Loading
Feedback

banjtheman 109

Badges

0
0
1

Activity

Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Mon
Wed
Fri

Ratings Progression

Loading...

Challenge Categories

Loading...

Challenges Entered

A new benchmark for Artificial Intelligence (AI) research in Reinforcement Learning

Latest submissions

See All
graded 4892
graded 4884
graded 4875
Gold 0
Silver 0
Bronze 1
Trustable
May 16, 2020

Badges

  • Has filled their profile page
    May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020
Participant Rating
Participant Rating
banjtheman has not joined any teams yet...

Unity Obstacle Tower Challenge

GitHub repo gone?

About 2 years ago

Hmm did the GitHub repo just go dark?
https://github.com/Unity-Technologies/obstacle-tower-challenge

getting 404 for all Unity-Technologies repos

How to claim Round 1 Prize?

About 2 years ago

Just got email with prize details

Submissions Q&A

Over 2 years ago

@arthurj
The core issue is that if any agent is importing any large libraries (rainbow, tensorflow, dopamine, etc) before the agent can load the environment the test will never be started no matter how long the timeout_wait parameter is.

The easiest way would be to add a delay to the environment container starting up to give time for the agent to load, without referring to hacky methods such as deferring imports.

In local docker testing the agent container must always be ready before the environment container starts as well.

[Announcement][git error] Large Model Weights in your Submission Repository

Over 2 years ago

@mohanty huzzah its going through now, go and enjoy that beer

[Announcement][git error] Large Model Weights in your Submission Repository

Over 2 years ago

@mohanty @shivam ok after making my build fail I noticed the commands the server is running is not running the correct commands

https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/9

Currently

git clone git@gitlab.aicrowd.com:USER/obstacle-tower-challenge.git
cd obstacle-tower-challenge
git checkout lfs

Which just makes a new detached branch and doesn’t pull down files
The server should use git lfs pull instead to pull down LFS files

git clone git@gitlab.aicrowd.com:USER/obstacle-tower-challenge.git
cd obstacle-tower-challenge
git lfs pull

this is a better and easier solution then the one I had before

[Announcement][git error] Large Model Weights in your Submission Repository

Over 2 years ago

@ChenKuanSun I found it easier to just delete the repo and start fresh and then use git lfs track before you add the file to the repo.

git init
git remote add YOUR_URL
git lfs install
git lfs track file
git add .
git commit -am “initial commit”
git tag -am “test” tag
git push -u origin master tag

[Announcement][git error] Large Model Weights in your Submission Repository

Over 2 years ago

@mohanty ok getting close created fresh repo with out the $store$ files and confirmed that the image runs the evaluation locally after some minor updates, but getting weird error on build server

DataLossError (see above for traceback): Checksum does not match: stored 3961381255 vs. calculated on the restored bytes 87951365

https://gitlab.aicrowd.com/banjtheman/fresh-obs/issues/2

Evalutation error : Unity environment took too long to respond

Over 2 years ago

Was able to get around this by using the defer import strategy so the mlagent text came 1 second after startup, bit hacky but seems to be only way to get test to run.

2019-03-13T14:16:19.307396414Z root
2019-03-13T14:16:20.443927313Z INFO:mlagents_envs:Start training by pressing the Play button in the Unity Editor.

[Announcement][git error] Large Model Weights in your Submission Repository

Over 2 years ago

@mohanty the git lfs files are being added to gitlab, but i don’t think the build server is pulling the lfs files.

getting this error

2019-03-13T14:16:44.691582001Z     raise OSError('Not a gzipped file (%r)' % magic)
2019-03-13T14:16:44.691735028Z OSError: Not a gzipped file (b've')

https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/2

When i clone repo with git clone, I also dont get lfs files

Evalutation error : Unity environment took too long to respond

Over 2 years ago

@mohanty yea, I’ve increased mine to 600 (even 30000 once) but all that does is keep the agent container idle, it looks like if the environment container starts before the agent displays

 INFO:mlagents_envs:Start training by pressing the Play button in the Unity Editor.

The test never starts, and this is easily reproducible on a local environment

Evalutation error : Unity environment took too long to respond

Over 2 years ago

@shivam seems to be happening again

2019-03-12T02:22:06.918029502Z root
2019-03-12T02:22:17.32389638Z INFO:mlagents_envs:Start training by pressing the Play button in the Unity Editor.

https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/9

The ~10 second startup time seems to be the issue as noted here Starter kit stuck "pending" state for a day

Is it possible to delay environment container say 60 seconds after agent container starts?

Is there a max repo size?

Over 2 years ago

Seeing similar issue, suspect its the same thing happens right when agent tries to load checkpoints

https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/12

Evalutation error : Unity environment took too long to respond

Over 2 years ago

@arthurj so i increased my timeout wait to 30000, and same result notice time stamps on my logs

2019-03-07T13:25:24.799545869Z root
....
2019-03-07T21:45:37.653813452Z    "The Unity environment took too long to respond. Make sure that :\n"
2019-03-07T21:45:37.653925272Z mlagents_envs.exception.UnityTimeOutException: The Unity environment took too long to respond. Make sure that :
2019-03-07T21:45:37.653939834Z 	 The environment does not need user interaction to launch
2019-03-07T21:45:37.653945019Z 	 The Academy and the External Brain(s) are attached to objects in the Scene
2019-03-07T21:45:37.653949289Z 	 The environment and the Python interface have compatible versions.
2019-03-07T21:45:37.653953562Z   In call to configurable 'create_otc_environment' (<function create_otc_environment at 0x7f7dfdfcaf28>)
2019-03-07T21:45:37.65395818Z   In call to configurable 'Runner' (<function Runner.__init__ at 0x7f7dfdfca158>)
2019-03-07T21:45:37.653962707Z   In call to configurable 'create_runner' (<function create_runner at 0x7f7dfdfa9f28>)

full logs https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/11

Is there a max repo size?

Over 2 years ago

Ha yup same issue just now, looks like the server’s docker volume might be full

Submission Failed: Evaluation Error

Over 2 years ago

@shivam had similar issue as well and could use logs for guidance…
https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/4

Also got another weird issue on a different submit

    Submission failed : uninitialized constant Api::ExternalGradersController::TermsNotAcceptedByParticipant

https://gitlab.aicrowd.com/banjtheman/obstacle-tower-challenge/issues/5

Is there a max repo size?

Over 2 years ago

Great the image was able to build successfully now

Is there a max repo size?

Over 2 years ago

So my repo size is 4.4GB under the 10GB limit stated there, my code is in gitlab so the problem is not there.

The problem seems to be in the build process in the repo2docker phase. The script works fine on my local box, but there seems to be some file size issue on their build server. Would like some guidance on if there is away to remedy that issue.

Is there a max repo size?

Over 2 years ago

Hey, I was able to submit my build for evaluation, however my model weights are pretty large(4.4GBs) and it seems the build server is out of space? Here is the error I’m getting…

Step 27/33 : COPY src/ ${HOME}
failed to copy files: failed to copy directory: Error processing tar file(exit status 1): write /checkpoints/$store$_observation_ckpt.100.gz: no space left on device

Is there a max repo size or a recommend way to add large files to our docker image?

banjtheman has not provided any information yet.

Notebooks

Create Notebook