Accueil Explorer Aide
S'inscrire Connexion
zhengchun
/
PaddleX
1
0
Fork 0
Fichiers Tickets 0 Pull Requests 0 Wiki
Aborescence: 41db77eb26
Branches Tags
PaddleX
/
docs
/
support_list
/
model_list_npu.en.md

model_list_npu.en.md 82 KB
Historique Raw

comments: true

PaddleX Model List (Huawei Ascend NPU)

PaddleX incorporates multiple pipelines, each containing several modules, and each module encompasses various models. You can select the appropriate models based on the benchmark data below. If you prioritize model accuracy, choose models with higher accuracy. If you prioritize model size, select models with smaller storage requirements.

Image Classification Module

Model Name Top-1 Accuracy (%) Model Size (M) Model Download Link
CLIP_vit_base_patch16_224 85.36 306.5 M Inference Model/Training Model
CLIP_vit_large_patch14_224 88.1 1.04 G Inference Model/Training Model
ConvNeXt_base_224 83.84 313.9 M Inference Model/Training Model
ConvNeXt_base_384 84.90 313.9 M Inference Model/Training Model
ConvNeXt_large_224 84.26 700.7 M Inference Model/Training Model
ConvNeXt_large_384 85.27 700.7 M Inference Model/Training Model
ConvNeXt_small 83.13 178.0 M Inference Model/Training Model
ConvNeXt_tiny 82.03 101.4 M Inference Model/Training Model
MobileNetV1_x0_5 63.5 4.8 M Inference Model/Training Model
MobileNetV1_x0_25 51.4 1.8 M Inference Model/Training Model
MobileNetV1_x0_75 68.8 9.3 M Inference Model/Training Model
MobileNetV1_x1_0 71.0 15.2 M Inference Model/Training Model
MobileNetV2_x0_5 65.0 7.1 M Inference Model/Training Model
MobileNetV2_x0_25 53.2 5.5 M Inference Model/Training Model
MobileNetV2_x1_0 72.2 12.6 M Inference Model/Training Model
MobileNetV2_x1_5 74.1 25.0 M Inference Model/Training Model
MobileNetV2_x2_0 75.2 41.2 M Inference Model/Training Model
MobileNetV3_large_x0_5 69.2 9.6 M Inference Model/Training Model
MobileNetV3_large_x0_35 64.3 7.5 M Inference Model/Training Model
MobileNetV3_large_x0_75 73.1 14.0 M Inference Model/Training Model
MobileNetV3_large_x1_0 75.3 19.5 M Inference Model/Training Model
MobileNetV3_large_x1_25 76.4 26.5 M Inference Model/Training Model
MobileNetV3_small_x0_5 59.2 6.8 M Inference Model/Training Model
MobileNetV3_small_x0_35 53.0 6.0 M Inference Model/Training Model
MobileNetV3_small_x0_75 66.0 8.5 M Inference Model/Training Model
MobileNetV3_small_x1_0 68.2 10.5 M Inference Model/Training Model
MobileNetV3_small_x1_25 70.7 13.0 M Inference Model/Training Model
MobileNetV4_conv_large 83.4 125.2 M Inference Model/Training Model
MobileNetV4_conv_medium 79.9 37.6 M Inference Model/Training Model
MobileNetV4_conv_small 74.6 14.7 M Inference Model/Training Model
MobileNetV4_hybrid_large 83.8 145.1 M Inference Model/Training Model
MobileNetV4_hybrid_medium 80.5 42.9 M Inference Model/Training Model
PP-HGNet_base 85.0 249.4 M Inference Model/Training Model
PP-HGNet_small 81.51 86.5 M Inference Model/Training Model
