Closed NHSE closed 2 years ago
The caffe model is generated by vitis-ai v1.1, and the elf is compiled under DNNDK v3.1 with specific dcf file which generated by inputing custom XSA design file in "dlet" tool. You can find the XSA file here I'm sorry that I'm not able to provide an complete tutorial soon.
Is it not possible to proceed with Vitisai v1.2?
How should I write the target in the Json file? I'm sorry to bother you
Is it not possible to proceed with Vitisai v1.2?
The Caffe model can be generated in Vitis-ai v1.2, but I'm not sure if the compilation of the elf file can be finished in Vitis-ai instead of DNNDK since Vitis-ai v1.2 appears not supporting the zynq-7000 series.
How should I write the target in the Json file? I'm sorry to bother you
I think you can have a look in C++ json libs like nlohmann/json Or this struct mapping approachs to sotre your data in struct and save in json.
If you need DNNDKv3.1, you can refer to my another forked jobs petalinux-docker. It can help you build a docker image with petalinux and DNNDKv3.1.
Can you tell me the tutorial?
How do I create an elf file with DNNDK version 3.1? I have to complete the execution of the custom YOLO by next Monday. Can you help me?
I succeeded Thank you!
I made an elf file using Vitis AI. However, command 'make' using the elf file does not proceed due to formatting errors. Do you know how to solve this problem?