question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Windows: build TensorFlow Lite with Bazel on Windows

See original GitHub issue

This is a tracking bug for building TF Lite on Windows with Bazel.

  • remove -Wno-implicit-fallthrough from @tensorflow//third_party/flatbuffers/flatbuffers.BUILD
  • remove single quotes from repro command’s --cxxflags
  • update README.md: use drive root as --output_user_root (long paths)
  • update README.md: build with “-c opt” (long paths)
  • edit WindowsJniLoader to load from bazel_tools, not from <reponame>/external/bazel_tools (must test on Android BusyBox)
  • create src/main/native/windows/BUILD.tools
  • add src/main/native/windows/*.{cc,h} to //src/main/native/windows:embedded_tools
  • Java SingleJar doesn’t support --nocompress_suffixes, and some build rule requires that --> migrate to C++ SingleJar on Windows (#2241)

Issue Analytics

  • State:closed
  • Created 6 years ago
  • Comments:15 (15 by maintainers)

github_iconTop GitHub Comments

9reactions
meteorcloudycommented, May 16, 2018

I think we can drop the priority to P2. I talked to people working on TensorFlow lite, they don’t seem to need this urgently. We can reconsider the priority after TF completely moves to Bazel on Windows, which is what I’m working on right now.

1reaction
meteorcloudycommented, Mar 4, 2019

I finally figured out why.

Looks like for TensorFlow Lite we should not run the ./configure script. The generated .tf_configure.bazelrc file contains flags that would cause the build to fail.

Can you re-try after deleting .tf_configure.bazelrc file?

Update: Problematic flags are:

--copt=-DWIN32_LEAN_AND_MEAN --host_copt=-DWIN32_LEAN_AND_MEAN
--distinct_host_configuration=false
Read more comments on GitHub >

github_iconTop Results From Across the Web

Build from source on Windows - Install - TensorFlow
Install Bazel, the build tool used to compile TensorFlow. For Bazel version, see the tested build configurations for Windows. Configure Bazel to build...
Read more >
How to build and install TensorFlow GPU/CPU for Windows ...
This is a tutorial how to build TensorFlow v1.10 with GPU (NVIDIA CUDA 9.2 ... Install Git for Windows; Install Bazel; Install MSYS2...
Read more >
Where do I find a step-by-step tutorial on how to build TF Lite ...
The build produced the following output: C:\tensorflow\tensorflow\bazel-out\x64_windows-opt\bin\tensorflow\lite: 07/30/2020 16:10 35,904 allocation.lib.
Read more >
How do I compile a TensorFlow Lite GPU delegate for Windows
Even though TFLite does not support the Windows platform there is a clear indication that it should be possible to compile a GPU...
Read more >
Trying to building tensorflow from source on Windows - YouTube
I looked into building tensorflow from source on Windows. The reason, TensorRT released and I wanted to test the new functionallity.
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found