PP-HGNet_tiny 79.83 52.4 M Inference Model/Training Model
PP-HGNetV2-B0 77.77 21.4 M Inference Model/Training Model
PP-HGNetV2-B1 79.18 22.6 M Inference Model/Training Model
PP-HGNetV2-B2 81.74 39.9 M Inference Model/Training Model
PP-HGNetV2-B3 82.98 57.9 M Inference Model/Training Model
PP-HGNetV2-B4 83.57 70.4 M Inference Model/Training Model
PP-HGNetV2-B5 84.75 140.8 M Inference Model/Training Model
PP-HGNetV2-B6 86.30 268.4 M Inference Model/Training Model
PP-LCNet_x0_5 63.14 6.7 M Inference Model/Training Model
PP-LCNet_x0_25 51.86 5.5 M Inference Model/Training Model
PP-LCNet_x0_35 58.09 5.9 M Inference Model/Training Model
PP-LCNet_x0_75 68.18 8.4 M Inference Model/Training Model
PP-LCNet_x1_0 71.32 10.5 M Inference Model/Training Model
PP-LCNet_x1_5 73.71 16.0 M Inference Model/Training Model
PP-LCNet_x2_0 75.18 23.2 M Inference Model/Training Model
PP-LCNet_x2_5 76.60 32.1 M Inference Model/Training Model
PP-LCNetV2_base 77.05 23.7 M Inference Model/Training Model
PP-LCNetV2_large 78.51 37.3 M Inference Model/Training Model
PP-LCNetV2_small 73.97 14.6 M Inference Model/Training Model
ResNet18_vd 72.3 41.5 M Inference Model/Training Model
ResNet18 71.0 41.5 M Inference Model/Training Model
ResNet34_vd 76.0 77.3 M Inference Model/Training Model
ResNet34 74.6 77.3 M Inference Model/Training Model
ResNet50_vd 79.1 90.8 M Inference Model/Training Model
ResNet50 76.5 90.8 M Inference Model/Training Model
ResNet101_vd 80.2 158.4 M Inference Model/Training Model
ResNet101 77.6 158.7 M Inference Model/Training Model
ResNet152_vd 80.6 214.3 M Inference Model/Training Model
ResNet152 78.3 214.2 M Inference Model/Training Model
ResNet200_vd 80.9 266.0 M Inference Model/Training Model
SwinTransformer_base_patch4_window7_224 83.37 310.5 M Inference Model/Training Model
SwinTransformer_base_patch4_window12_384 84.17 311.4 M Inference Model/Training Model
SwinTransformer_large_patch4_window7_224 86.19 694.8 M Inference Model/Training Model
SwinTransformer_large_patch4_window12_384 87.06 696.1 M Inference Model/Training Model
SwinTransformer_small_patch4_window7_224 83.21 175.6 M Inference Model/Training Model
SwinTransformer_tiny_patch4_window7_224 81.10 100.1 M Inference Model/Training Model
StarNet-S1 73.6 11.2 M Inference Model/Training Model
StarNet-S2 74.8 14.3 M Inference Model/Training Model
StarNet-S3 77.0 22.2 M Inference Model/Training Model
StarNet-S4 79.0 28.9 M Inference Model/Training Model
FasterNet-L 83.5 357.1 M Inference Model/Training Model
FasterNet-M 83.0 204.6 M Inference Model/Training Model
FasterNet-S 81.3 119.3 M Inference Model/Training Model
FasterNet-T0 71.9 15.1 M Inference Model/Training Model
FasterNet-T1 75.9 29.2 M Inference Model/Training Model
FasterNet-T2 79.1 57.4 M Inference Model/Training Model
Note: The above accuracy metrics refer to Top-1 Accuracy on the ImageNet-1k validation set.

