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.

Sign up for GitHub

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

improve test #279

Merged
merged 3 commits into from
Dec 12, 2023
Merged

improve test #279

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
48edf32
update readme
huangshiyu13 Dec 12, 2023
c1a6f56
Merge branch 'main' of github.com:OpenRL-Lab/openrl
huangshiyu13 Dec 12, 2023
e29fd90
improve test
huangshiyu13 Dec 12, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions Project.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ However, in many practical applications, it is important to develop reasonable a
In this paper, we propose an on-policy framework for discovering multiple strategies for the same task.
Experimental results show that our method efficiently finds diverse strategies in a wide variety of reinforcement learning tasks.

- Paper: [DGPO: Discovering Multiple Strategies with Diversity-Guided Policy Optimization](https://arxiv.org/abs/2207.05631)(AAMAS Extended Abstract 2023)
- Authors: Wenze Chen, Shiyu Huang, Yuan Chiang, Ting Chen, Jun Zhu
- Paper: [DGPO: Discovering Multiple Strategies with Diversity-Guided Policy Optimization](https://arxiv.org/abs/2207.05631)(AAAAI 2024)
- Authors: Wenze Chen, Shiyu Huang, Yuan Chiang, Tim Pearce, Wei-Wei Tu, Ting Chen, Jun Zhu


92 changes: 46 additions & 46 deletions openrl/buffers/offpolicy_replay_data.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,52 +97,52 @@ def __init__(
)
self.first_insert_flag = True

def dict_insert(self, data):
if self._mixed_obs:
for key in self.critic_obs.keys():
self.critic_obs[key][self.step + 1] = data["critic_obs"][key].copy()
for key in self.policy_obs.keys():
self.policy_obs[key][self.step + 1] = data["policy_obs"][key].copy()
for key in self.next_policy_obs.keys():
self.next_policy_obs[key][self.step + 1] = data["next_policy_obs"][
key
].copy()
for key in self.next_critic_obs.keys():
self.next_critic_obs[key][self.step + 1] = data["next_critic_obs"][
key
].copy()
else:
self.critic_obs[self.step + 1] = data["critic_obs"].copy()
self.policy_obs[self.step + 1] = data["policy_obs"].copy()
self.next_policy_obs[self.step + 1] = data["next_policy_obs"].copy()
self.next_critic_obs[self.step + 1] = data["next_critic_obs"].copy()

if "rnn_states" in data:
self.rnn_states[self.step + 1] = data["rnn_states"].copy()
if "rnn_states_critic" in data:
self.rnn_states_critic[self.step + 1] = data["rnn_states_critic"].copy()
if "actions" in data:
self.actions[self.step + 1] = data["actions"].copy()
if "action_log_probs" in data:
self.action_log_probs[self.step] = data["action_log_probs"].copy()

if "value_preds" in data:
self.value_preds[self.step] = data["value_preds"].copy()
if "rewards" in data:
self.rewards[self.step + 1] = data["rewards"].copy()
if "masks" in data:
self.masks[self.step + 1] = data["masks"].copy()

if "bad_masks" in data:
self.bad_masks[self.step + 1] = data["bad_masks"].copy()
if "active_masks" in data:
self.active_masks[self.step + 1] = data["active_masks"].copy()
if "action_masks" in data:
self.action_masks[self.step + 1] = data["action_masks"].copy()

if (self.step + 1) % self.episode_length != 0:
self.first_insert_flag = False
self.step = (self.step + 1) % self.episode_length
# def dict_insert(self, data):
# if self._mixed_obs:
# for key in self.critic_obs.keys():
# self.critic_obs[key][self.step + 1] = data["critic_obs"][key].copy()
# for key in self.policy_obs.keys():
# self.policy_obs[key][self.step + 1] = data["policy_obs"][key].copy()
# for key in self.next_policy_obs.keys():
# self.next_policy_obs[key][self.step + 1] = data["next_policy_obs"][
# key
# ].copy()
# for key in self.next_critic_obs.keys():
# self.next_critic_obs[key][self.step + 1] = data["next_critic_obs"][
# key
# ].copy()
# else:
# self.critic_obs[self.step + 1] = data["critic_obs"].copy()
# self.policy_obs[self.step + 1] = data["policy_obs"].copy()
# self.next_policy_obs[self.step + 1] = data["next_policy_obs"].copy()
# self.next_critic_obs[self.step + 1] = data["next_critic_obs"].copy()
#
# if "rnn_states" in data:
# self.rnn_states[self.step + 1] = data["rnn_states"].copy()
# if "rnn_states_critic" in data:
# self.rnn_states_critic[self.step + 1] = data["rnn_states_critic"].copy()
# if "actions" in data:
# self.actions[self.step + 1] = data["actions"].copy()
# if "action_log_probs" in data:
# self.action_log_probs[self.step] = data["action_log_probs"].copy()
#
# if "value_preds" in data:
# self.value_preds[self.step] = data["value_preds"].copy()
# if "rewards" in data:
# self.rewards[self.step + 1] = data["rewards"].copy()
# if "masks" in data:
# self.masks[self.step + 1] = data["masks"].copy()
#
# if "bad_masks" in data:
# self.bad_masks[self.step + 1] = data["bad_masks"].copy()
# if "active_masks" in data:
# self.active_masks[self.step + 1] = data["active_masks"].copy()
# if "action_masks" in data:
# self.action_masks[self.step + 1] = data["action_masks"].copy()
#
# if (self.step + 1) % self.episode_length != 0:
# self.first_insert_flag = False
# self.step = (self.step + 1) % self.episode_length

def init_buffer(self, raw_obs, action_masks=None):
critic_obs = get_critic_obs(raw_obs)
Expand Down
Loading