Image	Object Detection	rfcn		✅
Image	Image Classification	SE_ResNeXt50	✅
Image	Image Classification	senet_resnet101	✅	✅	✅
Image	Image Classification	senet_resnet50	✅	✅	✅
Image	Image Classification	se-res2net50	✅	✅
Image	Image Classification	S-GhostNet	✅
Image	Pose Estimation	simple_baselines	✅	✅
Image	Image Generation	SinGAN	✅
Image	Image Classification	single_path_nas	✅	✅
Image	Image Classification	sknet	✅	✅	✅
Image	Image Classification	snn_mlp		✅
Image	Object Detection	Spnas	✅
Image	Image Classification	SPPNet	✅	✅
Image	Image Classification	squeezenet	✅	✅
Image	Image Super Resolution	sr_ea	✅
Image	Image Super Resolution	SRGAN	✅	✅
Image	Image Classification	ssc_resnet50	✅	✅
Image	Object Detection	ssd_ghostnet	✅	✅	✅
Image	Object Detection	ssd_inception_v2		✅	✅
Image	Object Detection	ssd_inceptionv2	✅
Image	Object Detection	ssd_mobilenetV2	✅	✅	✅
Image	Object Detection	ssd_mobilenetV2_FPNlite	✅	✅	✅
Image	Object Detection	ssd_resnet_34		✅
Image	Object Detection	ssd_resnet34	✅		✅
Image	Object Detection	ssd_resnet50	✅
Image	Pose Estimation	StackedHourglass	✅
Image	Image Generation	StarGAN	✅	✅
Image	Image Generation	STGAN	✅	✅
Image	Traffic Prediction	stgcn	✅	✅
Image	Image Classification	stpm	✅	✅
Image	Image Classification	swin_transformer	✅	✅
Image	Temporal Localization	tall	✅
Image	Image Classification	TCN	✅	✅
Image	Scene Text Detection	textfusenet	✅
Image	Traffic Prediction	tgcn	✅	✅
Image	Image Classification	tinynet		✅
Image	Image Classification	TNT	✅	✅
Image	Object Detection	u2net	✅	✅
Image	Image Generation	U-GAT-IT	✅	✅
Image	Semantic Segmentation	UNet3+	✅	✅
Image	Re-Identification	VehicleNet	✅
Image	Image Classification	vgg19	✅	✅
Image	Image Classification	ViG	✅	✅
Image	Image Classification	vit_cifar	✅	✅
Image	Semantic Segmentation	vnet	✅	✅
Image	Image Classification	wave_mlp	✅	✅
Image	Image Super Resolution	wdsr	✅	✅
Image	Image Generation	wgan	✅
Image	Image Classification	wideresnet	✅	✅
Image	Instance Segmentation	Yolact++	✅
Image	Object Detection	yolov3_tiny	✅	✅
Image	Object Detection	yolox	✅
Multi Modal	Multi Modal	opt	✅	✅
Multi Modal	Multi Modal	TokenFusion	✅	✅
Multi Modal	Multi Modal	wukong	✅
Recommendation	Click-Through Rate Prediction	autodis	✅	✅
Recommendation	Click-Through Rate Prediction	DIEN	✅	✅
Recommendation	Click-Through Rate Prediction	dlrm	✅	✅
Recommendation	Click-Through Rate Prediction	EDCN	✅	✅
Recommendation	Click-Through Rate Prediction	Fat-DeepFFM	✅	✅
Recommendation	Click-Through Rate Prediction	mmoe	✅	✅
Text	Natural Language Understanding	albert	✅	✅
Text	Emotion Classification	atae_lstm	✅	✅
Text	Dialogue	dam	✅
Text	Language Model	gpt2	✅
Text	Knowledge Graph Embedding	hake		✅
Text	Natural Language Understanding	ktnet	✅	✅
Text	Named Entity Recognition	lstm_crf	✅
Text	Natural Language Understanding	luke	✅	✅
Text	Knowledge Graph Embedding	rotate	✅	✅
Text	Emotion Classification	senta	✅	✅
Text	Machine Translation	seq2seq	✅
Text	Word Embedding	skipgram	✅	✅
Text	Machine Translation	speech_transformer	✅
Text	Pre Training	ternarybert	✅	✅
Text	Natural Language Understanding	tprr	✅
Text	Natural Language Understanding	transformer_xl	✅	✅
Text	Knowledge Graph Embedding	transX		✅
Video	Video Classification	AttentionCluster	✅	✅
Video	Others	DYR	✅
Video	Video Classification	ecolite	✅
Video	Object Tracking	fairmot	✅	✅
Video	Video Classification	I3D	✅
Video	Object Tracking	JDE		✅
Video	video Segment	OSVOS		✅
Video	Video Classification	r2plus1d	✅	✅
Video	video Super Resolution	rbpn	✅
Video	Video Classification	resnet3d	✅
Video	Object Tracking	SiamFC	✅
Video	Object Tracking	siamRPN	✅	✅
Video	Video Classification	slowfast	✅	✅
Video	Video Classification	stnet	✅
Video	Object Tracking	tracktor		✅
Video	Object Tracking	tracktor++	✅	✅
Video	Video Classification	trn		✅
Video	Video Classification	tsm	✅	✅
Video	Video Classification	tsn	✅	✅

