PaddleX Model List (Cambricon MLU)¶
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 |
|---|---|---|---|
| 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 |
| 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-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 |
| 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 |
Note: The above accuracy metrics are Top-1 Accuracy on the ImageNet-1k validation set.
Object Detection Module¶
| Model Name | mAP (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| 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 |
Note: The above accuracy metrics are mAP(0.5:0.95) on the COCO2017 validation set.
Semantic Segmentation Module¶
| Model Name | mIoU (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| PP-LiteSeg-T | 73.10 | 28.5 M | Inference Model/Trained Model |
Note: The above accuracy metrics are based on the mIoU of the Cityscapes dataset.
Image Feature Module¶
| Model Name | recall@1（%） | Model Size | 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 |
Note: The above accuracy metrics are for AliProducts recall@1。
Abnormality Detection Module¶
| Model Name | Avg（%） | Model Size | Model Download Link |
|---|---|---|---|
| STFPM | 96.2 | 21.5 M | Inference Model/Trained Model |
Note: The above accuracy metrics are evaluated on the MVTec AD dataset using the average anomaly score.
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/Trained Model |
Note: The above accuracy metrics are evaluated on the WIDER-FACE validation set with an input size of 640*640.
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 | 100.1M | Inference Model/Trained Model |
Note: The evaluation set for the above accuracy metrics is PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, handwriting, and more 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 evaluation set for the above accuracy metrics is PaddleOCR's self-built Chinese dataset, covering street scenes, web images, documents, handwriting, and more scenarios, with 11,000 images for text recognition.
Layout Analysis Module¶
| Model Name | mAP (%) | Model Size (M) | Model Download Link |
|---|---|---|---|
| PicoDet_layout_1x | 86.8 | 7.4M | 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 |
| RLinear | 0.384 | 0.392 | 40K | Inference Model/Trained Model |
Note: The above accuracy metrics are measured on the ETTH1 dataset (evaluation results on the test set test.csv).