This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA2: Host emulation build for SDK0304

Part Number: TDA2

hi,dear

              In the paper called deep learning user guide , I could not bringup the host emulation mode ,the ref information and the error as following:

The command is so simple, however it did not work. Could you please help me figure out how to set it ?

BR

  • Hi,

    You need source code release of TIDL to build in host emulation mode.

    Do you have access to it?

    Thanks,
    Praveen
  • No,I did not have that code. However I really saw the makefile. There is the ./src directory in the make process. Could you please kind enough to offer me the code? I want to research that and do some important examples to show the capability of tda2x processer. If it can do well (especially DL related algorithm) and develop easily , there will be more cooperation.
  • Hi,

    Source code will be shared only through CDDS. Please contact you local TI FAE to get access to CDDS account and source code release.

    Thanks,
    Praveen
  • Any other method? The communication process of the FAE in China is too Inefficient. Let me offer you the order of your company's TDA2x EVM, is it OK ?
  • No, I cannot help here further. Please contact your local TI representative only, that is the only way many customer got the source code.

    Thanks,
    Praveen
  • In fact, the representative delayed me so long. I could not get the feedback from him/her. How can I get forward to my work? Could I talk the manager with yours or the departmant manager?
  • Hi ,
    The TIDL Host emulation is NOT needed for running a CNN network on TI device (It is only needed for de-bugging purpose). The binary release available in VSDK package could run CNN inference in TI devices, use this binary release for your network deployment.

    Regards,
    Kumar.D
  • hi, dear

    Now although I bought serveral tda2x boards, I have only got part service from the TI tech support in Shanghai of China. I could not get the source code of TIDL. So if I want to run my improved cnn model , I can only run it on the tda2x board not on PC simulator. I have imported the cnn model to the bin format which is suitable for TIDL. Yet the input data need to be resize to 368 X 368,and I should resize it by myself( It has to be the TI method to resize). Even worse, the configuration file (only seg , ssd detection included)on the tda2x board looks like that below:

    TIDL Configuration parameters
    -----------------------------

    ################################################################################
    IMP: Make sure the size of the file names (excluding extension) is not more
    than 8 characters.
    ################################################################################

    inputWidth=1024
    inputHeight=512
    inputFile=IN.RGB
    outputFile=OUT.BIN
    netFileName=NET.BIN
    paramFileName=PRM.BIN

    My cnn net is not for detection or classifcation, but pose recognition. I do not know if I can use this demo cfg file to run my cnn model successfully on the tda2x.

    If not , could you please be kind enough to offer me new cfg file or make some change for me to run?

    tks, looking forward to your reply.
  • Hi,

    this question is discussed here:
    e2e.ti.com/.../723315

    I will close this thread.

    Regards,
    Yordan