• Community

Announcements

2021.9.15 Set up repository models

models comes from the directory model_zoo of repository mindspore. This new repository doesn't contain any history of commits about the directory model_zoo in mindspore, you could refer to the repository mindspore for the past commits.

Related Website

Here is the ModelZoo for MindSpore which support different devices including Ascend, GPU, CPU and mobile.

If you are looking for exclusive models only for Ascend using different ML platform, you could refer to Ascend ModelZoo and corresponding gitee repository

If you are looking for some pretrained checkpoint of mindspore, you could refer to MindSpore Hub or Download Website.

Disclaimers

Mindspore only provides scripts that downloads and preprocesses public datasets. We do not own these datasets and are not responsible for their quality or maintenance. Please make sure you have permission to use the dataset under the dataset’s license. The models trained on these dataset are for non-commercial research and educational purpose only.

To dataset owners: we will remove or update all public content upon request if you don’t want your dataset included on Mindspore, or wish to update it in any way. Please contact us through a Github/Gitee issue. Your understanding and contribution to this community is greatly appreciated.

MindSpore is Apache 2.0 licensed. Please see the LICENSE file.

License

Apache License 2.0

FAQ

For more information about MindSpore framework, please refer to FAQ

• Q: How to resolve the lack of memory while using the model directly under "models" with errors such as Failed to alloc memory pool memory?

  A: The typical reason for insufficient memory when directly using models under "models" is due to differences in operating mode (PYNATIVE_MODE), operating environment configuration, and license control (AI-TOKEN).

  • PYNATIVE_MODE usually uses more memory than GRAPH_MODE , especially in the training graph that needs back propagation calculation, there are two ways to try to solve this problem.
    Method 1: You can try to use some smaller batch size;
    Method 2: Add context.set_context(mempool_block_size="XXGB"), where the current maximum effective value of "XX" can be set to "31".
    If method 1 and method 2 are used in combination, the effect will be better.
  • The operating environment will also cause similar problems due to the different configurations of NPU cores, memory, etc.;
  • Different gears of License control (AI-TOKEN ) will cause different memory overhead during execution. You can also try to use some smaller batch sizes.

• Q: How to resolve the error about the interface are not supported in some network operations, such as cann not import?

  A: Please check the version of MindSpore and the branch you fetch the modelzoo scripts. Some model scripits in latest branch will use new interface in the latest version of MindSpore.

• Q: What is Some RANK_TBAL_FILE which mentioned in many models?

  A: RANK_TABLE_FILE is the config file of cluster on Ascend while running distributed training. For more information, you could refer to the generator hccl_tools and Parallel Distributed Training Example

• Q: How to run the scripts on Windows system?

  A: Most the start-up scripts are written in bash, but we usually can't run bash directly on Windows. You can try start python directly without bash scripts. If you really need the start-up bash scripts, we suggest you the following method to get a bash environment on Windows:

  1. Use a virtual system or docker container with linux system. Then run the scripts in the virtual system or container.
  2. Use WSL, you could turn on the Windows Subsystem for Linux on Windows to obtain an linux system which could run the bash scripts.
  3. Use some bash tools on Windows, such as cygwin and git bash.

• Q: How to resolve the compile error point to gflags when infer on ascend310 with errors such as undefined reference to 'google::FlagRegisterer::FlagRegisterer'?

  A: Please check the version of GCC and gflags. You can refer to GCC and gflags to install GCC and gflags. You need to ensure that the components used are ABI compatible, for more information, please refer to _GLIBCXX_USE_CXX11_ABI.

• Q: How to solve the error when loading dataset in mindrecord format on Mac system, such as Invalid file, failed to open files for reading mindrecord files.?

  A: Please check the system limit with ulimit -a, if the number of file descriptors is 256 (default), you need to use ulimit -n 1024 to set it to 1024 (or larger). Then check whether the file is damaged or modified.

• Q: What should I do if I can't reach the accuracy while training with several servers instead of a single server?

  A: Most of the models has only been trained on single server with at most 8 pcs. Because the batch_size used in MindSpore only represent the batch size of single GPU/NPU, the global_batch_size will increase while training with multi-server. Different gloabl_batch_size requires different hyper parameter including learning_rate, etc. So you have to optimize these hyperparameters will training with multi-servers.