2.4 kB
TensorFlow.NET pack all required libraries in architecture-specific assemblies folders per NuGet standard [Deprecated] .
We changed to use Microsoft.ML.TensorFlow.Redist to maintain the TensorFlow library.
Download manually
Here are some pre-built TensorFlow binaries you can use for each platform:
- Linux
- Mac: https://storage.googleapis.com/tensorflow/libtensorflow/libtensorflow-cpu-darwin-x86_64-1.14.0.tar.gz
- Windows
Run in Linux
Install-Package TensorFlow.NET
Download Linux pre-built library and unzip libtensorflow.so and libtensorflow_framework.so into current running directory.
To run image recognition in Linux, please ensure some prerequisite libraries is install.
sudo apt install libc6-dev
sudo apt install libgdiplus
More information about System.Drawing on Linux.
Run in Mac OS
GPU Tensorflow for windows
Before running verify you installed CUDA and cuDNN
Build from source for Windows
https://www.tensorflow.org/install/source_windows
pacman -S git patch unzip
- Build static library
bazel build --config=opt //tensorflow:libtensorflow.so
- Build pip package
bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package
- Generate pip installation file
bazel-bin\tensorflow\tools\pip_package\build_pip_package C:/tmp/tensorflow_pkg
- Install from local wheel file.
pip install C:/tmp/tensorflow_pkg/tensorflow-1.14.0-cp36-cp36m-win_amd64.whl
Export more APIs
Add more api to c_api.h
TF_CAPI_EXPORT extern void AddControlInput(TF_Graph* graph, TF_Operation* op, TF_Operation* input);
TF_CAPI_EXPORT extern void UpdateEdge(TF_Graph* graph, TF_Output new_src, TF_Input dst, TF_Status* status);
TF_CAPI_EXPORT extern void RemoveAllControlInputs(TF_Graph* graph, TF_Operation* op);