Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Nishimura #1160

Open
wants to merge 47 commits into
base: master
Choose a base branch
from
Open

Nishimura #1160

wants to merge 47 commits into from

Conversation

NishiYusuke
Copy link

変更箇所には, コメントで nishimura と記載してあるので, それで検索すれば見つかると思います.
詳細は以下の通り.

gym-grasp/gym-grasp/envs/hand/grasp_block.py : 報酬関数部分の変更
baseline/her/rollout.py : 成功把持姿勢を npy ファイルで保存するように変更
めっちゃ雑な実装ですが, 良い感じに編集等してもらえると幸いです.
どうぞよろしくお願いします.

ToDo

  • 成功把持姿勢保存時に, オブジェクトIDとエピソード数なども一緒に保存
  • ちゃんと成功把持姿勢が保存されているか確認

getty708 and others added 30 commits November 22, 2018 18:47
Qtable: 中間出力の吐き出し
Chainer: 投射モデルの学習コード
install gym_grasp by pip install -e .
fix import path from dist_package.gym to gym_grasp
The number of action space of grasp block task is 21, not 20.
KazukiHigashi and others added 17 commits January 7, 2019 23:40
Just importing gym_grasp, we can use the 'GraspBlock-v0' environment.
When simulator slide up the hand, its position is vibration, not converge. But sometimes success the task. So I'll share this tentative version. I will upload the stable verion later.
tentative ver. of partially working RL
Chainer: Testコード変更 & 中間出力の獲得
This commit make the model works properly.
However the convergence of this model become slow.
adjust physics parameters for working of RL
environment tuning for learning well
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants