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/Trained Model |
| CLIP_vit_large_patch14_224 | 88.1 | 1.04 G | Inference Model/Trained Model |
| ConvNeXt_base_224 | 83.84 | 313.9 M | Inference Model/Trained Model |
| ConvNeXt_base_384 | 84.90 | 313.9 M | Inference Model/Trained Model |
| ConvNeXt_large_224 | 84.26 | 700.7 M | Inference Model/Trained Model |
| ConvNeXt_large_384 | 85.27 | 700.7 M | Inference Model/Trained Model |
| ConvNeXt_small | 83.13 | 178.0 M | Inference Model/Trained Model |
| ConvNeXt_tiny | 82.03 | 101.4 M | Inference Model/Trained Model |
| MobileNetV1_x0_5 | 63.5 | 4.8 M | Inference Model/Trained Model |
| MobileNetV1_x0_25 | 51.4 | 1.8 M | Inference Model/Trained Model |
| MobileNetV1_x0_75 | 68.8 | 9.3 M | Inference Model/Trained Model |
| MobileNetV1_x1_0 | 71.0 | 15.2 M | Inference Model/Trained Model |
| MobileNetV2_x0_5 | 65.0 | 7.1 M | Inference Model/Trained Model |
| MobileNetV2_x0_25 | 53.2 | 5.5 M | Inference Model/Trained Model |
| MobileNetV2_x1_0 | 72.2 | 12.6 M | Inference Model/Trained Model |
| MobileNetV2_x1_5 | 74.1 | 25.0 M | Inference Model/Trained Model |
| MobileNetV2_x2_0 | 75.2 | 41.2 M | Inference Model/Trained Model |
| MobileNetV3_large_x0_5 | 69.2 | 9.6 M | Inference Model/Trained Model |
| MobileNetV3_large_x0_35 | 64.3 | 7.5 M | Inference Model/Trained Model |
| MobileNetV3_large_x0_75 | 73.1 | 14.0 M | Inference Model/Trained Model |
| MobileNetV3_large_x1_0 | 75.3 | 19.5 M | Inference Model/Trained Model |
| MobileNetV3_large_x1_25 | 76.4 | 26.5 M | Inference Model/Trained Model |
| MobileNetV3_small_x0_5 | 59.2 | 6.8 M | Inference Model/Trained Model |
| MobileNetV3_small_x0_35 | 53.0 | 6.0 M | Inference Model/Trained Model |
| MobileNetV3_small_x0_75 | 66.0 | 8.5 M | Inference Model/Trained Model |
| MobileNetV3_small_x1_0 | 68.2 | 10.5 M | Inference Model/Trained Model |
| MobileNetV3_small_x1_25 | 70.7 | 13.0 M | Inference Model/Trained Model |
| MobileNetV4_conv_large | 83.4 | 125.2 M | Inference Model/Trained Model |
| MobileNetV4_conv_medium | 79.9 | 37.6 M | Inference Model/Trained Model |
| MobileNetV4_conv_small | 74.6 | 14.7 M | Inference Model/Trained Model |
| PP-HGNet_base | 85.0 | 249.4 M | Inference Model/Trained Model |
| PP-HGNet_small | 81.51 | 86.5 M | Inference Model/Trained Model |
| PP-HGNet_tiny | 79.83 | 52.4 M | Inference Model/Trained Model |
| PP-HGNetV2-B0 | 77.77 | 21.4 M | Inference Model/Trained Model |
| PP-HGNetV2-B1 | 79.18 | 22.6 M | Inference Model/Trained Model |
| PP-HGNetV2-B2 | 81.74 | 39.9 M | Inference Model/Trained Model |
| PP-HGNetV2-B3 | 82.98 | 57.9 M | Inference Model/Trained Model |
| PP-HGNetV2-B4 | 83.57 | 70.4 M | Inference Model/Trained Model |
| PP-HGNetV2-B5 | 84.75 | 140.8 M | Inference Model/Trained Model |
| PP-HGNetV2-B6 | 86.30 | 268.4 M | Inference Model/Trained Model |
| PP-LCNet_x0_5 | 63.14 | 6.7 M | Inference Model/Trained Model |
| PP-LCNet_x0_25 | 51.86 | 5.5 M | Inference Model/Trained Model |
| PP-LCNet_x0_35 | 58.09 | 5.9 M | Inference Model/Trained Model |
| PP-LCNet_x0_75 | 68.18 | 8.4 M | Inference Model/Trained Model |
| PP-LCNet_x1_0 | 71.32 | 10.5 M | Inference Model/Trained Model |
| PP-LCNet_x1_5 | 73.71 | 16.0 M | Inference Model/Trained Model |
| PP-LCNet_x2_0 | 75.18 | 23.2 M | Inference Model/Trained Model |
| PP-LCNet_x2_5 | 76.60 | 32.1 M | Inference Model/Trained Model |
| PP-LCNetV2_base | 77.05 | 23.7 M | Inference Model/Trained Model |
| PP-LCNetV2_large | 78.51 | 37.3 M | Inference Model/Trained Model |
| PP-LCNetV2_small | 73.97 | 14.6 M | Inference Model/Trained Model |
| ResNet18_vd | 72.3 | 41.5 M | Inference Model/Trained Model |
| ResNet18 | 71.0 | 41.5 M | Inference Model/Trained Model |
| ResNet34_vd | 76.0 | 77.3 M | Inference Model/Trained Model |
| ResNet34 | 74.6 | 77.3 M | Inference Model/Trained Model |
| ResNet50_vd | 79.1 | 90.8 M | Inference Model/Trained Model |
| ResNet50 | 76.5 | 90.8 M | Inference Model/Trained Model |
| ResNet101_vd | 80.2 | 158.4 M | Inference Model/Trained Model |
| ResNet101 | 77.6 | 158.7 M | Inference Model/Trained Model |
| ResNet152_vd | 80.6 | 214.3 M | Inference Model/Trained Model |
| ResNet152 | 78.3 | 214.2 M | Inference Model/Trained Model |
| ResNet200_vd | 80.9 | 266.0 M | Inference Model/Trained Model |
| SwinTransformer_base_patch4_window7_224 | 83.37 | 310.5 M | Inference Model/Trained Model |
| SwinTransformer_base_patch4_window12_384 | 84.17 | 311.4 M | Inference Model/Trained Model |
| SwinTransformer_large_patch4_window7_224 | 86.19 | 694.8 M | Inference Model/Trained Model |
| SwinTransformer_large_patch4_window12_384 | 87.06 | 696.1 M | Inference Model/Trained Model |
| SwinTransformer_small_patch4_window7_224 | 83.21 | 175.6 M | Inference Model/Trained Model |
| SwinTransformer_tiny_patch4_window7_224 | 81.10 | 100.1 M | Inference Model/Trained Model |
Note: The above accuracy metrics refer to Top-1 Accuracy on the ImageNet-1k validation set.
Image Multi-label Classification Module¶
| 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 |
Note: The above accuracy metrics are for the multi-label classification task mAP of COCO2017.
Object Detection Module¶
| Model Name | mAP (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| Cascade-FasterRCNN-ResNet50-FPN | 41.1 | 245.4 M | Inference Model/Trained Model |
| Cascade-FasterRCNN-ResNet50-vd-SSLDv2-FPN | 45.0 | 246.2 M | Inference Model/Trained Model |
| CenterNet-DLA-34 | 37.6 | 75.4 M | Inference Model/Trained Model |
| CenterNet-ResNet50 | 38.9 | 319.7 M | Inference Model/Trained Model |
| DETR-R50 | 42.3 | 159.3 M | Inference Model/Trained Model |
| FasterRCNN-ResNet34-FPN | 37.8 | 137.5 M | Inference Model/Trained Model |
| FasterRCNN-ResNet50 | 36.7 | 120.2 M | Inference Model/Trained Model |
| FasterRCNN-ResNet50-FPN | 38.4 | 148.1 M | Inference Model/Trained Model |
| FasterRCNN-ResNet50-vd-FPN | 39.5 | 148.1 M | Inference Model/Trained Model |
| FasterRCNN-ResNet50-vd-SSLDv2-FPN | 41.4 | 148.1 M | Inference Model/Trained Model |
| FasterRCNN-ResNet101 | 39.0 | 188.1 M | Inference Model/Trained Model |
| FasterRCNN-ResNet101-FPN | 41.4 | 216.3 M | Inference Model/Trained Model |
| FasterRCNN-ResNeXt101-vd-FPN | 43.4 | 360.6 M | Inference Model/Trained Model |
| FasterRCNN-Swin-Tiny-FPN | 42.6 | 159.8 M | Inference Model/Trained Model |
| FCOS-ResNet50 | 39.6 | 124.2 M | Inference Model/Trained Model |
| PicoDet-L | 42.6 | 20.9 M | Inference Model/Trained Model |
| PicoDet-M | 37.5 | 16.8 M | Inference Model/Trained Model |
| PicoDet-S | 29.1 | 4.4 M | Inference Model/Trained Model |
