how can I link tensorflow with tensorrt specific version?
suhyung-code42 opened this issue · 1 comments
I have installed various TensorRT versions. and below is my system.
(tf2.0) suhyung@suhyung:~$ dpkg -l | grep nvinfer
ii libnvinfer-bin 7.0.0-1+cuda10.0 amd64 TensorRT binaries
ii libnvinfer-dev 7.0.0-1+cuda10.0 amd64 TensorRT development libraries and headers
ii libnvinfer-dev-cross-aarch64 5.0.3-1+cuda10.0 all TensorRT cross development libraries and headers
ii libnvinfer-doc 7.0.0-1+cuda10.0 all TensorRT documentation
ii libnvinfer-plugin-dev 7.0.0-1+cuda10.0 amd64 TensorRT plugin libraries
ii libnvinfer-plugin7 7.0.0-1+cuda10.0 amd64 TensorRT plugin libraries
ii libnvinfer-samples 7.0.0-1+cuda10.0 all TensorRT samples
ii libnvinfer5 5.0.2-1+cuda10.0 amd64 TensorRT runtime libraries
ii libnvinfer5-cross-aarch64 5.0.3-1+cuda10.0 all TensorRT cross compilation runtime libraries
ii libnvinfer6 6.0.1-1+cuda10.0 amd64 TensorRT runtime libraries
ii libnvinfer7 7.0.0-1+cuda10.0 amd64 TensorRT runtime libraries
ii python3-libnvinfer 7.0.0-1+cuda10.0 amd64 Python 3 bindings for TensorRT
ii python3-libnvinfer-dev 7.0.0-1+cuda10.0 amd64 Python 3 development package for TensorRT
I am checking tensorrt version which is linked with tensorflow using below code
from tensorflow.compiler.tf2tensorrt.wrap_py_utils import get_linked_tensorrt_version
from tensorflow.compiler.tf2tensorrt.wrap_py_utils import get_loaded_tensorrt_version
import tensorflow
compiled_version = get_linked_tensorrt_version()
loaded_version = get_loaded_tensorrt_version()
tensorflow_version = tensorflow.version
print("Linked TensorRT version: %s" % str(compiled_version))
print("Loaded TensorRT version: %s" % str(loaded_version))
print("TensorFlow version: %s" %str(tensorflow_version))
and result is shown like below
Linked TensorRT version: (5, 1, 5)
Loaded TensorRT version: (5, 0, 3)
TensorFlow version: 2.0.0
`
I don't understand why 5.1.5 and 5.0.3 are shown.
This is the sequence of work I've done.
- install tensorrt 5.x
- install tensorrt 6.x
- install tensorrt 7.x
- make conda env (tf2.0)
- install tensorflow-gpu==2.0 on conda env
But as you see, tensorflow is linked with 5.x
How can I link with 7.0 ?
Additionally, I am using CUDA 10.0.
Thank you
If user want to use specific version library with tf, user should build install binary by self after setting configure.