Image Multi-label Classification Module

Note: The above accuracy metrics are for the multi-label classification task mAP of COCO2017.
Model Name mAP (%) Model Storage Size Model Download Link
CLIP_vit_base_patch16_448_ML 89.15 325.6 M Inference Model/Training Model
PP-HGNetV2-B0_ML 80.98 39.6 M Inference Model/Training Model
PP-HGNetV2-B4_ML 87.96 88.5 M Inference Model/Training Model
PP-HGNetV2-B6_ML 91.06 286.5 M Inference Model/Training Model
PP-LCNet_x1_0_ML 77.96 29.4 M Inference Model/Training Model
ResNet50_ML 83.42 108.9 M Inference Model/Training Model

Pedestrian Attribute Module

Model Name mA (%) Model Size Model Download Link
PP-LCNet_x1_0_pedestrian_attribute 92.2 6.7 M Inference Model/Training Model
Note: The above accuracy metrics are mA on PaddleX's internal self-built dataset.

Vehicle Attribute Module

Model Name mA (%) Model Size Model Download Link
PP-LCNet_x1_0_vehicle_attribute 91.7 6.7 M Inference Model/Training Model
Note: The above accuracy metrics are mA on the VeRi dataset.

Object Detection Module

Model Name mAP (%) Model Size (M) Model Download Link
Cascade-FasterRCNN-ResNet50-FPN 41.1 245.4 M Inference Model/Training Model
Cascade-FasterRCNN-ResNet50-vd-SSLDv2-FPN 45.0 246.2 M Inference Model/Training Model
CenterNet-DLA-34 37.6 75.4 M Inference Model/Training Model
CenterNet-ResNet50 38.9 319.7 M Inference Model/Training Model
DETR-R50 42.3 159.3 M Inference Model/Training Model
FasterRCNN-ResNet34-FPN 37.8 137.5 M Inference Model/Training Model
FasterRCNN-ResNet50 36.7 120.2 M Inference Model/Training Model
FasterRCNN-ResNet50-FPN 38.4 148.1 M Inference Model/Training Model
FasterRCNN-ResNet50-vd-FPN 39.5 148.1 M Inference Model/Training Model
FasterRCNN-ResNet50-vd-SSLDv2-FPN 41.4 148.1 M Inference Model/Training Model
FasterRCNN-ResNet101 39.0 188.1 M Inference Model/Training Model
FasterRCNN-ResNet101-FPN 41.4 216.3 M Inference Model/Training Model
FasterRCNN-ResNeXt101-vd-FPN 43.4 360.6 M Inference Model/Training Model
FasterRCNN-Swin-Tiny-FPN 42.6 159.8 M Inference Model/Training Model
FCOS-ResNet50 39.6 124.2 M Inference Model/Training Model
PicoDet-L 42.6 20.9 M Inference Model/Training Model
PicoDet-M 37.5 16.8 M Inference Model/Training Model
PicoDet-S 29.1 4.4 M Inference Model/Training Model
PicoDet-XS 26.2 5.7M Inference Model/Training Model
PP-YOLOE_plus-L 52.9 185.3 M Inference Model/Training Model
PP-YOLOE_plus-M 49.8 83.2 M Inference Model/Training Model
PP-YOLOE_plus-S 43.7 28.3 M Inference Model/Training Model
PP-YOLOE_plus-X 54.7 349.4 M Inference Model/Training Model
RT-DETR-H 56.3 435.8 M Inference Model/Training Model
RT-DETR-L 53.0 113.7 M Inference Model/Training Model
RT-DETR-R18 46.5 70.7 M Inference Model/Training Model
RT-DETR-R50 53.1 149.1 M Inference Model/Training Model
RT-DETR-X 54.8 232.9 M Inference Model/Training Model
YOLOv3-DarkNet53 39.1 219.7 M Inference Model/Training Model
YOLOv3-MobileNetV3 31.4 83.8 M Inference Model/Training Model
YOLOv3-ResNet50_vd_DCN 40.6 163.0 M Inference Model/Training Model
Note: The above accuracy metrics are for COCO2017 validation set mAP(0.5:0.95).

Small Object Detection Module

Model Name mAP（%） Model Size Model Download Link
PP-YOLOE_plus_SOD-S 25.1 77.3 M Inference Model/Training Model
PP-YOLOE_plus_SOD-L 31.9 325.0 M Inference Model/Training Model
PP-YOLOE_plus_SOD-largesize-L 42.7 340.5 M Inference Model/Training Model
YOLOX-S 40.4 32.0 M Inference Model/Training Model
YOLOX-T 32.9 18.1 M Inference Model/Training Model
YOLOX-M 46.9 90.0 M Inference Model/Training Model
YOLOX-N 26.1 3.4M 推理模型/训练模型
Note: The above accuracy metrics are for VisDrone-DET validation set mAP(0.5:0.95)。

Pedestrian Detection Module

Model Name mAP（%） Model Size Model Download Link
PP-YOLOE-L_human 48.0 196.1 M Inference Model/Training Model
PP-YOLOE-S_human 42.5 28.8 M Inference Model/Training Model
Note: The above accuracy metrics are for CrowdHuman validation set mAP(0.5:0.95)。

