Path and file IO changes to make the scripts work on my machine

[dbdean]

I've had to make a few changes to the file IO paths and whether folders can be created or deleted to get these scripts to work on my machine.

[andrewljohnson]

Thanks, I appreciate it.

I'll run your changes on my box next week and verify, then accept the PR.

[dbdean]

@andrewljohnson, have you had a chance to review and/or verify this pull request?

[andrewljohnson]

I have been working on this today, but I found issues with the Dockerfile I was trying to fix. I was surprised you got this far without changing the Dockerfile.

On Sun, Jan 29, 2017 at 6:45 PM, David Dean notifications@github.com wrote:

@andrewljohnson https://github.com/andrewljohnson, have you had a chance to review and/or verify this pull request?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#issuecomment-275969688, or mute the thread https://github.com/notifications/unsubscribe-auth/AAbzf-dQH-3mLOK-DFSK_0ZA97Wj8qcxks5rXU7TgaJpZM4LosLO .

[dbdean]

I didn't have any problems with the Dockerfile last time I built it, at least on the master branch. I'm removing my cache and trying again now to see if anything breaks for me.

[andrewljohnson]

I had to make this change to let "make dev" work (non-cpu):

https://github.com/trailbehind/DeepOSM/commit/38cb5391d3a3853505fb0ba0bf87743d257fc5db

And now I'm working on "make dev-gpu" where I ran into this issue late in the install:

The command '/bin/sh -c apt-get update && apt-get install -y --no-install-recommends --force-yes cuda-core-$CUDA_PKG_VERSION cuda-misc-headers-$CUDA_PKG_VERSION cuda-command-line-tools-$CUDA_PKG_VERSION cuda-license-$CUDA_PKG_VERSION cuda-nvrtc-dev-$CUDA_PKG_VERSION cuda-cusolver-dev-$CUDA_PKG_VERSION cuda-cublas-dev-$CUDA_PKG_VERSION cuda-cufft-dev-$CUDA_PKG_VERSION cuda-curand-dev-$CUDA_PKG_VERSION cuda-cusparse-dev-$CUDA_PKG_VERSION cuda-npp-dev-$CUDA_PKG_VERSION cuda-cudart-dev-$CUDA_PKG_VERSION cuda-driver-dev-$CUDA_PKG_VERSION && cd /tmp && apt-get download gpu-deployment-kit && rm -rf /var/lib/apt/lists/*' returned a non-zero code: 100 Makefile:28: recipe for target 'dev-gpu' failed make: *** [dev-gpu] Error 1

Andrew Johnson Founder gaiagps.com http://gaiagps.com

On Sun, Jan 29, 2017 at 6:59 PM, David Dean notifications@github.com wrote:

I didn't have any problems with the Dockerfile last time I built it. I'm removing my cache and trying again now to see if anything breaks for me.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#issuecomment-275970923, or mute the thread https://github.com/notifications/unsubscribe-auth/AAbzf5M_fj1rtM-L7G3l1xG7O60nyPdvks5rXVITgaJpZM4LosLO .

[dbdean]

Hmm, I'm not working on the feature-analysis branch (or GPU stuff at the moment). I'm in master.

I'll try and build on that branch then, see what I get.

[andrewljohnson]

I don't think the branch is relevant though, since the Dockerfiles haven't changed. This branch was just where I left off, I figured I should merge it and then integrate your change if needed.

I may have solved this now too, waiting for the build to run right now...

Andrew Johnson Founder gaiagps.com http://gaiagps.com

On Sun, Jan 29, 2017 at 7:10 PM, David Dean notifications@github.com wrote:

Hmm, I'm not working on the feature-analysis branch (or GPU stuff at the moment). I'm in master.

I'll try and build on that branch then, see what I get.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#issuecomment-275971814, or mute the thread https://github.com/notifications/unsubscribe-auth/AAbzf9LXdZJAO25QXb1-MgMWNDUrI70pks5rXVSKgaJpZM4LosLO .

[dbdean]

I'll wait and see how you go, then :+1:

[dbdean]

The shutil.rmtree one is because on first run the cache directories aren't there yet, and trying to remove them when they don't already exists raises an exception.

The try/except around makedirs is because makedirs raises an exception if the directory already exists (in this case, the parent of the cache_filename).

[dbdean]

Thanks!

On Mon., 30 Jan. 2017, 14:37 Andrew L. Johnson, notifications@github.com wrote:

Merged #73 https://github.com/trailbehind/DeepOSM/pull/73.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#event-940824442, or mute the thread https://github.com/notifications/unsubscribe-auth/AMIsPkLUzxJU-lTpzsmqU96c-Yz-0QV3ks5rXWkBgaJpZM4LosLO .

-- http://dbdean.com

[andrewljohnson]

Likewise!

On Sun, Jan 29, 2017 at 8:41 PM, David Dean notifications@github.com wrote:

Thanks!

On Mon., 30 Jan. 2017, 14:37 Andrew L. Johnson, notifications@github.com wrote:

Merged #73 https://github.com/trailbehind/DeepOSM/pull/73.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#event-940824442, or mute the thread https://github.com/notifications/unsubscribe- auth/AMIsPkLUzxJU-lTpzsmqU96c-Yz-0QV3ks5rXWkBgaJpZM4LosLO .

-- http://dbdean.com

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/trailbehind/DeepOSM/pull/73#issuecomment-275978938, or mute the thread https://github.com/notifications/unsubscribe-auth/AAbzf9AjhvFIpItyjb2p9yYkOaVOEOx7ks5rXWnhgaJpZM4LosLO .