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.

Caffe Jacinto for Windows



Hello Team,

We are planning to work on object detection task on Jacinto 7 series.

Could you please suggest which training repo/framework should we need to use on windows 10?

Could you please also share the necessary links to setup for the same.

Thanks & Regards,

Chandrashekhar 

  • Hi Chandrashekhar,


    We are happy to provide the necessary information and get you started.

    I have to admit that I don't have much expertise with Deep Learning on Windows. These days most of the Deep Learning training activities are done using Linux machines - so it is best that we use a Linux machine (Ubuntu OS if possible) with GPUs. 

    Another option is to use a cloud service such as AWS or Google Cloud - they have Ubuntu machines with GPUs. 

    Coming to actual training framework, both PyTorch and Tensorflow/Keras are popular these days for Deep Learning training and Jacinto7 SDK supports both of these. Do you have any prior experience in these  - what does your team predominantly use?

    This is our landing page related to Deep Learning Training:

    https://github.com/TexasInstruments/jacinto-ai-devkit

    And this is a ModelZoo with example models for Jacinto 7:

    https://git.ti.com/cgit/jacinto-ai/jacinto-ai-modelzoo/about/

    One you get familiarized with the information in the above pages, we can discuss more specific details.

    Best regards, Manu

  • To be specific, we have used Ubuntu 18.04 OS with Nvidia GPUs. Any of the recent Nvidia GPUs will be fine.

    Do you have any further questions?