Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
S
seminar-breakout
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
0
Issues
0
List
Boards
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Analytics
Analytics
CI / CD
Repository
Value Stream
Wiki
Wiki
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
Shashank Suhas
seminar-breakout
Commits
13d4171f
You need to sign in or sign up before continuing.
Commit
13d4171f
authored
Jun 07, 2016
by
Yuxin Wu
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
add history frame player
parent
5f98e6ca
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
58 additions
and
0 deletions
+58
-0
tensorpack/RL/history.py
tensorpack/RL/history.py
+58
-0
No files found.
tensorpack/RL/history.py
0 → 100644
View file @
13d4171f
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# File: history.py
# Author: Yuxin Wu <ppwwyyxxc@gmail.com>
import
numpy
as
np
from
collections
import
deque
from
.envbase
import
ProxyPlayer
__all__
=
[
'HistoryFramePlayer'
]
class
HistoryFramePlayer
(
ProxyPlayer
):
""" Include history frames in state, or use black images"""
def
__init__
(
self
,
player
,
hist_len
):
""" :params hist_len: total length of the state, including the current
and `hist_len-1` history"""
super
(
HistoryFramePlayer
,
self
)
.
__init__
(
player
)
self
.
history
=
deque
(
maxlen
=
hist_len
)
s
=
self
.
player
.
current_state
()
self
.
history
.
append
(
s
)
def
current_state
(
self
):
assert
len
(
self
.
history
)
!=
0
diff_len
=
self
.
history
.
maxlen
-
len
(
self
.
history
)
if
diff_len
==
0
:
return
np
.
concatenate
(
self
.
history
,
axis
=
2
)
zeros
=
[
np
.
zeros_like
(
self
.
history
[
0
])
for
k
in
range
(
diff_len
)]
for
k
in
self
.
history
:
zeros
.
append
(
k
)
return
np
.
concatenate
(
zeros
,
axis
=
2
)
def
action
(
self
,
act
):
r
,
isOver
=
self
.
player
.
action
(
act
)
s
=
self
.
player
.
current_state
()
self
.
history
.
append
(
s
)
if
isOver
:
# s would be a new episode
self
.
history
.
clear
()
self
.
history
.
append
(
s
)
return
(
r
,
isOver
)
def
restart_episode
(
self
):
super
(
HistoryFramePlayer
,
self
)
.
restart_episode
()
self
.
history
.
clear
()
self
.
history
.
append
(
self
.
player
.
current_state
())
class
TimePointHistoryFramePlayer
(
HistoryFramePlayer
):
""" Include history from a list of time points in the past"""
def
__init__
(
self
,
player
,
hists
):
""" hists: a list of positive integers. 1 means the last frame"""
queue_size
=
max
(
hists
)
+
1
super
(
TimePointHistoryFramePlayer
,
self
)
.
__init__
(
player
,
queue_size
)
self
.
hists
=
hists
def
current_state
(
self
):
# TODO
pass
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment