TensorFlow (TF) can be built from source easily and installed as a Python wheel package. I used the following steps to build it using Python3 and with support for CUDA and TensorRT:
- Install Python3 pre-requisites:
$ sudo apt install python3-dev python3-pip
- Install necessary Python3 packages locally:
$ pip3 install -U --user six numpy wheel setuptools mock $ pip3 install -U --user keras_applications==1.0.6 --no-deps $ pip3 install -U --user keras_preprocessing==1.0.5 --no-deps
These packages are installed to your
~/.local/lib/python3.x/site-packages directory. TF documentation also installs the latest pip3 from PyPI. However, doing that causes the infamous “Cannot import name main” error, so I do not do that.
- TF uses Bazel as its build tool. Install it as described here. I ended up placing its binary in my
~/bin. Since my
~/binis in my
PATH, the Bazel binary can be executed from any place.
I recommend creating a
tensorflow_rootdirectory. This is because the TF packaging tends to write out to a location outside the TF source directory. Also, TF needs to access other libraries. So this root directory makes it easy to create all TF related directories under one umbrella.
Clone the TF Git repository inside the root directory:
$ cd tensorflow_root $ git clone firstname.lastname@example.org:tensorflow/tensorflow.git $ cd tensorflow
- Configure the build process using:
Some of the questions it asks and my replies:
- Please specify the location of python.
- Please input the desired Python library path to use.
- Enable: XLA JIT, CUDA and TensorRT. Be careful, TF might not work with latest versions of CUDA, cuDNN and TensorRT. I used CUDA 10.0 and cuDNN 7.3 and TensorRT 5.0.
- Did not enable: OpenCL SYCL and ROCm.
- It is time to build TF. The command to build is:
$ bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package
The first build can take 2-4 hours to complete.
- Now we are ready to package the build artifacts into a Python package. Specify where you want that package to be placed in the packaging command:
$ ./bazel-bin/tensorflow/tools/pip_package/build_pip_package /path_to_tensorflow_root/tensorflow_pkg
I found that it had generated a Python wheel file named:
- We are ready to install TF. Make sure you remove older versions of TF:
$ pip3 uninstall tensorflow tensorflow-estimator
- Install TF from the Python wheel package:
$ pip3 install -U --user tensorflow-1.13.1-cp36-cp36m-linux_x86_64.whl
- Check if TF is installed correctly:
$ python3 -c "import tensorflow"
Tried with: Tensorflow 1.13.1 and Ubuntu 18.04