Changes to webapp to allow debugging of upload problems and providing feedback out of sequence #39
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… of sequence,
- add media UUID to allow checking upload completion using GCP console
- changed order_by from "-created_at" to Comparision.order_by("+id") for comparisons because learning phase was uploading media using Comparison.order_by("+created_at")
…server - skips upload Google Cloud Platform add nav for browsing comparisons
|
I added a new args param to serve videos from the local server. By skipping the Google Cloud uploads, I was able to complete a training run with 600 feedbacks before my video rendering processes died. |
…learner separately; continue training reward predictor after loading weights; and server content locally (skip GCP) —load_predictor_weights [walker-splits-3-20_weights.150.h5] —train_predictor [0,1] —load_policy_weights [walker-splits-4-0.policy_weights.120.ckpt] --content_server [local] E.g. python rl_teacher/teach.py -p human -e $RL_ENV -n $RL_NAME-$RL_LABELS --content_server local --load_predictor_weights walker-splits-3-20_weights.150.h5 --load_policy_weights walker-splits-4-0.policy_weights.120.ckpt --train_predictor 1
|
Hey @mixuala, I'd like to keep this version of the code simple and not add new features to it. However, I'd bet that many people would find your code useful to them. Perhaps you could put together a description of the fork that you're working on, and then we can link to it to the Extensions section of the README |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I'm still having problems getting the agent to complete a learning task from human feedback. right now I'm stuck on #38 which keeps me from offering feedback after about 3000 sec.
But I made a few changes to the webapp to make it easier to debug uploads from the webapp, and also to provide feedback out of sequence.
/experiments/[name]/[comparison_id]for offering feedback out of sequence,