Swift for Tensorflow Installation Guide

(Vibhutha Kumarage) #1

Hello Everyone,
I wrote this blog post as a Guide to install Swift-TF on your ubuntu local machine or VM
https://vibhuthasak.github.io/posts/swift-tensorflow-guide/

3 Likes

#2

Thanks VIbhutha! These instructions got me up and running.

However, one thing I noticed is that when I run swift, instead of seeing what you say, I also see these errors on the console:

alexis@topobigbox:~/Desktop/swifttoolchain_downloads$ swift
error: ld-2.27.so 0xffffffff0005c564: adding range [0x1464a-0x146ba) which has a base that is less than the function's low PC 0x14dc0. Please file a bug and attach the file at the start of this error message
error: ld-2.27.so 0xffffffff0005c564: adding range [0x146d0-0x146d6) which has a base that is less than the function's low PC 0x14dc0. Please file a bug and attach the file at the start of this error message
error: ld-2.27.so 0xffffffff0005c5c5: adding range [0x1464a-0x146ba) which has a base that is less than the function's low PC 0x14dc0. Please file a bug and attach the file at the start of this error message
error: ld-2.27.so 0xffffffff0005c5c5: adding range [0x146d0-0x146d6) which has a base that is less than the function's low PC 0x14dc0. Please file a bug and attach the file at the start of this error message
Welcome to Swift version 5.0-dev (LLVM dcb9eb74a7, Clang 95cdf7c9af, Swift 434442b530).
Type :help for assistance.
  1>

I am curious if you ever saw this error as well, because I believe I have exactly the same configuration as you (full version numbers are https://gist.github.com/algal/09ea60872f9e5aa1cf54882a2b0ef4b9 ).

1 Like

(Pedro Cuenca) #3

I get the same errors in the console, even though everything seems to be working fine. I used the nightly Ubuntu + Cuda toolchain.

0 Likes

(tyoc213) #4

Well, I get that error from the CPU version (which seems to work), but have tried to compile or run the niglty or thde 0.2 version on a Ubuntu 18.04 and I get error: failed to stop process at REPL breakpoint

And installing docker-cuda and running docker run -u $(id -u):$(id -g) -p 8888:8888 tensorflow/tensorflow:latest-gpu-jupyter

I get

0 Likes

(Fred Monroe) #5

I get this same error as well

error: ld-2.27.so 0xffffffff0005c564 etc....
0 Likes

(Edwin) #6

I get the same error.

1 Like

(Edwin) #7

Update: Reinstalled CUDA and Cudnn, start error ended up being resolved.

2 Likes