Semantic Segmentation Module

Model Name mIoU (%) Model Size (M) Model Download Link
Deeplabv3_Plus-R50 80.36 94.9 M Inference Model/Training Model
Deeplabv3_Plus-R101 81.10 162.5 M Inference Model/Training Model
Deeplabv3-R50 79.90 138.3 M Inference Model/Training Model
Deeplabv3-R101 80.85 205.9 M Inference Model/Training Model
OCRNet_HRNet-W18 80.67 43.1 M Inference Model/Training Model
OCRNet_HRNet-W48 82.15 249.8 M Inference Model/Training Model
PP-LiteSeg-T 73.10 28.5 M Inference Model/Training Model
SegFormer-B0 (slice) 76.73 13.2 M Inference Model/Training Model
SegFormer-B1 (slice) 78.35 48.5 M Inference Model/Training Model
SegFormer-B2 (slice) 81.60 96.9 M Inference Model/Training Model
SegFormer-B3 (slice) 82.47 167.3 M Inference Model/Training Model
SegFormer-B4 (slice) 82.38 226.7 M Inference Model/Training Model
SegFormer-B5 (slice) 82.58 229.7 M Inference Model/Training Model
Note: The above accuracy metrics are for Cityscapes dataset mIoU.
Model Name mIoU (%) Model Size Model Download Link
SeaFormer_base(slice) 40.92 30.8 M Inference Model/Training Model
SeaFormer_large (slice) 43.66 49.8 M Inference Model/Training Model
SeaFormer_small (slice) 38.73 14.3 M Inference Model/Training Model
SeaFormer_tiny (slice) 34.58 6.1 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on the ADE20k dataset. "slice" indicates that the input image has been cropped.

Instance Segmentation Module

Model Name Mask AP Model Size (M) Model Download Link
Mask-RT-DETR-H 50.6 449.9 M Inference Model/Training Model
Mask-RT-DETR-L 45.7 113.6 M Inference Model/Training Model
Mask-RT-DETR-M 42.7 66.6 M Inference Model/Training Model
Mask-RT-DETR-S 41.0 51.8 M Inference Model/Training Model
Mask-RT-DETR-X 47.5 237.5 M Inference Model/Training Model
Cascade-MaskRCNN-ResNet50-FPN 36.3 254.8 M Inference Model/Training Model
Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN 39.1 254.7 M Inference Model/Training Model
MaskRCNN-ResNet50-FPN 35.6 157.5 M Inference Model/Training Model
MaskRCNN-ResNet50-vd-FPN 36.4 157.5 M Inference Model/Training Model
MaskRCNN-ResNet50 32.8 127.8 M Inference Model/Training Model
MaskRCNN-ResNet101-FPN 36.6 225.4 M Inference Model/Training Model
MaskRCNN-ResNet101-vd-FPN 38.1 225.1 M Inference Model/Training Model
MaskRCNN-ResNeXt101-vd-FPN 39.5 370.0 M Inference Model/Training Model
PP-YOLOE_seg-S 32.5 31.5 M Inference Model/Training Model
Note: The above accuracy metrics are for COCO2017 validation set Mask AP(0.5:0.95).

Image Feature Module

Model Name recall@1（%） Model Size Model Download Link
PP-ShiTuV2_rec 84.2 16.3 M Inference Model/Training Model
PP-ShiTuV2_rec_CLIP_vit_base 88.69 306.6 M Inference Model/Training Model
PP-ShiTuV2_rec_CLIP_vit_large 91.03 1.05 G Inference Model/Training Model
Note: The above accuracy metrics are for AliProducts recall@1。

Main Body Detection Module

Model Name mAP（%） Model Size Model Download Link
PP-ShiTuV2_det 41.5 27.6 M Inference Model/Training Model
Note: The above accuracy metrics are for PaddleClas主体检测数据集 mAP(0.5:0.95)。

Vehicle Detection Module

Model Name mAP（%） Model Size Model Download Link
PP-YOLOE-L_vehicle 63.9 196.1 M Inference Model/Training Model
PP-YOLOE-S_vehicle 61.3 28.8 M Inference Model/Training Model
Note: The above accuracy metrics are for PPVehicle validation set mAP(0.5:0.95)。

Face Detection Module

Model Name AP (%)
Easy/Medium/Hard		 Model Size Model Download Link
PicoDet_LCNet_x2_5_face 93.7/90.7/68.1 28.9 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on the WIDER-FACE validation set with an input size of 640*640.

Abnormality Detection Module

Model Name Avg（%） Model Size Model Download Link
STFPM 96.2 21.5 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on the MVTec AD dataset using the average anomaly score.

