Hello
I am trying to run a quantization model on TI EVM Jacinto 7. According to the documentation (https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/07_00_00_11/exports/docs/tidl_j7_01_02_00_09/ti_dl/docs/user_guide_html/md_tidl_model_import.html) we can quantize a model but cannot run it on the EVM as this is not supported on the current release (In Model Format-version, point-3). We tried this as well and we were able to create a quantized model but we couldn't perform inference on it. But the cloud EVM which is provided through the EdgeAI platform does support inference using a quantized model.
Can anyone clarify why EdgeAi cloud EVM support inference on a quantized model and the same thing is not supported on in-house EVM?