| PicoDet-XS | 26.2 | 5.7M | Inference Model/Trained Model |
| PP-YOLOE_plus-L | 52.9 | 185.3 M | Inference Model/Trained Model |
| PP-YOLOE_plus-M | 49.8 | 83.2 M | Inference Model/Trained Model |
| PP-YOLOE_plus-S | 43.7 | 28.3 M | Inference Model/Trained Model |
| PP-YOLOE_plus-X | 54.7 | 349.4 M | Inference Model/Trained Model |
| RT-DETR-H | 56.3 | 435.8 M | Inference Model/Trained Model |
| RT-DETR-L | 53.0 | 113.7 M | Inference Model/Trained Model |
| RT-DETR-R18 | 46.5 | 70.7 M | Inference Model/Trained Model |
| RT-DETR-R50 | 53.1 | 149.1 M | Inference Model/Trained Model |
| RT-DETR-X | 54.8 | 232.9 M | Inference Model/Trained Model |
| YOLOv3-DarkNet53 | 39.1 | 219.7 M | Inference Model/Trained Model |
| YOLOv3-MobileNetV3 | 31.4 | 83.8 M | Inference Model/Trained Model |
| YOLOv3-ResNet50_vd_DCN | 40.6 | 163.0 M | Inference Model/Trained Model |
Note: The above accuracy metrics are for COCO2017 validation set mAP(0.5:0.95).
小目标检测模块¶
| 模型名称 | mAP（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| PP-YOLOE_plus_SOD-S | 25.1 | 77.3 M | Inference Model/Trained Model |
| PP-YOLOE_plus_SOD-L | 31.9 | 325.0 M | Inference Model/Trained Model |
| PP-YOLOE_plus_SOD-largesize-L | 42.7 | 340.5 M | Inference Model/Trained Model |
注：以上精度指标为 VisDrone-DET 验证集 mAP(0.5:0.95)。
行人检测模块¶
| 模型名称 | mAP（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| PP-YOLOE-L_human | 48.0 | 196.1 M | Inference Model/Trained Model |
| PP-YOLOE-S_human | 42.5 | 28.8 M | Inference Model/Trained Model |
注：以上精度指标为 CrowdHuman 验证集 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/Trained Model |
| Deeplabv3_Plus-R101 | 81.10 | 162.5 M | Inference Model/Trained Model |
| Deeplabv3-R50 | 79.90 | 138.3 M | Inference Model/Trained Model |
| Deeplabv3-R101 | 80.85 | 205.9 M | Inference Model/Trained Model |
| OCRNet_HRNet-W48 | 82.15 | 249.8 M | Inference Model/Trained Model |
| PP-LiteSeg-T | 73.10 | 28.5 M | Inference Model/Trained Model |
Note: The above accuracy metrics are for Cityscapes dataset mIoU.
Instance Segmentation Module¶
| Model Name | Mask AP | Model Size (M) | Model Download Link |
|---|---|---|---|
| Mask-RT-DETR-H | 50.6 | 449.9 M | Inference Model/Trained Model |
| Mask-RT-DETR-L | 45.7 | 113.6 M | Inference Model/Trained Model |
| Mask-RT-DETR-M | 42.7 | 66.6 M | Inference Model/Trained Model |
| Mask-RT-DETR-S | 41.0 | 51.8 M | Inference Model/Trained Model |
| Mask-RT-DETR-X | 47.5 | 237.5 M | Inference Model/Trained Model |
| Cascade-MaskRCNN-ResNet50-FPN | 36.3 | 254.8 M | Inference Model/Trained Model |
| Cascade-MaskRCNN-ResNet50-vd-SSLDv2-FPN | 39.1 | 254.7 M | Inference Model/Trained Model |
| MaskRCNN-ResNet50-FPN | 35.6 | 157.5 M | Inference Model/Trained Model |
| MaskRCNN-ResNet50-vd-FPN | 36.4 | 157.5 M | Inference Model/Trained Model |
| MaskRCNN-ResNet50 | 32.8 | 127.8 M | Inference Model/Trained Model |
| MaskRCNN-ResNet101-FPN | 36.6 | 225.4 M | Inference Model/Trained Model |
| MaskRCNN-ResNet101-vd-FPN | 38.1 | 225.1 M | Inference Model/Trained Model |
| MaskRCNN-ResNeXt101-vd-FPN | 39.5 | 370.0 M | Inference Model/Trained Model |
| PP-YOLOE_seg-S | 32.5 | 31.5 M | Inference Model/Trained Model |
Note: The above accuracy metrics are for COCO2017 validation set Mask AP(0.5:0.95).
图像特征模块¶
| 模型名称 | recall@1（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| PP-ShiTuV2_rec_CLIP_vit_base | 88.69 | 306.6 M | Inference Model/Trained Model |
| PP-ShiTuV2_rec_CLIP_vit_large | 91.03 | 1.05 G | Inference Model/Trained Model |
注：以上精度指标为 AliProducts recall@1。
主体检测模块¶
| 模型名称 | mAP（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| PP-ShiTuV2_det | 41.5 | 27.6 M | Inference Model/Trained Model |
注：以上精度指标为 PaddleClas主体检测数据集 mAP(0.5:0.95)。
车辆检测模块¶
| 模型名称 | mAP（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| PP-YOLOE-L_vehicle | 63.9 | 196.1 M | Inference Model/Trained Model |
| PP-YOLOE-S_vehicle | 61.3 | 28.8 M | Inference Model/Trained Model |
注：以上精度指标为 PPVehicle 验证集 mAP(0.5:0.95)。
异常检测模块¶
| 模型名称 | Avg（%） | 模型存储大小 | Model Download Link |
|---|---|---|---|
| STFPM | 96.2 | 21.5 M | Inference Model/Trained Model |
注：以上精度指标为 MVTec AD 验证集 平均异常分数。
Text Detection Module¶
| Model Name | Detection Hmean (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| PP-OCRv4_mobile_det | 77.79 | 4.2 M | Inference Model/Trained Model |
| PP-OCRv4_server_det | 82.69 | 100.1 M | Inference Model/Trained 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/Trained Model |
| PP-OCRv4_server_rec | 79.20 | 71.2 M | Inference Model/Trained 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/Trained 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/Trained 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/Trained Model |
Note: The above accuracy metrics are measured on the PubtabNet English table recognition dataset.
Layout Analysis Module¶
| Model Name | mAP (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| PicoDet_layout_1x | 86.8 | 7.4M | Inference Model/Trained Model |
| PicoDet-L_layout_3cls | 89.3 | 22.6 M | Inference Model/Trained Model |
| RT-DETR-H_layout_3cls | 95.9 | 470.1 M | Inference Model/Trained Model |
| RT-DETR-H_layout_17cls | 92.6 | 470.2 M | Inference Model/Trained 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/Trained Model |
| NLinear | 0.386 | 0.392 | 40K | Inference Model/Trained Model |
| Nonstationary | 0.600 | 0.515 | 55.5 M | Inference Model/Trained Model |
| PatchTST | 0.385 | 0.397 | 2.0M | Inference Model/Trained Model |
| RLinear | 0.384 | 0.392 | 40K | Inference Model/Trained Model |
| TiDE | 0.405 | 0.412 | 31.7M | Inference Model/Trained Model |
| TimesNet | 0.417 | 0.431 | 4.9M | Inference Model/Trained 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/Trained Model |
| DLinear_ad | 98.98 | 93.96 | 96.41 | 112K | Inference Model/Trained Model |
| Nonstationary_ad | 98.55 | 88.95 | 93.51 | 1.8M | Inference Model/Trained Model |
| PatchTST_ad | 98.78 | 90.70 | 94.57 | 320K | Inference Model/Trained Model |
| TimesNet_ad | 98.37 | 94.80 | 96.56 | 1.3M | Inference Model/Trained 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/Trained Model |
Note: The above accuracy metrics are measured on the UWaveGestureLibrary: Training, Evaluation datasets.