can not run with precision_mode="INT8"

NVIDIA-AI-IOT / tf_trt_models

TensorFlow models accelerated with NVIDIA TensorRT

BSD 3-Clause "New" or "Revised" License

684 stars 244 forks source link

can not run with precision_mode="INT8" #37

Open guapizyq opened 5 years ago

guapizyq commented 5 years ago

error is fillowing Node Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/add should have an input named 'Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/add/y' but it is not available