diff --git a/docs/sirider/s1/app-development/README.md b/docs/sirider/s1/app-development/README.md new file mode 100644 index 000000000..2794a58b0 --- /dev/null +++ b/docs/sirider/s1/app-development/README.md @@ -0,0 +1,9 @@ +--- +sidebar_position: 4 +--- + +# 应用开发 + +主要介绍上层应用开发，比如 QT, WiringX, Mraa 等 + + diff --git a/docs/sirider/s1/app-development/zhouyi_model_zoo.md b/docs/sirider/s1/app-development/zhouyi_model_zoo.md new file mode 100644 index 000000000..a00a60efe --- /dev/null +++ b/docs/sirider/s1/app-development/zhouyi_model_zoo.md @@ -0,0 +1,266 @@ +--- +sidebar_position: 2 +--- +# 周易 Model Zoo + +[周易 Model Zoo](https://github.com/Arm-China/Model_zoo) 仓库提供了一套人工智能模型，供周易 SDK 参考使用。 + +#### **FTP 模型下载 (推荐 FTP 工具 [FileZilla](https://filezilla-project.org/))** + - `Host`: sftp://sftp01.armchina.com + - `Account`: zhouyi.armchina + - `Password`: 114r3cJd + + +| Model | Framework | Input Shape | Model Source | Quant Model | +|---------------------------|-----------|---------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------|-------------| +| mobilenet_v2 | caffe | [1,3,224,224] | https://github.com/shicai/MobileNet-Caffe | No | +| inception_v4 | tf1 | [1,299,299,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| deeplab_v2 | onnx | [1, 3, 513, 513] | https://github.com/kazuto1011/deeplab-pytorch | No | +| mtcnn_o | caffe | [1,3,48,48] | https://github.com/CongWeilin/mtcnn-caffe | No | +| inception_v4 | caffe | [1,3,299,299] | https://github.com/soeaver/caffe-model/tree/master/cls | No | +| deepspeech_v2 | onnx | [1,385,161,1] | https://github.com/tensorflow/models/tree/archive/research/deep_speech | No | +| wavenet | onnx | [1,390,23] | https://github.com/buriburisuri/speech-to-text-wavenet | No | +| shufflenet_v2 | onnx | [1,3,224,224] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| mobilenet_v2_ssd | onnx | [1,300,300,3] | https://github.com/tensorflow/models/tree/archive/research/object_detection/models | No | +| facenet | tf1 | [1, 160, 160, 3] | https://github.com/davidsandberg/facenet | No | +| yolo_v3_tiny | onnx | [1,416,416,3] | https://github.com/onnx/models/tree/main/vision/object_detection_segmentation/tiny-yolov3 | No | +| yolo_v3 | caffe | [1, 3, 608, 608] | https://github.com/foss-for-synopsys-dwc-arc-processors/synopsys-caffe-models/tree/master/caffe_models/yolo_v3 | No | +| yolo_v2_416 | tf1 | [1,416,416,3] | https://github.com/wojciechmo/yolo2 | No | +| yolo_v2_416 | caffe | [1,3,416,416] | https://github.com/tsingjinyun/caffe-yolov2 | No | +| yolo_v2_416 | onnx | [1,3,416,416] | https://github.com/wojciechmo/yolo2 | No | +| inception_v4 | onnx | [1,3,299,299] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| faster_rcnn | caffe | [1, 3, 224, 224] | https://github.com/rbgirshick/py-faster-rcnn | No | +| shufflenet_v2 | tf1 | [1,224,224,3] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| resnet_v1_101 | caffe | [1,3,224,224] | https://github.com/SnailTyan/caffe-model-zoo | No | +| mtcnn_p | caffe | [1,3,12,12] | https://github.com/CongWeilin/mtcnn-caffe | No | +| resnet_v1_50 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| alexnet | onnx | [1,1,28,28] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2_ssd | tf1 | [1,300,300,3] | https://github.com/tensorflow/models/tree/archive/research/object_detection/models | No | +| inception_v3 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| shufflenet_v2 | caffe | [1,3,224,224] | https://github.com/Ewenwan/ShuffleNet-2 | No | +| deepspeech_v2 | tf1 | [1,385,161,1] | https://github.com/tensorflow/models/tree/archive/research/deep_speech | No | +| inception_v3 | caffe | [1, 3, 299, 299] | https://github.com/soeaver/caffe-model/tree/master/cls | No | +| mobilenet_v2_ssd | caffe | [1,3,300,300] | https://github.com/chuanqi305/MobileNet-SSD | No | +| mtcnn_r | caffe | [1,3,24,24] | https://github.com/CongWeilin/mtcnn-caffe | No | +| resnet_v1_101 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| efficientnet_b5 | tf1 | [1, 456, 456, 3] | https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet | No | +| yolo_v3 | tf1 | [1, 416, 416, 3] | https://github.com/qqwweee/keras-yolo3 | No | +| alexnet | tf1 | [1,28,28,1] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| inception_v3 | onnx | [1, 3, 299, 299] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | caffe | [1,3,224,224] | https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md | No | +| resnet_v1_101 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| wavenet | tf1 | [1,390,23] | https://github.com/buriburisuri/speech-to-text-wavenet | No | +| alexnet | tflite | [1,28,28,1] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| shufflenet_v2 | tflite | [1,224,224,3] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| inception_v3 | tflite | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| inception_v4 | tflite | [1,299,299,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_101 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| alexnet | caffe | [1,3,227,227] | https://github.com/BVLC/caffe/tree/master/models/bvlc_alexnet | No | +| vgg_16 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | caffe | [1,3,224,224] | https://github.com/SnailTyan/caffe-model-zoo | No | +| deeplab_v3 | tflite | [1,257,257,3] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| mobilenet_v3 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet | No | +| peleenet | caffe | [1,3,224,224] | https://github.com/Robert-JunWang/PeleeNet/tree/master/caffe | No | +| fcn | caffe | [1,3,224,224] | https://github.com/shelhamer/fcn.berkeleyvision.org/tree/master/voc-fcn8s-atonce | No | +| deeplab_v3 | onnx | [1,3,513,513] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| erfnet | caffe | [1,3,512,1024] | https://github.com/Yuelong-Yu/ERFNet-Caffe | No | +| deeplab_v3 | tf1 | [1,513,513,3] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| centerface | onnx | [1, 3, 640, 640] | https://github.com/ttanzhiqiang/onnx_tensorrt_project | No | +| vgg_ssd | caffe | [1, 3, 300, 300] | https://github.com/weiliu89/caffe/tree/ssd | No | +| icnet | caffe | [1, 3, 1025, 2049] | https://github.com/hszhao/ICNet | No | +| resnet_34_ssd | onnx | [1, 3, 1200, 1200] | https://github.com/mlcommons/inference/tree/master/vision/classification_and_detection | No | +| resnet_v2_101 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| squeezenet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| sphereface | caffe | [1, 3, 112, 96] | https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/Sphereface/README.md | No | +| dpn_92 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| resnet_v2_152 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| se_resnet_50 | tf1 | [1, 224, 224, 3] | https://github.com/HiKapok/TF-SENet | No | +| srcnn | tf1 | [1, 33, 33, 1] | https://github.com/tegg89/SRCNN-Tensorflow | No | +| nasnet_mobile | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| super_resolution | onnx | [1, 1, 224, 224] | https://github.com/onnx/models/tree/master/vision/super_resolution/sub_pixel_cnn_2016 | No | +| squeezenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/squeezenet | No | +| resnet_v2_101 | caffe | [1, 3, 448, 448] | https://github.com/soeaver/caffe-model | No | +| densenet_121 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| yolo_v2 | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov2-coco | No | +| yolo_v2_tiny | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/tiny-yolov2 | No | +| inception_v2 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| lightface | onnx | [1, 3, 240, 320] | https://hailo.ai/devzone-model-zoo/face-detection/ | No | +| face_boxes | onnx | [1,3,1024,1024] | https://github.com/zisianw/FaceBoxes.PyTorch | No | +| vgg_cnn_s | caffe | [10, 3, 224, 224] | https://gist.github.com/ksimonyan/fd8800eeb36e276cd6f9 | No | +| mobilenet_v1 | caffe | [1, 3, 224, 224] | https://github.com/shicai/MobileNet-Caffe | No | +| regnet_x | onnx | [1, 3, 224, 224] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| yolact_regnetx | onnx | [1, 3, 512, 512] | https://hailo.ai/devzone-model-zoo/instance-segmentation/ | No | +| inception_resnet_v2 | caffe | [1, 3,331,331] | https://github.com/soeaver/caffe-model | No | +| resnext_50 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| resnet_v2_50 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| pnasnet_large | tf1 | [1, 331, 331, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| fast_depth | onnx | [1,3,224,224] | https://github.com/dwofk/fast-depth | No | +| efficientnet_lite | tf1 | [1, 280, 280, 3] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| centernet | tf1 | [1, 512, 512, 3] | https://hailo.ai/devzone-model-zoo/object-detection/ | No | +| inception_v1 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/inception_v1 | No | +| unet_bio | tf1 | [1,256, 256,1] | https://github.com/zhixuhao/unet | No | +| shufflenet_v1 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/shufflenet | No | +| resnext_101 | caffe | [1, 3, 224, 224] | https://www.deepdetect.com/models/resnext/ | No | +| se_resnet_50 | caffe | [1, 3, 225, 225] | https://github.com/soeaver/caffe-model | No | +| xception | caffe | [1, 3, 299, 299] | https://github.com/soeaver/caffe-model | No | +| resnet_v1_152 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| densenet_169 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| mixnet | tf1 | [1,224,224,3] | https://github.com/openvinotoolkit/open_model_zoo/tree/master/models/public/mixnet-l | No | +| mnasnet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| ann | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/ann | No | +| duc | onnx | [1, 3, 800, 800] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/duc | No | +| efficientnet_lite | tflite | [1, 300, 300, 3] | https://tfhub.dev/tensorflow/efficientnet/lite4/classification/2 | No | +| se_inception | caffe | [1, 3, 224, 224] | https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/se-inception/README.md | No | +| rnn_t_encoder | onnx | [1,249,240] | https://github.com/mlcommons/inference/tree/master/speech_recognition/rnnt | No | +| nasnet_mobile | tf1 | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| yolo_v5 | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5 | No | +| fcn16s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| dilation_8 | caffe | [1, 3, 900, 900] | https://github.com/fyu/dilation | No | +| zfnet_512 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/zfnet-512 | No | +| mobilenet_v2_ssd_lite | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| inception_v2 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| age_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | No | +| rnn_t_decoder | onnx | [1, 1],[1, 2, 320],[1, 2, 320],[1, 1, 1024] | https://github.com/mlcommons/inference/tree/master/speech_recognition/rnnt | No | +| stacked_hourglass | tf1 | [1, 256, 256, 3] | https://github.com/yuanyuanli85/Stacked_Hourglass_Network_Keras | No | +| resnet_v2_152 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| yolo_v1_tiny | caffe | [1, 3, 448, 448] | https://github.com/xingwangsfu/caffe-yolo | No | +| googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/googlenet | No | +| mobilenet_v1_224 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| poly_lanenet | onnx | [1, 3, 640, 640] | https://hailo.ai/devzone-model-zoo/driving-lane-detection/ | No | +| squeezenet | caffe | [10, 3, 227, 227] | https://github.com/forresti/SqueezeNet | No | +| inception_v2 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/inception_v2 | No | +| vgg_19 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| efficientnet_lite | onnx | [1,224,224,3] | https://github.com/onnx/models/tree/master/vision/classification/efficientnet-lite4 | No | +| caffenet | caffe | [10, 3, 227, 227] | https://github.com/BVLC/caffe/tree/master/models/bvlc_reference_caffenet | No | +| densenet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| mobilenet_v3 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet | No | +| googlenet | caffe | [10, 3, 224, 224] | https://github.com/BVLC/caffe/tree/master/models/bvlc_googlenet | No | +| densenet_121 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/densenet-121 | No | +| nasnet_large | tf1 | [1, 331, 331,3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| arcface | onnx | [1, 3, 112, 112] | https://github.com/onnx/models/tree/master/vision/body_analysis/arcface | No | +| dpn_68_extra | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| bisenet_v2 | tf1 | [4, 512, 1024, 3] | https://github.com/MaybeShewill-CV/bisenetv2-tensorflow | No | +| caffenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/caffenet | No | +| yolo_v4 | onnx | [1, 416, 416, 3] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov4 | No | +| inception_v1 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| fcn32s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| inception_v1 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| mobilenet_v1 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| resnet_v2_101 | tflite | [1, 299, 299, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| densenet_201 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| efficientnet_l | tf1 | [1, 300, 300, 3] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| pspnet | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/pspnet | No | +| inception_resnet_v2 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| se_resnet_101 | caffe | [1, 3, 224, 224] | https://github.com/hujie-frank/SENet | No | +| apcnet | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/apcnet | No | +| inception_resnet_v2 | tflite | [1, 299, 299, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| efficientdet | tflite | [ 1,512, 512, 3] | https://tfhub.dev/tensorflow/lite-model/efficientdet/lite3/detection/default/1 | No | +| yolo_v5_s | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5/ | No | +| gender_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | No | +| resnet_v1_ssd | tf1 | [1, 640, 640, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| retinanet | onnx | [1, 3, 480, 640] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/retinanet | No | +| deeplab_v3_xception | tf1 | [1,513,513,3] | https://github.com/tensorflow/models/blob/66264b2353aeeca3d4b340908a9590571495d5a6/research/deeplab/g3doc/model_zoo.md | No | +| yolo_v5 | tflite | [1, 640, 640, 3] | https://github.com/ultralytics/yolov5 | No | +| deeplabv3_mobilenetv2 | onnx | [1, 3, 512, 512] | https://github.com/zym1119/DeepLabv3_MobileNetv2_PyTorch | No | +| yolox_s | onnx | [1, 3, 640, 640] | https://github.com/Megvii-BaseDetection/YOLOX/tree/main/demo/ONNXRuntime | No | +| mobilenet_v1_ssd | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/ssd-mobilenetv1 | No | +| vision_transformer | onnx | [1,3,224,224] | https://github.com/jeonsworld/ViT-pytorch | No | +| ssd | onnx | [1, 3, 1200, 1200] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/ssd | No | +| inception_v2_ssd | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| yolo_v4 | tf1 | [1,416,416,3] | https://github.com/hunglc007/tensorflow-yolov4-tflite | No | +| mobilenet_v1_ssd | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| fcn8s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| drn_38 | onnx | [1, 3, 512, 1024] | https://github.com/fyu/drn | No | +| fsrcnn | onnx | [1, 1, 85, 85] | https://github.com/yjn870/FSRCNN-pytorch | No | +| pointnet | onnx | [32, 9, 4096] | https://github.com/yanx27/Pointnet_Pointnet2_pytorch | No | +| mobilenet_edgetpu | tflite | [1,224,224,3] | https://github.com/mlcommons/mobile_models/blob/main/v0_7/tflite/mobilenet_edgetpu_224_1.0_uint8.tflite | No | +| gru_l | tf1 | [1, 49, 10] | https://github.com/UT2UH/ML-KWS-for-ESP32/tree/master/Pretrained_models/GRU | No | +| swin_transformer | onnx | [1,3,384,384] | https://github.com/microsoft/Swin-Transformer | No | +| codeformer_256 | onnx | [1,3,256,256] | https://github.com/sczhou/CodeFormer/releases/tag/v0.1.0 | No | +| resnet_18 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/main/vision/classification/resnet/model | No | +| resnext_101 | onnx | [1, 3, 224, 224] | https://github.com/Cadene/pretrained-models.pytorch | No | +| unet_3d | onnx | [1, 3, 224, 224, 32] | https://zenodo.org/record/3904138#.YbBtatDP1PY | No | +| sne_roadseg | onnx | [1,3,384,1248] | https://github.com/hlwang1124/SNE-RoadSeg | No | +| maskrcnn | pytorch | [0] | https://pytorch.org/vision/main/models/mask_rcnn.html | No | +| yolo_v6s | onnx | [1,3,640,640] | https://github.com/DefTruth/lite.ai.toolkit | No | +| swin_transformer_base | onnx | [1,3,224,224] | https://github.com/microsoft/Swin-Transformer/blob/main/MODELHUB.md | No | +| 3d_unet | onnx | [0] | https://github.com/mlcommons/inference/tree/master/vision/medical_imaging/3d-unet-kits19 | No | +| xception | tf2 | [1, 299, 299, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| efficientnet_b5 | tf2 | [1,456,456,3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v1_ssd | tflite | [300, 300] | https://tfhub.dev/tensorflow/lite-model/ssd_mobilenet_v1/1/metadata/2 | No | +| transformer_official | tf1 | [1, 32] | https://github.com/Kyubyong/transformer | No | +| ViT_B_16 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/vision_transformer.html | No | +| efficientnet_b4_quant | tflite | [1,380,380,3] | https://ai-benchmark.com/download.html | Yes | +| dped_quant | tflite | [1,1536,2048,3] | https://ai-benchmark.com/download.html | Yes | +| shufflenet_v2 | onnx | [1,3,224,224] | https://github.com/TropComplique/shufflenet-v2-tensorflow | Yes | +| vgg_quant | tflite | [1,256,256,1] | https://ai-benchmark.com/download.html | Yes | +| caffenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/caffenet | Yes | +| mobilebert_quant | tflite | [1,384], [1,384], [1,384] | https://ai-benchmark.com/download.html | Yes | +| shufflenet_v2 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/shufflenetv2.html | Yes | +| fcn32s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | Yes | +| imdn_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| squeezenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/squeezenet | Yes | +| resnet_v1_50 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/resnet.html | Yes | +| vsr_quant | tflite | [1,540,960,3] | https://ai-benchmark.com/download.html | Yes | +| punet_quant | tflite | [1,544,960,4] | https://ai-benchmark.com/download.html | Yes | +| dped_instance_quant | tflite | [1,1024,1536,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v2_quant | tflite | [1,224,224,3] | https://ai-benchmark.com/download.html | Yes | +| vgg_16 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | Yes | +| mobilenet_v2_b8_quant | tflite | [8,224,224,3] | https://ai-benchmark.com/download.html | Yes | +| unet_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| resnet_v1_50 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | Yes | +| mobilenet_v3_quant | tflite | [1,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| crnn_quant | tflite | [1,64,200,3] | https://ai-benchmark.com/download.html | Yes | +| deeplab_v3_plus_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| age_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | Yes | +| googlenet | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/googlenet.html | Yes | +| inception_v3 | pytorch | [1,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.inception_v3 | Yes | +| srgan_quant | tflite | [1,256,256,3] | https://ai-benchmark.com/download.html | Yes | +| mv3_depth_quant | tflite | [1,1024,1536,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v3 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/mobilenetv3.html | Yes | +| mobiledet_ssd | tflite | [1, 320, 320, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | Yes | +| pynet_quant | tflite | [1,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v2 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | Yes | +| lstm_quant | tflite | [1,32,500,1] | https://ai-benchmark.com/download.html | Yes | +| yolo_v4_tiny_quant | tflite | [1,416,416,3] | https://ai-benchmark.com/download.html | Yes | +| inception_v3_quant | tflite | [1, 346, 346, 3] | https://ai-benchmark.com/download.html | Yes | +| esrgan_quant | tflite | [1,128,128,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v3_b4_quant | tflite | [4,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| xlsr_quant | tflite | [1,360,640,3] | https://ai-benchmark.com/download.html | Yes | +| efficientnet_b5 | pytorch | [1, 3, 456, 456] | https://pytorch.org/vision/stable/models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.efficientnet_b5 | No | +| petr | pytorch | [6, 3, 320, 800] | https://github.com/megvii-research/PETR | No | +| fcn | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/fcn.html | No | +| centernet_resnet50 | pytorch | [1, 3, 512, 512] | https://github.com/bubbliiiing/centernet-pytorch | No | +| resnet_34_ssd | tf1 | [1, 1200, 1200, 3] | https://github.com/mlcommons/inference/tree/r0.5/v0.5/classification_and_detection | No | +| densenet_121 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| yolo_v5s6 | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s6.pt | No | +| resnet_v1_101 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| shufflenet_v2 | pytorch | [1,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.shufflenet_v2_x1_0 | No | +| yolo_v4 | pytorch | [1, 3, 416, 416] | https://github.com/bubbliiiing/yolov4-pytorch/tree/master | No | +| resnet_50_v1_5 | tf1 | [1, 224, 224, 3] | https://zenodo.org/ | No | +| alexnet | pytorch | [2,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.alexnet.html#torchvision.models.alexnet | No | +| yolo_v4_tiny | pytorch | [1, 3, 416, 416] | https://github.com/bubbliiiing/yolov4-tiny-pytorch | No | +| se_resnext_50 | tf1 | [1, 224, 224, 3] | https://github.com/HiKapok/TF-SENet | No | +| vgg_16 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/vgg.html | No | +| mobilenet_v2 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| sea_former | onnx | [1, 3, 512, 1024] | https://github.com/fudan-zvg/SeaFormer/tree/main/seaformer-seg | No | +| resnet_v2_50 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v2_101 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| vgg_16 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v1_224 | tflite | [1, 224, 224, 3] | https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet_v1.md | No | +| vgg_19 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| yolox_l | onnx | [1,3,640,640] | https://github.com/Megvii-BaseDetection/YOLOX | No | +| yolo_v4_tiny | tflite | [1,416,416,3] | https://github.com/hunglc007/tensorflow-yolov4-tflite/tree/master | No | +| swin_transformer_tiny_224 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/swin_transformer.html | No | +| inception_v3 | tf2 | [1, 299, 299, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| nasnet_mobile | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v3 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v1_50 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v1_50 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/resnet.html | No | diff --git a/docs/sirider/s1/app-development/zhouyi_npu.md b/docs/sirider/s1/app-development/zhouyi_npu.md new file mode 100644 index 000000000..b06af0787 --- /dev/null +++ b/docs/sirider/s1/app-development/zhouyi_npu.md @@ -0,0 +1,148 @@ +--- +sidebar_position: 1 +--- +# 周易Z2 AIPU + +“周易” AIPU 是由安谋中国针对深度学习而自主研发的创新性 AI 专用处理器，它采用了创新性的架构设计，提供完整的硬件和软件生态，并且具有 PPA 最佳平衡。 +安谋中国还为“周易” AIPU 的客户提供很多工具来帮助他们进行开发，包括仿真器、编译器和调试器等进行数据的采集、分析。 +“周易” AIPU 也支持业界主流的 AI 规模框架，包括 TensorFlow、ONNX 等，未来也将支持更多不同的扩展框架。 + +“周易” Z2 AIPU 将主要面向中高端安防、智能座舱和 ADAS、边缘服务器等应用场景。 +## 周易 Z2 AIPU 使用教程 + +### x86 PC 端安装 周易 AIPU SDK +周易 SDK 是一个全栈平台，可为用户提供快速上市的开发和部署能力。 +![image](https://user-images.githubusercontent.com/85479712/198521602-49e13a31-bb49-424f-b782-5108274d63c3.png) + +- 在[瑞莎下载站](https://dl.radxa.com/sirider/s1/)下载周易 Z2 SDK 安装包后解压安装 + ```bash + tar -xvf Zhouyi_Z2.tar.gz + cd Zhouyi_Z2 && bash +x SETUP.SH + ``` +- 安装后得到的完整 SDK 文件如下 + + - `AI610-SDK-r1p3-AIoT` : ARM ZhouYi Z2 工具包 + + - `siengine` : siengine 提供的 ARM ZhouYi Z2 模型编译(nn-compiler-user-case-example)及板子部署(nn-runtime-user-case-example)的 demos + +- 配置 nn-compiler 环境 + ```bash + cd AI610-SDK-r1p3-AIoT/AI610-SDK-r1p3-00eac0/Out-Of-Box/out-of-box-nn-compiler + pip3 install -r lib_dependency.txt + ``` + 因为此 SDK 不包含模拟功能, 故安装过程会出现安装 AIPUSimProfiler 的报错，可以忽略 + + 若使用 venv 的用户请在 env_setup.sh 中 pip3 install 部分去掉 --user 选项 + ```bash + source env_setup.sh + ``` + +### x86 PC 端模型转换 +nn-compiler 可以将 TensorFlow、ONNX 等框架模型转换成可以在周易 AIPU 进行硬件加速推理的模型文件 +:::tip +此案例中将介绍开箱即用案例：resnet50 目标分类 + +完整 SDK 文档请参 `AI610-SDK-r1p3-AIoT/AI610-SDK-r1p3-00eac0/AI610-DOC-1001-r1p3-eac0` +::: + +- 进入 siengine nn-compiler-user-case-example 目录 + + 如没配置好 nn-compiler 环境， 请按照 [x86 PC 端安装 AIPU SDK](#x86-pc-端安装-aipu-sdk)进行配置 + + ```bash + cd siengine/nn-compiler-user-case-example/onnx + ``` + +- 生成量化校准集 + ```bash + python3 generate_calibration_data.py + ``` +- 生成用于模型推理的照片文件 + ```bash + python3 generate_input_binary.py + ``` + 文件在 ./resnet50/input_3_224_224.bin + +- (可选) 配置 build.cfg （开箱即用案例已提供） + ```bash + vim ./resnet50/build.cfg + ``` +- 生成 aipu 模型 + ```bash + cd ./restnet50 + aipubuild build.cfg + ``` + 在 ./restnet50 中得到 aipu_mlperf_resnet50.bin + +### 板端使用周易 Z2 推理 AIPU 模型 +在使用周易 Z2 AIPU 推理前需要在 x86 主机进行交叉编译生成可执行文件 `aiputest`，然后拷贝到 Sirider S1 中执行 +#### 在 x86 PC 端交叉编译二进制可执行文件 +- 安装 [gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu](https://releases.linaro.org/components/toolchain/binaries/latest-7/aarch64-linux-gnu/) 交叉编译工具链 + ```bash + tar -xvf gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu.tar + cp -r gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu /opt + ``` +- 编译 aiputest + + - 修改 UMDSRC 变量 + ```bash + cd siengine/nn-runtime-user-case-example + vim CMakeLists.txt + #set(UMDSRC "${CMAKE_SOURCE_DIR}/../AI610-SDK-${AIPU_VERSION}-00eac0/AI610-SDK-1012-${AIPU_VERSION}-eac0/Linux-driver/driver/umd") + set(UMDSRC "${CMAKE_SOURCE_DIR}/../../AI610-SDK-${AIPU_VERSION}-AIoT/AI610-SDK-r1p3-00eac0/AI610-SDK-1012-${AIPU_VERSION}-eac0/Linux-driver/driver/umd") + ``` + - 交叉编译 + ```bash + mkdir build && cd build + cmake -DCMAKE_BUILD_TYPE=Release .. + make + ``` + 编译生成的文件在 `siengine/nn-runtime-user-case-example/out/linux/aipu_test` + +#### 在 Sirider S1 进行板端推理 +- 将生成的 `aipu_mlperf_resnet50.bin` 模型文件，`input_3_224_224.bin` 照片文件，`aipu_test` 可执行文件，`out/linux/libs` 动态库文件夹复制到 Sirider S1 中 +- 执行 aipu_test + ```bash + export LD_LIBRARY_PATH=$LD_LIBRARY_PATH: + ./aipu_test aipu_mlperf_resnet50.bin input_3_224_224.bin + ``` + ```bash + (aiot-focal_overlayfs)root@linux:~/ssd# ./aipu_test aipu_mlperf_resnet50.bin input_3_224_224.bin + usage: ./aipu_test aipu.bin input0.bin + aipu_init_context success + aipu_load_graph_helper success: aipu_mlperf_resnet50.bin + aipu_create_job success + Frame #0 + aipu_finish_job success + No profiler data + get output tensor 0 success (1/1) + output_desc zero_point: 0.0000 scale: 5.5835 + idx: 637 fval: 21.4919 + idx: 749 fval: 19.8800 + idx: 415 fval: 16.1189 + idx: 412 fval: 15.0443 + idx: 791 fval: 14.1488 + Frame #1 + aipu_finish_job success + No profiler data + get output tensor 0 success (1/1) + output_desc zero_point: 0.0000 scale: 5.5835 + idx: 637 fval: 21.4919 + idx: 749 fval: 19.8800 + idx: 415 fval: 16.1189 + idx: 412 fval: 15.0443 + idx: 791 fval: 14.1488 + aipu_clean_job success + aipu_unload_graph success + aipu_deinit_ctx success + ``` + 两次的推理总时间 + ```bash + real 0m0.043s + user 0m0.008s + sys 0m0.023s + ``` + + 这里结果仅显示 推理结果的标签值，最大置信度 637 即对应 [imagenet1000](https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a) 中的 `mailbag, postbag` + + ![input.webp](/img/sirider/s1/aipu_1.webp) \ No newline at end of file diff --git a/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/README.md b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/README.md new file mode 100644 index 000000000..325ebf399 --- /dev/null +++ b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/README.md @@ -0,0 +1,9 @@ +--- +sidebar_position: 4 +--- + +# Application development + +Introduces upper-layer application development, such as QT, WiringX, Mraa, etc. + + diff --git a/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_model_zoo.md b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_model_zoo.md new file mode 100644 index 000000000..93830a4b9 --- /dev/null +++ b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_model_zoo.md @@ -0,0 +1,266 @@ +--- +sidebar_position: 2 +--- + +# Zhouyi Model Zoo + +The [Zhouyi Model Zoo](https://github.com/Arm-China/Model_zoo) repository provides a set of AI models for reference used by Zhouyi SDK. + +#### **FTP Model Download (Recommended FTP Tool: [FileZilla](https://filezilla-project.org/))** + - `Host`: sftp://sftp01.armchina.com + - `Account`: zhouyi.armchina + - `Password`: 114r3cJd + +| Model | Framework | Input Shape | Model Source | Quant Model | +|---------------------------|-----------|---------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------|-------------| +| mobilenet_v2 | caffe | [1,3,224,224] | https://github.com/shicai/MobileNet-Caffe | No | +| inception_v4 | tf1 | [1,299,299,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| deeplab_v2 | onnx | [1, 3, 513, 513] | https://github.com/kazuto1011/deeplab-pytorch | No | +| mtcnn_o | caffe | [1,3,48,48] | https://github.com/CongWeilin/mtcnn-caffe | No | +| inception_v4 | caffe | [1,3,299,299] | https://github.com/soeaver/caffe-model/tree/master/cls | No | +| deepspeech_v2 | onnx | [1,385,161,1] | https://github.com/tensorflow/models/tree/archive/research/deep_speech | No | +| wavenet | onnx | [1,390,23] | https://github.com/buriburisuri/speech-to-text-wavenet | No | +| shufflenet_v2 | onnx | [1,3,224,224] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| mobilenet_v2_ssd | onnx | [1,300,300,3] | https://github.com/tensorflow/models/tree/archive/research/object_detection/models | No | +| facenet | tf1 | [1, 160, 160, 3] | https://github.com/davidsandberg/facenet | No | +| yolo_v3_tiny | onnx | [1,416,416,3] | https://github.com/onnx/models/tree/main/vision/object_detection_segmentation/tiny-yolov3 | No | +| yolo_v3 | caffe | [1, 3, 608, 608] | https://github.com/foss-for-synopsys-dwc-arc-processors/synopsys-caffe-models/tree/master/caffe_models/yolo_v3 | No | +| yolo_v2_416 | tf1 | [1,416,416,3] | https://github.com/wojciechmo/yolo2 | No | +| yolo_v2_416 | caffe | [1,3,416,416] | https://github.com/tsingjinyun/caffe-yolov2 | No | +| yolo_v2_416 | onnx | [1,3,416,416] | https://github.com/wojciechmo/yolo2 | No | +| inception_v4 | onnx | [1,3,299,299] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| faster_rcnn | caffe | [1, 3, 224, 224] | https://github.com/rbgirshick/py-faster-rcnn | No | +| shufflenet_v2 | tf1 | [1,224,224,3] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| resnet_v1_101 | caffe | [1,3,224,224] | https://github.com/SnailTyan/caffe-model-zoo | No | +| mtcnn_p | caffe | [1,3,12,12] | https://github.com/CongWeilin/mtcnn-caffe | No | +| resnet_v1_50 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| alexnet | onnx | [1,1,28,28] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2_ssd | tf1 | [1,300,300,3] | https://github.com/tensorflow/models/tree/archive/research/object_detection/models | No | +| inception_v3 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| shufflenet_v2 | caffe | [1,3,224,224] | https://github.com/Ewenwan/ShuffleNet-2 | No | +| deepspeech_v2 | tf1 | [1,385,161,1] | https://github.com/tensorflow/models/tree/archive/research/deep_speech | No | +| inception_v3 | caffe | [1, 3, 299, 299] | https://github.com/soeaver/caffe-model/tree/master/cls | No | +| mobilenet_v2_ssd | caffe | [1,3,300,300] | https://github.com/chuanqi305/MobileNet-SSD | No | +| mtcnn_r | caffe | [1,3,24,24] | https://github.com/CongWeilin/mtcnn-caffe | No | +| resnet_v1_101 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| efficientnet_b5 | tf1 | [1, 456, 456, 3] | https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet | No | +| yolo_v3 | tf1 | [1, 416, 416, 3] | https://github.com/qqwweee/keras-yolo3 | No | +| alexnet | tf1 | [1,28,28,1] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| inception_v3 | onnx | [1, 3, 299, 299] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | caffe | [1,3,224,224] | https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md | No | +| resnet_v1_101 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| wavenet | tf1 | [1,390,23] | https://github.com/buriburisuri/speech-to-text-wavenet | No | +| alexnet | tflite | [1,28,28,1] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| mobilenet_v2 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | No | +| shufflenet_v2 | tflite | [1,224,224,3] | https://github.com/TropComplique/shufflenet-v2-tensorflow | No | +| inception_v3 | tflite | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| inception_v4 | tflite | [1,299,299,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_101 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| vgg_16 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| alexnet | caffe | [1,3,227,227] | https://github.com/BVLC/caffe/tree/master/models/bvlc_alexnet | No | +| vgg_16 | tf1 | [1,224,224,3] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | No | +| resnet_v1_50 | caffe | [1,3,224,224] | https://github.com/SnailTyan/caffe-model-zoo | No | +| deeplab_v3 | tflite | [1,257,257,3] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| mobilenet_v3 | tflite | [1,224,224,3] | https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet | No | +| peleenet | caffe | [1,3,224,224] | https://github.com/Robert-JunWang/PeleeNet/tree/master/caffe | No | +| fcn | caffe | [1,3,224,224] | https://github.com/shelhamer/fcn.berkeleyvision.org/tree/master/voc-fcn8s-atonce | No | +| deeplab_v3 | onnx | [1,3,513,513] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| erfnet | caffe | [1,3,512,1024] | https://github.com/Yuelong-Yu/ERFNet-Caffe | No | +| deeplab_v3 | tf1 | [1,513,513,3] | https://github.com/tensorflow/models/tree/archive/research/deeplab | No | +| centerface | onnx | [1, 3, 640, 640] | https://github.com/ttanzhiqiang/onnx_tensorrt_project | No | +| vgg_ssd | caffe | [1, 3, 300, 300] | https://github.com/weiliu89/caffe/tree/ssd | No | +| icnet | caffe | [1, 3, 1025, 2049] | https://github.com/hszhao/ICNet | No | +| resnet_34_ssd | onnx | [1, 3, 1200, 1200] | https://github.com/mlcommons/inference/tree/master/vision/classification_and_detection | No | +| resnet_v2_101 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| squeezenet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| sphereface | caffe | [1, 3, 112, 96] | https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/Sphereface/README.md | No | +| dpn_92 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| resnet_v2_152 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| se_resnet_50 | tf1 | [1, 224, 224, 3] | https://github.com/HiKapok/TF-SENet | No | +| srcnn | tf1 | [1, 33, 33, 1] | https://github.com/tegg89/SRCNN-Tensorflow | No | +| nasnet_mobile | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| super_resolution | onnx | [1, 1, 224, 224] | https://github.com/onnx/models/tree/master/vision/super_resolution/sub_pixel_cnn_2016 | No | +| squeezenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/squeezenet | No | +| resnet_v2_101 | caffe | [1, 3, 448, 448] | https://github.com/soeaver/caffe-model | No | +| densenet_121 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| yolo_v2 | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov2-coco | No | +| yolo_v2_tiny | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/tiny-yolov2 | No | +| inception_v2 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| lightface | onnx | [1, 3, 240, 320] | https://hailo.ai/devzone-model-zoo/face-detection/ | No | +| face_boxes | onnx | [1,3,1024,1024] | https://github.com/zisianw/FaceBoxes.PyTorch | No | +| vgg_cnn_s | caffe | [10, 3, 224, 224] | https://gist.github.com/ksimonyan/fd8800eeb36e276cd6f9 | No | +| mobilenet_v1 | caffe | [1, 3, 224, 224] | https://github.com/shicai/MobileNet-Caffe | No | +| regnet_x | onnx | [1, 3, 224, 224] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| yolact_regnetx | onnx | [1, 3, 512, 512] | https://hailo.ai/devzone-model-zoo/instance-segmentation/ | No | +| inception_resnet_v2 | caffe | [1, 3,331,331] | https://github.com/soeaver/caffe-model | No | +| resnext_50 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| resnet_v2_50 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| pnasnet_large | tf1 | [1, 331, 331, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| fast_depth | onnx | [1,3,224,224] | https://github.com/dwofk/fast-depth | No | +| efficientnet_lite | tf1 | [1, 280, 280, 3] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| centernet | tf1 | [1, 512, 512, 3] | https://hailo.ai/devzone-model-zoo/object-detection/ | No | +| inception_v1 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/inception_v1 | No | +| unet_bio | tf1 | [1,256, 256,1] | https://github.com/zhixuhao/unet | No | +| shufflenet_v1 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/shufflenet | No | +| resnext_101 | caffe | [1, 3, 224, 224] | https://www.deepdetect.com/models/resnext/ | No | +| se_resnet_50 | caffe | [1, 3, 225, 225] | https://github.com/soeaver/caffe-model | No | +| xception | caffe | [1, 3, 299, 299] | https://github.com/soeaver/caffe-model | No | +| resnet_v1_152 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| densenet_169 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| mixnet | tf1 | [1,224,224,3] | https://github.com/openvinotoolkit/open_model_zoo/tree/master/models/public/mixnet-l | No | +| mnasnet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| ann | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/ann | No | +| duc | onnx | [1, 3, 800, 800] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/duc | No | +| efficientnet_lite | tflite | [1, 300, 300, 3] | https://tfhub.dev/tensorflow/efficientnet/lite4/classification/2 | No | +| se_inception | caffe | [1, 3, 224, 224] | https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/se-inception/README.md | No | +| rnn_t_encoder | onnx | [1,249,240] | https://github.com/mlcommons/inference/tree/master/speech_recognition/rnnt | No | +| nasnet_mobile | tf1 | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| yolo_v5 | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5 | No | +| fcn16s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| dilation_8 | caffe | [1, 3, 900, 900] | https://github.com/fyu/dilation | No | +| zfnet_512 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/zfnet-512 | No | +| mobilenet_v2_ssd_lite | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| inception_v2 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| age_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | No | +| rnn_t_decoder | onnx | [1, 1],[1, 2, 320],[1, 2, 320],[1, 1, 1024] | https://github.com/mlcommons/inference/tree/master/speech_recognition/rnnt | No | +| stacked_hourglass | tf1 | [1, 256, 256, 3] | https://github.com/yuanyuanli85/Stacked_Hourglass_Network_Keras | No | +| resnet_v2_152 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| yolo_v1_tiny | caffe | [1, 3, 448, 448] | https://github.com/xingwangsfu/caffe-yolo | No | +| googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/googlenet | No | +| mobilenet_v1_224 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| poly_lanenet | onnx | [1, 3, 640, 640] | https://hailo.ai/devzone-model-zoo/driving-lane-detection/ | No | +| squeezenet | caffe | [10, 3, 227, 227] | https://github.com/forresti/SqueezeNet | No | +| inception_v2 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/inception_v2 | No | +| vgg_19 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| efficientnet_lite | onnx | [1,224,224,3] | https://github.com/onnx/models/tree/master/vision/classification/efficientnet-lite4 | No | +| caffenet | caffe | [10, 3, 227, 227] | https://github.com/BVLC/caffe/tree/master/models/bvlc_reference_caffenet | No | +| densenet | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| mobilenet_v3 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet | No | +| googlenet | caffe | [10, 3, 224, 224] | https://github.com/BVLC/caffe/tree/master/models/bvlc_googlenet | No | +| densenet_121 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/densenet-121 | No | +| nasnet_large | tf1 | [1, 331, 331,3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| arcface | onnx | [1, 3, 112, 112] | https://github.com/onnx/models/tree/master/vision/body_analysis/arcface | No | +| dpn_68_extra | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| bisenet_v2 | tf1 | [4, 512, 1024, 3] | https://github.com/MaybeShewill-CV/bisenetv2-tensorflow | No | +| caffenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/caffenet | No | +| yolo_v4 | onnx | [1, 416, 416, 3] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov4 | No | +| inception_v1 | tf1 | [1, 224, 224, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| fcn32s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| inception_v1 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| mobilenet_v1 | tflite | [1, 224, 224, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| resnet_v2_101 | tflite | [1, 299, 299, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| densenet_201 | caffe | [1, 3, 224, 224] | https://github.com/soeaver/caffe-model | No | +| efficientnet_l | tf1 | [1, 300, 300, 3] | https://hailo.ai/devzone-model-zoo/about-object-detection/ | No | +| pspnet | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/pspnet | No | +| inception_resnet_v2 | tf1 | [1, 299, 299, 3] | https://github.com/tensorflow/models/tree/master/research/slim#Pretrained | No | +| se_resnet_101 | caffe | [1, 3, 224, 224] | https://github.com/hujie-frank/SENet | No | +| apcnet | onnx | [1, 3, 1024, 2048] | https://github.com/open-mmlab/mmsegmentation/tree/master/configs/apcnet | No | +| inception_resnet_v2 | tflite | [1, 299, 299, 3] | https://www.tensorflow.org/lite/guide/hosted_models?hl=zh-cn | No | +| efficientdet | tflite | [ 1,512, 512, 3] | https://tfhub.dev/tensorflow/lite-model/efficientdet/lite3/detection/default/1 | No | +| yolo_v5_s | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5/ | No | +| gender_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | No | +| resnet_v1_ssd | tf1 | [1, 640, 640, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| retinanet | onnx | [1, 3, 480, 640] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/retinanet | No | +| deeplab_v3_xception | tf1 | [1,513,513,3] | https://github.com/tensorflow/models/blob/66264b2353aeeca3d4b340908a9590571495d5a6/research/deeplab/g3doc/model_zoo.md | No | +| yolo_v5 | tflite | [1, 640, 640, 3] | https://github.com/ultralytics/yolov5 | No | +| deeplabv3_mobilenetv2 | onnx | [1, 3, 512, 512] | https://github.com/zym1119/DeepLabv3_MobileNetv2_PyTorch | No | +| yolox_s | onnx | [1, 3, 640, 640] | https://github.com/Megvii-BaseDetection/YOLOX/tree/main/demo/ONNXRuntime | No | +| mobilenet_v1_ssd | onnx | [1, 3, 416, 416] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/ssd-mobilenetv1 | No | +| vision_transformer | onnx | [1,3,224,224] | https://github.com/jeonsworld/ViT-pytorch | No | +| ssd | onnx | [1, 3, 1200, 1200] | https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/ssd | No | +| inception_v2_ssd | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| yolo_v4 | tf1 | [1,416,416,3] | https://github.com/hunglc007/tensorflow-yolov4-tflite | No | +| mobilenet_v1_ssd | tf1 | [1, 300, 300, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | No | +| fcn8s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | No | +| drn_38 | onnx | [1, 3, 512, 1024] | https://github.com/fyu/drn | No | +| fsrcnn | onnx | [1, 1, 85, 85] | https://github.com/yjn870/FSRCNN-pytorch | No | +| pointnet | onnx | [32, 9, 4096] | https://github.com/yanx27/Pointnet_Pointnet2_pytorch | No | +| mobilenet_edgetpu | tflite | [1,224,224,3] | https://github.com/mlcommons/mobile_models/blob/main/v0_7/tflite/mobilenet_edgetpu_224_1.0_uint8.tflite | No | +| gru_l | tf1 | [1, 49, 10] | https://github.com/UT2UH/ML-KWS-for-ESP32/tree/master/Pretrained_models/GRU | No | +| swin_transformer | onnx | [1,3,384,384] | https://github.com/microsoft/Swin-Transformer | No | +| codeformer_256 | onnx | [1,3,256,256] | https://github.com/sczhou/CodeFormer/releases/tag/v0.1.0 | No | +| resnet_18 | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/main/vision/classification/resnet/model | No | +| resnext_101 | onnx | [1, 3, 224, 224] | https://github.com/Cadene/pretrained-models.pytorch | No | +| unet_3d | onnx | [1, 3, 224, 224, 32] | https://zenodo.org/record/3904138#.YbBtatDP1PY | No | +| sne_roadseg | onnx | [1,3,384,1248] | https://github.com/hlwang1124/SNE-RoadSeg | No | +| maskrcnn | pytorch | [0] | https://pytorch.org/vision/main/models/mask_rcnn.html | No | +| yolo_v6s | onnx | [1,3,640,640] | https://github.com/DefTruth/lite.ai.toolkit | No | +| swin_transformer_base | onnx | [1,3,224,224] | https://github.com/microsoft/Swin-Transformer/blob/main/MODELHUB.md | No | +| 3d_unet | onnx | [0] | https://github.com/mlcommons/inference/tree/master/vision/medical_imaging/3d-unet-kits19 | No | +| xception | tf2 | [1, 299, 299, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| efficientnet_b5 | tf2 | [1,456,456,3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v1_ssd | tflite | [300, 300] | https://tfhub.dev/tensorflow/lite-model/ssd_mobilenet_v1/1/metadata/2 | No | +| transformer_official | tf1 | [1, 32] | https://github.com/Kyubyong/transformer | No | +| ViT_B_16 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/vision_transformer.html | No | +| efficientnet_b4_quant | tflite | [1,380,380,3] | https://ai-benchmark.com/download.html | Yes | +| dped_quant | tflite | [1,1536,2048,3] | https://ai-benchmark.com/download.html | Yes | +| shufflenet_v2 | onnx | [1,3,224,224] | https://github.com/TropComplique/shufflenet-v2-tensorflow | Yes | +| vgg_quant | tflite | [1,256,256,1] | https://ai-benchmark.com/download.html | Yes | +| caffenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/caffenet | Yes | +| mobilebert_quant | tflite | [1,384], [1,384], [1,384] | https://ai-benchmark.com/download.html | Yes | +| shufflenet_v2 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/shufflenetv2.html | Yes | +| fcn32s | onnx | [1, 3, 224, 224] | https://github.com/wkentaro/pytorch-fcn/ | Yes | +| imdn_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| squeezenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/classification/squeezenet | Yes | +| resnet_v1_50 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/resnet.html | Yes | +| vsr_quant | tflite | [1,540,960,3] | https://ai-benchmark.com/download.html | Yes | +| punet_quant | tflite | [1,544,960,4] | https://ai-benchmark.com/download.html | Yes | +| dped_instance_quant | tflite | [1,1024,1536,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v2_quant | tflite | [1,224,224,3] | https://ai-benchmark.com/download.html | Yes | +| vgg_16 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | Yes | +| mobilenet_v2_b8_quant | tflite | [8,224,224,3] | https://ai-benchmark.com/download.html | Yes | +| unet_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| resnet_v1_50 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets | Yes | +| mobilenet_v3_quant | tflite | [1,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| crnn_quant | tflite | [1,64,200,3] | https://ai-benchmark.com/download.html | Yes | +| deeplab_v3_plus_quant | tflite | [1,1024,1024,3] | https://ai-benchmark.com/download.html | Yes | +| age_googlenet | onnx | [1, 3, 224, 224] | https://github.com/onnx/models/tree/master/vision/body_analysis/age_gender | Yes | +| googlenet | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/googlenet.html | Yes | +| inception_v3 | pytorch | [1,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.inception_v3 | Yes | +| srgan_quant | tflite | [1,256,256,3] | https://ai-benchmark.com/download.html | Yes | +| mv3_depth_quant | tflite | [1,1024,1536,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v3 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/mobilenetv3.html | Yes | +| mobiledet_ssd | tflite | [1, 320, 320, 3] | https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md | Yes | +| pynet_quant | tflite | [1,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v2 | onnx | [1,3,224,224] | https://github.com/tensorflow/models/tree/archive/research/slim/nets/mobilenet | Yes | +| lstm_quant | tflite | [1,32,500,1] | https://ai-benchmark.com/download.html | Yes | +| yolo_v4_tiny_quant | tflite | [1,416,416,3] | https://ai-benchmark.com/download.html | Yes | +| inception_v3_quant | tflite | [1, 346, 346, 3] | https://ai-benchmark.com/download.html | Yes | +| esrgan_quant | tflite | [1,128,128,3] | https://ai-benchmark.com/download.html | Yes | +| mobilenet_v3_b4_quant | tflite | [4,512,512,3] | https://ai-benchmark.com/download.html | Yes | +| xlsr_quant | tflite | [1,360,640,3] | https://ai-benchmark.com/download.html | Yes | +| efficientnet_b5 | pytorch | [1, 3, 456, 456] | https://pytorch.org/vision/stable/models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.efficientnet_b5 | No | +| petr | pytorch | [6, 3, 320, 800] | https://github.com/megvii-research/PETR | No | +| fcn | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/fcn.html | No | +| centernet_resnet50 | pytorch | [1, 3, 512, 512] | https://github.com/bubbliiiing/centernet-pytorch | No | +| resnet_34_ssd | tf1 | [1, 1200, 1200, 3] | https://github.com/mlcommons/inference/tree/r0.5/v0.5/classification_and_detection | No | +| densenet_121 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| yolo_v5s6 | onnx | [1, 3, 640, 640] | https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s6.pt | No | +| resnet_v1_101 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| shufflenet_v2 | pytorch | [1,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.shufflenet_v2_x1_0 | No | +| yolo_v4 | pytorch | [1, 3, 416, 416] | https://github.com/bubbliiiing/yolov4-pytorch/tree/master | No | +| resnet_50_v1_5 | tf1 | [1, 224, 224, 3] | https://zenodo.org/ | No | +| alexnet | pytorch | [2,3,224,224] | https://pytorch.org/vision/stable/models/generated/torchvision.models.alexnet.html#torchvision.models.alexnet | No | +| yolo_v4_tiny | pytorch | [1, 3, 416, 416] | https://github.com/bubbliiiing/yolov4-tiny-pytorch | No | +| se_resnext_50 | tf1 | [1, 224, 224, 3] | https://github.com/HiKapok/TF-SENet | No | +| vgg_16 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/vgg.html | No | +| mobilenet_v2 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| sea_former | onnx | [1, 3, 512, 1024] | https://github.com/fudan-zvg/SeaFormer/tree/main/seaformer-seg | No | +| resnet_v2_50 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v2_101 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| vgg_16 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v1_224 | tflite | [1, 224, 224, 3] | https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet_v1.md | No | +| vgg_19 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| yolox_l | onnx | [1,3,640,640] | https://github.com/Megvii-BaseDetection/YOLOX | No | +| yolo_v4_tiny | tflite | [1,416,416,3] | https://github.com/hunglc007/tensorflow-yolov4-tflite/tree/master | No | +| swin_transformer_tiny_224 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/swin_transformer.html | No | +| inception_v3 | tf2 | [1, 299, 299, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| nasnet_mobile | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| mobilenet_v3 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v1_50 | tf2 | [1, 224, 224, 3] | https://www.tensorflow.org/versions/r2.6/api_docs/python/tf/keras/applications | No | +| resnet_v1_50 | pytorch | [1, 3, 224, 224] | https://pytorch.org/vision/stable/models/resnet.html | No | diff --git a/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_npu.md b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_npu.md new file mode 100644 index 000000000..4017c2557 --- /dev/null +++ b/i18n/en/docusaurus-plugin-content-docs/current/sirider/s1/app-development/zhouyi_npu.md @@ -0,0 +1,159 @@ +--- +sidebar_position: 1 +--- + +# Zhouyi Z2 AIPU + +The "Zhouyi" AIPU is an innovative AI-specific processor independently developed by Arm China for deep learning. It features an innovative architecture design, providing a complete hardware and software ecosystem with the best balance of PPA (Performance, Power, Area). +Arm China also provides a range of tools for "Zhouyi" AIPU customers to assist in development, including simulators, compilers, and debuggers for data collection and analysis. +The "Zhouyi" AIPU supports mainstream AI frameworks, including TensorFlow and ONNX, and will support more extended frameworks in the future. + +The "Zhouyi" Z2 AIPU is primarily targeted at high-end security, intelligent cockpits and ADAS (Advanced Driver Assistance Systems), edge servers, and other application scenarios. + +## Zhouyi Z2 AIPU User Guide + +### Install Zhouyi AIPU SDK on x86 PC + +The Zhouyi SDK is a full-stack platform that provides users with rapid development and deployment capabilities. + +![image](https://user-images.githubusercontent.com/85479712/198521602-49e13a31-bb49-424f-b782-5108274d63c3.png) + +- Download the Zhouyi Z2 SDK installation package from the [Radxa Download Station](https://dl.radxa.com/sirider/s1/) and extract it for installation: + ```bash + tar -xvf Zhouyi_Z2.tar.gz + cd Zhouyi_Z2 && bash +x SETUP.SH + ``` +- After installation, the complete SDK files are as follows: + + - `AI610-SDK-r1p3-AIoT`: ARM Zhouyi Z2 toolkit + + - `siengine`: Demos provided by siengine for ARM Zhouyi Z2 model compilation (nn-compiler-user-case-example) and board deployment (nn-runtime-user-case-example) + +- Configure the nn-compiler environment: + ```bash + cd AI610-SDK-r1p3-AIoT/AI610-SDK-r1p3-00eac0/Out-Of-Box/out-of-box-nn-compiler + pip3 install -r lib_dependency.txt + ``` + Since this SDK does not include simulation functionality, errors may occur when installing AIPUSimProfiler. These can be ignored. + + If using a virtual environment (venv), please remove the --user option from the pip3 install part in env_setup.sh: + ```bash + source env_setup.sh + ``` + +### Model Conversion on x86 PC + +The nn-compiler can convert models from frameworks like TensorFlow and ONNX into models that can be accelerated by the Zhouyi AIPU for inference. + +:::tip +This case introduces an out-of-the-box example: resnet50 for object classification. + +For the complete SDK documentation, please refer to `AI610-SDK-r1p3-AIoT/AI610-SDK-r1p3-00eac0/AI610-DOC-1001-r1p3-eac0`. +::: + +- Enter the siengine nn-compiler-user-case-example directory. + + If the nn-compiler environment is not configured, please follow [Install Zhouyi AIPU SDK on x86 PC](#install-zhouyi-aipu-sdk-on-x86-pc) to configure. + + ```bash + cd siengine/nn-compiler-user-case-example/onnx + ``` + +- Generate the quantization calibration set: + ```bash + python3 generate_calibration_data.py + ``` +- Generate image files for model inference: + ```bash + python3 generate_input_binary.py + ``` + The file is located in ./resnet50/input_3_224_224.bin. + +- (Optional) Configure build.cfg (provided in out-of-the-box example): + ```bash + vim ./resnet50/build.cfg + ``` +- Generate the aipu model: + ```bash + cd ./restnet50 + aipubuild build.cfg + ``` + The aipu model is generated in ./restnet50 as aipu_mlperf_resnet50.bin. + +### Use Zhouyi Z2 for AIPU Model Inference on the Board + +Before using Zhouyi Z2 AIPU for inference, a cross-compiled executable file `aiputest` needs to be generated on the x86 host and then copied to the Sirider S1 for execution. + +#### Cross-compiling the Binary Executable on x86 PC + +- Install the [gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu](https://releases.linaro.org/components/toolchain/binaries/latest-7/aarch64-linux-gnu/) cross-compilation toolchain: + ```bash + tar -xvf gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu.tar + cp -r gcc-linaro-7.5.0-2019.12-x86_64_aarch64-linux-gnu /opt + ``` +- Compile aiputest: + + - Modify the UMDSRC variable: + ```bash + cd siengine/nn-runtime-user-case-example + vim CMakeLists.txt + #set(UMDSRC "${CMAKE_SOURCE_DIR}/../AI610-SDK-${AIPU_VERSION}-00eac0/AI610-SDK-1012-${AIPU_VERSION}-eac0/Linux-driver/driver/umd") + set(UMDSRC "${CMAKE_SOURCE_DIR}/../../AI610-SDK-${AIPU_VERSION}-AIoT/AI610-SDK-r1p3-00eac0/AI610-SDK-1012-${AIPU_VERSION}-eac0/Linux-driver/driver/umd") + ``` + - Cross-compile: + ```bash + mkdir build && cd build + cmake -DCMAKE_BUILD_TYPE=Release .. + make + ``` + The compiled file is located in `siengine/nn-runtime-user-case-example/out/linux/aipu_test`. + +#### Inference on the Sirider S1 + +- Copy the generated `aipu_mlperf_resnet50.bin` model file, `input_3_224_224.bin` image file, `aipu_test` executable file, and `out/linux/libs` dynamic library folder to the Sirider S1. +- Execute aipu_test: + ```bash + export LD_LIBRARY_PATH=$LD_LIBRARY_PATH: + ./aipu_test aipu_mlperf_resnet50.bin input_3_224_224.bin + ``` + ```bash + (aiot-focal_overlayfs)root@linux:~/ssd# ./aipu_test aipu_mlperf_resnet50.bin input_3_224_224.bin + usage: ./aipu_test aipu.bin input0.bin + aipu_init_context success + aipu_load_graph_helper success: aipu_mlperf_resnet50.bin + aipu_create_job success + Frame #0 + aipu_finish_job success + No profiler data + get output tensor 0 success (1/1) + output_desc zero_point: 0.0000 scale: 5.5835 + idx: 637 fval: 21.4919 + idx: 749 fval: 19.8800 + idx: 415 fval: 16.1189 + idx: 412 fval: 15.0443 + idx: 791 fval: 14.1488 + Frame #1 + aipu_finish_job success + No profiler data + get output tensor 0 success (1/1) + output_desc zero_point: 0.0000 scale: 5.5835 + idx: 637 fval: 21.4919 + idx: 749 fval: 19.8800 + idx: 415 fval: 16.1189 + idx: 412 fval: 15.0443 + idx: 791 fval: 14.1488 + aipu_clean_job success + aipu_unload_graph success + aipu_deinit_ctx success + ``` + The total time for two inferences: + ```bash + real 0m0.043s + user 0m0.008s + sys 0m0.023s + ``` + + The result here only shows the labels of the inference results, with the highest confidence being 637, corresponding to `mailbag, postbag` in [imagenet1000](https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a). + +![input.webp](/img/sirider/s1/aipu_1.webp) + diff --git a/static/img/sirider/s1/aipu_1.webp b/static/img/sirider/s1/aipu_1.webp new file mode 100644 index 000000000..29a0875d7 Binary files /dev/null and b/static/img/sirider/s1/aipu_1.webp differ