Comments

• Hi, Our model input size is 100x100x3 and output size is 1x3. 1st test, the output is: 9.65505123, -2.9817071, -7.66724682 Since first number is largest among the three, the correct result is "0". 2nd test, the output is -1.56184661, …

  in Same deep learning but different results between RGB565 and RGBA8888 Comment by LeoRHTang June 2021
• Hi, Ethon: Thanks for your reply. We found the problem. We use another USB dongle and run in two environment: (1) kneron/toolchain:520 + host_lib v0.8 (2) kneron/toolchain:latest + host_lib v0.9 and they work in both cases. We think we should upd…

  in Bind multiple models in a single binary file Comment by LeoRHTang April 2021
• Hi, We found the root cause of this problem. There are differences between #Original and # New model. Some of the OPs are removed during model conversion, for example: "Batch Normalization" and "ReLu". We suspect that the conv…

  in Run pre-processing in model head Comment by LeoRHTang April 2021
• Hi, kidd: Thanks for you reply. # Original @batch_compile_input_params.json "img_preprocess_method": ["customized"], @script/img_preprocess.py def normalization(x, mode, **kwargs): ... if mode == 'customized': #-123 - 1…

  in Run pre-processing in model head Comment by LeoRHTang March 2021