sungbinchoi 201

Evaluation server hardware specifications

Is GPU available on evaluation server?
I think GPU is necessary if we want to experiment with global observation.

View post 0

The evaluation result does not match my local testing

My code is also running on TF 1.13.1.
But TF version 1.13.1 is specified on my environment.yml file.
I’m still confused about what went wrong.

View post 0

The evaluation result does not match my local testing

I have a same problem.
Evaluation successful but only 1 episode with 4 steps.

View post 0

Sample efficiency and human dataset

Let me ask another question about MINERL_TRAINING_MAX_INSTANCES = 5 in competition_submission_starter_template. I’m not quite sure how INSTANCE is defined.
Does it mean that max 5 gym env could be run in parallel while training?
Or max 5 parallel trainer perform update function independently and somehow merge gradients periodically?
Or max 5 processes (regardless of doing what kind of job) can be deployed while training?

View post 0

Evaluation successful but actually it is not

Could you look into this issue?
https://gitlab.aicrowd.com/sungbinchoi/competition_submission_starter_template/issues/7

Evaluation successful
but there’re only 1 episode with 4 step run.
I think something went wrong while evaluating.

Instance-1 (Reward: 0.0, Steps: 4)
Episode-1
State : Episode Complete
Reward : 0.000
Steps : 4

View post 0

Please send me evaluation error log on gitlab

My previous evaluation error has been solved now.

But could you look into this issue?
https://gitlab.aicrowd.com/sungbinchoi/competition_submission_starter_template/issues/7

Evaluation successful
but there’re only 1 episode with 4 step run.
I think something went wrong while evaluating.

Instance-1 (Reward: 0.0, Steps: 4)
Episode-1
State : Episode Complete
Reward : 0.000
Steps : 4

View post 0

Evaluation error on log on MineRL task

Hello.

I got evaluation error on MineRL task.

https://gitlab.aicrowd.com/sungbinchoi/competition_submission_starter_template/issues/3

Could you send me evaluation error log to me?

My email address is
sungbin.choi.1@gmail.com

View post 0

Please send me evaluation error log on gitlab

Hello.

I got evaluation error on MineRL task.

https://gitlab.aicrowd.com/sungbinchoi/competition_submission_starter_template/issues/3

Could you send me evaluation error log to me?

My email address is
sungbin.choi.1@gmail.com

View post 0

Is submission system stuck?

Hello. I’m Sungbin Choi. I’m participating at Unity Obstacle Tower Challenge.
Usually before, it took less than an hour to get evaluation results,
but today now it’s been more than 3 hours and I still cannot get evaluation results back.
I need to see evaluation results to choose and submit best one at last.
Since today is last day of the challenge, I’m worried I cannot choose one.

View post 0

How can I get my evaluation error log?

Oh, now I found the problem. It was about file permission issue.
Thanks~

View post 1

How can I get my evaluation error log?

Hello. I’m SungbinChoi participating at Unity Obstacle Tower Challenge.
I tried to submit a few runs today but evaluation error occurred every time.
I want to see error logs ASAP to fix the problem.
https://gitlab.aicrowd.com/sungbinchoi/obstacle-tower-challenge

I’m also wondering what should I do if I encounter evaluation error again.
Do I need to send direct message to request error log again or is there any other way around?

View post 0

Submissions are stuck

I think it’s stuck again.

View post 0

Possible bug with tower-seed in env 2.0

Yes, I guess the bug is not fixed yet.
I’ll post a comment on that github issue link.
->
Now I got the answer.
They said tower-seed value range is intentionally limited to 100 in Round 2.

View post 1

Possible bug with tower-seed in env 2.0

Although I defined specific tower-seed when I reset the environment,
random seed is assigned instead
when tower-seed >= 100.
It looks like env 2.1 has the same problem.

View post 0