INT8 calibration file not specified/accessible. INT8 calibration can be done through setDynamicRange API in 'NvDsInferCreateNetwork' implementation

I have dowloaded the quantized pruned peoplenet-resnet34 model from ngc website. And, I set the tlt-encoded-model to its path but I am getting the above error. Can someone explain why I am getting this error and how can I fix it?

deepstream_basic.txt (2.0 KB)
model_basic.txt (930 Bytes)

• GTX1660
• DeepStream Version 5.1

Hey, what’s the error?

[INT8 calibration file not specified/accessible. INT8 calibration can be done through setDynamicRange API in ‘NvDsInferCreateNetwork’ implementation](https://forums.developer.nvidia.com/t/int8-calibration-file-not-specified-accessible-int8-calibration-can-be-done-through-setdynamicrange-api-in-nvdsinfercreatenetwork-implementation/182719)

Can you check if you provide a int8 calibration file?

How do I provide that?

refer Gst-nvinfer — DeepStream 5.1 Release documentation , check int8-calib-file

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.