tensorflow: TFlite got error output when enable hexagon delegate

Please make sure that this is a bug. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:bug_template

System information - Have I written custom code (as opposed to using a stock example script provided in TensorFlow): - OS Platform and Distribution ( Linux Ubuntu 16.04): - Mobile device (OPPO reno ace) if the issue happens on mobile device: - TensorFlow installed from (source or binary): - TensorFlow version (use command below): - Python version: - Bazel version (if compiling from source): - GCC/Compiler version (if compiling from source): - CUDA/cuDNN version: - GPU model and memory:

You can collect some of this information using our environment capture script You can also obtain the TensorFlow version with: 1. TF 1.0: python -c "import tensorflow as tf; print(tf.GIT_VERSION, tf.VERSION)" 2. TF 2.0: python -c "import tensorflow as tf; print(tf.version.GIT_VERSION, tf.version.VERSION)"

Describe the current behavior I use this model for testing: https://storage.googleapis.com/mirror.tensorflow.org/storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_0.75_quant_2018_06_29.zip I set all input values as 1(just for tesing), then print the first 50 values of output[0], see below picture, on the left is the values when running on CPU, on the right is the values when enable hexagon delegate,

CPU and DSP got diffrent output

Describe the expected behavior CPU and DSP should get the same output values

Code to reproduce the issue Provide a reproducible test case that is the bare minimum necessary to generate the problem.

Other info / logs Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.