⚡️An Easy-to-use and Fast Deep Learning Model Deployment Toolkit for ☁️Cloud 📱Mobile and 📹Edge. Including Image, Video, Text and Audio 20+ main stream scenarios and 150+ SOTA models with end-to-end optimization, multi-platform and multi-framework support.
使用文档中的Unet分割模型转成rknn部署到瑞芯微上提示该错误:
配置文件 模型转换命令:
python paddle_infer_shape.py --model_dir ./ppyolo_r50vd_dcn_1x_coco \ --model_filename model.pdmodel \ --params_filename model.pdiparams \ --save_dir ./inference \ --input_shape_dict="{'image':[1,3,608,608], 'scale_factor':[1,2],'im_shape':[1,2]}"
paddle2onnx --model_dir ./inference_seg --model_filename model.pdmodel --params_filename model.pdiparams --save_file Unet_cityscapes_without_argmax_infer/unet_lane_seg.onnx --enable_dev_version False --opset_version 12 --enable_onnx_checker True
其中--enable_dev_version改为True的话会提示错误:
python3 ../FastDeploy/tools/rknpu2/export.py --config_path=/data/project/rknn-toolkit2/models/Unet_cityscapes_without_argmax_infer/config.yml --target_platform=RK3588
模型转换后预测代码:
预测错误:
【FastDeploy版本】:develop