search

koenvervloesem / tensorflow-addons-on-arm

Build TensorFlow Addons for ARM on the Raspberry Pi
MIT License
5 stars 0 forks source link

rpi 4b aarch64 build error #3

Open genius0182 opened 3 years ago

genius0182 commented 3 years ago

Build configurations successfully written to .bazelrc

Starting local Bazel server and connecting to it... WARNING: /root/.cache/bazel/_bazel_root/05fec5a36c83f24c9f73cdaccf3bbd7b/external/local_config_tf/BUILD:2012:1: target '_pywrap_tensorflow_internal.so' is both a rule and a file; please choose another name for the rule INFO: Analyzed target //:build_pip_pkg (37 packages loaded, 239 targets configured). INFO: Found 1 target... INFO: Deleting stale sandbox base /root/.cache/bazel/_bazel_root/05fec5a36c83f24c9f73cdaccf3bbd7b/sandbox ERROR: /opt/build_soft/tensorflow-addons-on-arm/addons-0.7.1/tensorflow_addons/custom_ops/layers/BUILD:7:1: Linking of rule '//tensorflow_addons/custom_ops/layers:_correlation_cost_ops.so' failed (Exit 1) /usr/bin/ld.gold: error: cannot find -l_pywrap_tensorflow_internal collect2: error: ld returned 1 exit status Target //:build_pip_pkg failed to build Use --verbose_failures to see the command lines of failed build steps. INFO: Elapsed time: 24.981s, Critical Path: 8.54s INFO: 0 processes. FAILED: Build did NOT complete successfully

koenvervloesem commented 3 years ago

That's understandable, as TensorFlow Addons on ARM (from the README's introduction) "builds TensorFlow Addons for the Raspberry Pi's ARMv7 processor architecture" and you need a 32-bit Raspberry Pi OS for it. You seem to be running it on 64-bit Raspberry Pi OS?

I actually do want to try to build it on aarch64 some time, but I'm not running any 64-bit Raspberry Pi OS machines at the moment, so it won't happen soon.

genius0182 commented 3 years ago

Yes,I running it on 64-bit Raspberry Pi OS. Do you 64-bit need to change anything?

koenvervloesem commented 3 years ago

I just haven't looked yet at building it for 64-bit ARM. You should find some help in the repository of TensorFlow Addons: https://github.com/tensorflow/addons

genius0182 commented 3 years ago

Inquiries are already being made. Thank you very much.

genius0182 commented 3 years ago

I used "ld -l_pywrap_tensorflow_internal --verbose" command to find the following file failed to open,an error is as follows `

attempt to open //usr/local/lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.so failed attempt to open //usr/local/lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.a failed attempt to open //lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.so failed attempt to open //lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.a failed attempt to open //usr/lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.so failed attempt to open //usr/lib/aarch64-linux-gnu/lib_pywrap_tensorflow_internal.a failed attempt to open //usr/local/lib/lib_pywrap_tensorflow_internal.so failed attempt to open //usr/local/lib/lib_pywrap_tensorflow_internal.a failed attempt to open //lib/lib_pywrap_tensorflow_internal.so failed attempt to open //lib/lib_pywrap_tensorflow_internal.a failed attempt to open //usr/lib/lib_pywrap_tensorflow_internal.so failed attempt to open //usr/lib/lib_pywrap_tensorflow_internal.a failed attempt to open //usr/aarch64-linux-gnu/lib/lib_pywrap_tensorflow_internal.so failed attempt to open //usr/aarch64-linux-gnu/lib/lib_pywrap_tensorflow_internal.a failed ld: cannot find -l_pywrap_tensorflow_internal ` image

I would like to ask how the file “lib_pywrap_tensorflow_internal.so” is generated？

koenvervloesem commented 3 years ago

I'm not sure, shouldn't this be installed with Q-engineering's TensorFlow package?

Have you tried this patch?

genius0182 commented 3 years ago

thank you