Text Detection Module

Model Name Detection Hmean (%) Model Size (M) Model Download Link
PP-OCRv4_mobile_det 77.79 4.2 M Inference Model/Training Model
PP-OCRv4_server_det 82.69 100.1 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, and handwritten scenarios, with 500 images for detection.

Text Recognition Module

Model Name Recognition Avg Accuracy (%) Model Size (M) Model Download Link
PP-OCRv4_mobile_rec 78.20 10.6 M Inference Model/Training Model
PP-OCRv4_server_rec 79.20 71.2 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, and handwritten scenarios, with 11,000 images for text recognition.
Model Name Recognition Avg Accuracy (%) Model Size (M) Model Download Link
ch_SVTRv2_rec 68.81 73.9 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition A-Rank.
Model Name Recognition Avg Accuracy (%) Model Size (M) Model Download Link
ch_RepSVTR_rec 65.07 22.1 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition B-Rank.

Table Structure Recognition Module

Model Name Accuracy (%) Model Size (M) Model Download Link
SLANet 76.31 6.9 M Inference Model/Training Model
SLANet_plus 63.69 6.9 M Inference Model/Training Model
Note: The above accuracy metrics are measured on the PubtabNet English table recognition dataset.

Image Rectification Module

Model Name MS-SSIM (%) Model Size Model Download Link
UVDoc 54.40 30.3 M Inference Model/Training Model
Note: The above accuracy metrics are measured on a self-built image rectification dataset by PaddleX.

Seal Text Detection Module

Model Name Detection Hmean (%) Model Size Model Download Link
PP-OCRv4_mobile_seal_det 96.47 4.7 M Inference Model/Training Model
PP-OCRv4_server_seal_det 98.21 108.3 M Inference Model/Training Model
Note: The above accuracy metrics are evaluated on a self-built seal dataset by PaddleX, containing 500 seal images.

Document Orientation Classification Module

Model Name Top-1 Acc (%) Model Size Model Download Link
PP-LCNet_x1_0_doc_ori 99.26 7.1 M Inference Model/Training Model
Note: The above accuracy metrics are Top-1 Acc on PaddleX's internal self-built dataset.

Layout Detection Module

Model Name mAP (%) Model Size (M) Model Download Link
PicoDet_layout_1x 86.8 7.4M Inference Model/Training Model
PicoDet-L_layout_3cls 89.3 22.6 M Inference Model/Training Model
RT-DETR-H_layout_3cls 95.9 470.1 M Inference Model/Training Model
RT-DETR-H_layout_17cls 92.6 470.2 M Inference Model/Training Model
Note: The evaluation set for the above accuracy metrics is PaddleOCR's self-built layout analysis dataset, containing 10,000 images.

Time Series Forecasting Module

Model Name MSE MAE Model Size (M) Model Download Link
DLinear 0.382 0.394 72K Inference Model/Training Model
NLinear 0.386 0.392 40K Inference Model/Training Model
Nonstationary 0.600 0.515 55.5 M Inference Model/Training Model
PatchTST 0.385 0.397 2.0M Inference Model/Training Model
RLinear 0.384 0.392 40K Inference Model/Training Model
TiDE 0.405 0.412 31.7M Inference Model/Training Model
TimesNet 0.417 0.431 4.9M Inference Model/Training Model
Note: The above accuracy metrics are measured on the ETTH1 dataset (evaluation results on the test set test.csv).

Time Series Anomaly Detection Module

Model Name Precision Recall F1-Score Model Size (M) Model Download Link
AutoEncoder_ad 99.36 84.36 91.25 52K Inference Model/Training Model
DLinear_ad 98.98 93.96 96.41 112K Inference Model/Training Model
Nonstationary_ad 98.55 88.95 93.51 1.8M Inference Model/Training Model
PatchTST_ad 98.78 90.70 94.57 320K Inference Model/Training Model
TimesNet_ad 98.37 94.80 96.56 1.3M Inference Model/Training Model
Note: The above accuracy metrics are measured on the PSM dataset.

Time Series Classification Module

Model Name Acc (%) Model Size (M) Model Download Link
TimesNet_cls 87.5 792K Inference Model/Training Model
Note: The above accuracy metrics are measured on the UWaveGestureLibrary: Training, Evaluation datasets.