Error during building Deep Speech binaries with native client

Deep Speech
#1

problem:*
During building binaries with native client according to https://github.com/mozilla/DeepSpeech/blob/v0.5.1/native_client/README.md following errors are got.
iiit_admin@gpu-sys-5038a-i:~/DeepSpeech-0.5.1/DeepSpeech-0.5.1/tensorflow$ bazel build --config=monolithic --config=cuda -c opt --copt=-O3 --copt="-D_GLIBCXX_USE_CXX11_ABI=0" --copt=-fvisibility=hidden //native_client:libdeepspeech.so //native_client:generate_trie
WARNING: The following configs were expanded more than once: [cuda]. For repeatable flags, repeats are counted twice and may lead to unexpected behavior.
DEBUG: /home/iiit_admin/.cache/bazel/_bazel_iiit_admin/a351f61158fc8ce2a298ac1a721f4723/external/bazel_tools/tools/cpp/lib_cc_configure.bzl:115:5:
Auto-Configuration Warning: ‘TMP’ environment variable is not set, using ‘C:\Windows\Temp’ as default
INFO: Analysed 2 targets (0 packages loaded, 0 targets configured).
INFO: Found 2 targets…
ERROR: /home/iiit_admin/DeepSpeech-0.5.1/DeepSpeech-0.5.1/tensorflow/native_client/BUILD:16:1: Executing genrule //native_client:ds_git_version failed (Exit 1)
INFO: Elapsed time: 3.963s, Critical Path: 3.59s, Remote (0.00% of the time): [queue: 0.00%, setup: 0.00%, process: 0.00%]
INFO: 41 processes: 41 local.
FAILED: Build did NOT complete successfully

System details
OS Platform :Ubuntu 18.04
TensorFlow installed from (source or binary): source
TensorFlow version: Mozilla’s tensorflow r1.13
Python version: 3.7.1
Installed using virtualenv? pip? conda?: virtual env
Bazel version (if compiling from source): 0.19.2
CUDA/cuDNN version: 10.0.130
GPU model and memory: GeForce GTX 1080 2 GPU cards
Deepspeech -v 0.5.1

Please suggest.Thanks in advance.

#2

You need to investigate what is happening. This is the script being ran: https://github.com/mozilla/DeepSpeech/blob/v0.5.1/native_client/ds_git_version.sh

Maybe you lack realpath ?

#3

@Shruthi_Sridhar Also, may I ask why you need to rebuild ? You seem to use a setup where prebuilt binaries works well.

#4

Thanks for the reply.I am checking for the prebuilt native client binaries in https://github.com/mozilla/DeepSpeech/releases/tag/v0.5.1. I could not find native client for linux platform gpu version. please suggest which one to use…

#5

By Real path ,does it mean path of tensorflow and Deepspeech directories?

#6

No, the realpath binary.

#7

This ? https://github.com/mozilla/DeepSpeech/releases/download/v0.5.1/native_client.amd64.cuda.linux.tar.xz

#8

I am using this processor : Intel® Xeon® CPU E5-2620 v4 @ 2.10GHz × 16
Graphics : GeForce GTX 1080 Ti/PCIe/SSE2. So is the binary native client amd64 fine to use?

#9

I don’t get the question, it’s a 64-bits x86 CPU, so it’s fine …