What layer configurations can be combined?

hyunhyun
in AI Model migration

I have two models with similar sizes and layers. One model says the memory is too large to be converted, and one model has been converted. The only difference between the two models is batch normalization and resolution. What layer configurations can be combined?

Tagged:

Comments

  • Maria ChenMaria Chen

    Hi Hyun,

    Sorry for the late reply. Could you provide more information on the issue, such as what command you ran and what message you got (in a screenshot)? It would also be helpful for us if you could provide the two models, but it wasn't convenient, could you give us each model's batch normalization and resolution so we could look more into it? Thank you for your help.

  • Andy HsiehAndy Hsieh

    @hyun

    Hi Hyun,

    I don't know what your "layer configurations" are referring to.

    Not sure where you see the "memory size too large" message either.

    If you see the message "memory size too large" when doing Batch Compile with Kneron Toolchain, it may be because your model input size has become larger.

    Additationly, please confirm that the "Hardware Supported Operators" are different according to the Kneron NPU platform you use.

    Hardware Supported Operators: https://doc.kneron.com/docs/#toolchain/appendix/operators/#_top

The discussion has been closed due to inactivity. To continue with the topic, please feel free to post a new discussion.

© Kneron, Inc.