Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
B
baselines
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Model registry
Operate
Environments
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
xzhaoma
baselines
Commits
2c63e825
Commit
2c63e825
authored
5 years ago
by
Erik Nygren
Browse files
Options
Downloads
Patches
Plain Diff
updated observation and training to handle multi-speed
parent
cebde3d8
No related branches found
Branches containing commit
No related tags found
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
torch_training/render_agent_behavior.py
+6
-6
6 additions, 6 deletions
torch_training/render_agent_behavior.py
torch_training/training_navigation.py
+12
-7
12 additions, 7 deletions
torch_training/training_navigation.py
with
18 additions
and
13 deletions
torch_training/render_agent_behavior.py
+
6
−
6
View file @
2c63e825
...
...
@@ -38,7 +38,7 @@ min_dist = 5
observation_builder
=
TreeObsForRailEnv
(
max_depth
=
2
)
# Use a the malfunction generator to break agents from time to time
stochastic_data
=
{
'
prop_malfunction
'
:
0.
1
,
# Percentage of defective agents
stochastic_data
=
{
'
prop_malfunction
'
:
0.
0
,
# Percentage of defective agents
'
malfunction_rate
'
:
30
,
# Rate of malfunction occurence
'
min_duration
'
:
3
,
# Minimal duration of malfunction
'
max_duration
'
:
20
# Max duration of malfunction
...
...
@@ -48,10 +48,10 @@ stochastic_data = {'prop_malfunction': 0.1, # Percentage of defective agents
TreeObservation
=
TreeObsForRailEnv
(
max_depth
=
2
)
# Different agent types (trains) with different speeds.
speed_ration_map
=
{
1.
:
0.25
,
# Fast passenger train
1.
/
2.
:
0.
25
,
# Fast freight train
1.
/
3.
:
0.
25
,
# Slow commuter train
1.
/
4.
:
0.
25
}
# Slow freight train
speed_ration_map
=
{
1.
:
1.
,
# Fast passenger train
1.
/
2.
:
0.
0
,
# Fast freight train
1.
/
3.
:
0.
0
,
# Slow commuter train
1.
/
4.
:
0.
0
}
# Slow freight train
env
=
RailEnv
(
width
=
x_dim
,
height
=
y_dim
,
...
...
@@ -103,7 +103,7 @@ action_prob = [0] * action_size
agent_obs
=
[
None
]
*
env
.
get_num_agents
()
agent_next_obs
=
[
None
]
*
env
.
get_num_agents
()
agent
=
Agent
(
state_size
,
action_size
,
"
FC
"
,
0
)
with
path
(
torch_training
.
Nets
,
"
navigator_checkpoint100.pth
"
)
as
file_in
:
with
path
(
torch_training
.
Nets
,
"
navigator_checkpoint1
2
00.pth
"
)
as
file_in
:
agent
.
qnetwork_local
.
load_state_dict
(
torch
.
load
(
file_in
))
record_images
=
False
...
...
This diff is collapsed.
Click to expand it.
torch_training/training_navigation.py
+
12
−
7
View file @
2c63e825
...
...
@@ -37,7 +37,7 @@ def main(argv):
min_dist
=
5
# Use a the malfunction generator to break agents from time to time
stochastic_data
=
{
'
prop_malfunction
'
:
0.
1
,
# Percentage of defective agents
stochastic_data
=
{
'
prop_malfunction
'
:
0.
0
,
# Percentage of defective agents
'
malfunction_rate
'
:
30
,
# Rate of malfunction occurence
'
min_duration
'
:
3
,
# Minimal duration of malfunction
'
max_duration
'
:
20
# Max duration of malfunction
...
...
@@ -47,10 +47,10 @@ def main(argv):
TreeObservation
=
TreeObsForRailEnv
(
max_depth
=
2
)
# Different agent types (trains) with different speeds.
speed_ration_map
=
{
1.
:
0.25
,
# Fast passenger train
1.
/
2.
:
0.
25
,
# Fast freight train
1.
/
3.
:
0.
25
,
# Slow commuter train
1.
/
4.
:
0.
25
}
# Slow freight train
speed_ration_map
=
{
1.
:
1.
,
# Fast passenger train
1.
/
2.
:
0.
0
,
# Fast freight train
1.
/
3.
:
0.
0
,
# Slow commuter train
1.
/
4.
:
0.
0
}
# Slow freight train
env
=
RailEnv
(
width
=
x_dim
,
height
=
y_dim
,
...
...
@@ -120,7 +120,7 @@ def main(argv):
# Reset environment
obs
=
env
.
reset
(
True
,
True
)
register_action_state
=
np
.
zeros
(
env
.
get_num_agents
(),
dtype
=
bool
)
final_obs
=
agent_obs
.
copy
()
final_obs_next
=
agent_next_obs
.
copy
()
...
...
@@ -138,6 +138,11 @@ def main(argv):
# Action
for
a
in
range
(
env
.
get_num_agents
()):
if
env
.
agents
[
a
].
speed_data
[
'
position_fraction
'
]
==
0.
:
register_action_state
[
a
]
=
True
else
:
register_action_state
[
a
]
=
False
action
=
agent
.
act
(
agent_obs
[
a
],
eps
=
eps
)
action_prob
[
action
]
+=
1
action_dict
.
update
({
a
:
action
})
...
...
@@ -155,7 +160,7 @@ def main(argv):
final_obs
[
a
]
=
agent_obs
[
a
].
copy
()
final_obs_next
[
a
]
=
agent_next_obs
[
a
].
copy
()
final_action_dict
.
update
({
a
:
action_dict
[
a
]})
if
not
done
[
a
]:
if
not
done
[
a
]
and
register_action_state
[
a
]
:
agent
.
step
(
agent_obs
[
a
],
action_dict
[
a
],
all_rewards
[
a
],
agent_next_obs
[
a
],
done
[
a
])
score
+=
all_rewards
[
a
]
/
env
.
get_num_agents
()
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment