TensorFlow and Go on Raspberry Pi
This is a guide which I ran through for building libtensorflow
on Raspberry Pi.
0. Used Hardwares and Softwares
All steps were taken on my Raspberry Pi 3 B model with:
- Minimum GPU memory allocated (16MB)
- 1GB of swap memory
- External USB HDD (as root partition)
and software versions were:
- Raspbian (Stretch) / gcc 6.3.0
- Tensorflow 1.3.0
- Protobuf 3.1.0
- Bazel 0.5.1
Before the beginning, I had to install dependencies:
for protobuf
for bazel
1. Install Protobuf
I cloned the protobuf’s repository:
and started building:
It took less than an hour to finish.
I could see the version of installed protobuf with:
2. Install Bazel
a. download
I got a zip file of bazel from here and unzipped it:
b. edit bootstrap files
In the unzipped directory, I opened the scripts/bootstrap/compile.sh
file:
searched for lines that looked like following:
and appended -J-Xmx500M
to the last line so that the whole lines would look like:
It was for enlarging the max heap size of Java.
c. compile
After that, started building with:
It also took about an hour.
d. install
After the compilation had finished, I could find the compiled binary in output
directory.
Copied it into /usr/local/bin
directory:
3. Build libtensorflow.so
(I referenced this document for following processes)
a. download
Got the tensorflow go code with:
b. edit files
In the downloaded directory, I checked out the latest tag and replaced lib64
to lib
in the files with:
Raspberry Pi still runs on 32bit OS, so they had to be changed like this.
After that, I commented #define IS_MOBILE_PLATFORM
out in tensorflow/core/platform/platform.h
:
If it is not commented out, bazel will build for mobile platforms like iOS
or Android
, not Raspberry Pi.
To do this easily, just run:
Finally, it was time to configure and build tensorflow.
c. configure and build
I had to answer to some questions here.
Then I started building libtensorflow.so
with:
My Pi became unresponsive many times during this process, but I kept it going on.
d. install
After a long time of struggle, (it took nearly 7 hours for me!)
I finally got libtensorflow.so
compiled in bazel-bin/tensorflow/
.
So I copied it into /usr/local/lib/
:
All done. Time to test!
4. Go Test
I ran a test for validating the installation:
then I could see:
Ok, it works!
Edit: As this instruction says, I had to regenerate operations before the test:
5. Further Test
I wanted to see a simple go program running, so I wrote this code:
and ran it with go run sample.go
:
See the result?
From now on, I can write tensorflow applications in go, on Raspberry Pi! :-)
98. Trouble shooting
Build failure due to a problem with Eigen
Back in the day with Tensorflow 1.2.0, I encountered this issue while building, but it’s still not fixed yet in 1.3.0.
So I had to work around this problem again by editing tensorflow/workspace.bzl
from:
to:
and starting again from the beginning:
Then I could build it without further problems.
I hope it would be fixed on future releases.
99. Wrap-up
Installing TensorFlow on Raspberry Pi is not easy yet. (There’s a kind project which makes it super easy though!)
Installing libtensorflow.so
is a lot more difficult, because it takes too much time to build it.
But it is worth trying; managing TensorFlow graphs in golang will be handy for people who don’t love python - just like me.
999. If you need one,
You don’t have time to build it yourself, but still need the compiled file?
Good, take it here.
I cannot promise, but will try keeping it up-to-date whenever a newer version of tensorflow comes out.