 sp
|
24dec631aa
|
record gifs when evaluating
This currently does not use tensorboards feature, we have to leave this
as a TBD. The gifs will be left in the experiments log_dir
|
10 months ago |
|
sp
|
d7e7a2411b
|
use shield in evaluation when full shielding
|
11 months ago |
|
sp
|
2a0b75bb15
|
added helper for ShieldingConfig
|
11 months ago |
|
sp
|
028c942625
|
evaluate sb3 training
WIP: Not a 100% sure whether the masking will be used in the evaluation
|
11 months ago |
|
sp
|
a69dab422c
|
log to stdout
|
11 months ago |
|
sp
|
f68a4052f9
|
add expname suffix
|
11 months ago |
|
sp
|
c387d99a6c
|
include unique id in experiment name
|
11 months ago |
|
sp
|
b696dac5f6
|
configure logger manually to change csv filename
|
11 months ago |
|
sp
|
d2aa224bc8
|
move isodate to expname
|
11 months ago |
|
sp
|
8cbbef4006
|
ensure that exp log directory exists
|
11 months ago |
|
sp
|
703b213248
|
log to csv and tensorboard only
|
11 months ago |
|
sp
|
95a06bd0f0
|
pass nocleanup flag correctly
|
11 months ago |
|
sp
|
d194537f56
|
shortened expname
|
11 months ago |
|
sp
|
09ef1aac10
|
fixed nocleanup argument
|
11 months ago |
|
sp
|
efbad0cc27
|
log to subdirectory
|
11 months ago |
|
sp
|
5ab83b7460
|
only create ShieldHandler when necessary
this also renames camelcase variable logDir
|
11 months ago |
|
sp
|
7c39ab3b87
|
pass cleanup flag to handler
|
11 months ago |
|
sp
|
11d5c3c811
|
store shield files in local tmp dir
|
11 months ago |
|
sp
|
29799bc52f
|
resynthesize on reset is default False
|
11 months ago |
|
sp
|
7627645e63
|
changed default expname
|
11 months ago |
|
sp
|
2bcb38f6af
|
refactored training without shield
|
11 months ago |
|
sp
|
36c04f1b81
|
set sb3 device to auto
This automatically detects whether a GPU can be used for training.
|
11 months ago |
|
sp
|
71854bae01
|
init evalCallback for training with sb3
|
11 months ago |
|
sp
|
59c795348e
|
changes in sb3 rl training
- included callbacks for initial image and info plotting
- switched to CnnPolicy
- changed GRID_TO_PRISM_BINARY to environment var M2P_BINARY
|
11 months ago |
|
sp
|
315b0c8e7d
|
added useful sb3 callbacks
|
11 months ago |
|
sp
|
62c1198f25
|
removed observation changes from shielding wrapper
|
11 months ago |
|
sp
|
16490a74f1
|
init Miniwrapper to switch to WxHxC observations
|
11 months ago |
|
sp
|
7ccbe8f9bc
|
changes according to refactoring of utils
|
11 months ago |
|
sp
|
372006a1da
|
major refactor in utils
- introduced common_parser for arguments
- the shield dict uses minigrid.core.State instead of strings
- switched shield query to minigrid get_symbolic_state
|
11 months ago |
 Thomas Knoll
|
175171c035
|
added jpg remove on gif create
|
11 months ago |
|
sp
|
521c71eba4
|
removed erroneous file write
|
11 months ago |
|
sp
|
4486dba2c9
|
disabled shield creation at reset WIP
|
11 months ago |
|
sp
|
aa2cec0e0f
|
disabled default shield creation on reset
|
11 months ago |
|
sp
|
7eeb816013
|
changed default env
|
11 months ago |
|
sp
|
c7c2296c71
|
changed early stopping criterion in tuner
|
11 months ago |
|
Thomas Knoll
|
555511bd34
|
args for turn prob
|
11 months ago |
|
Thomas Knoll
|
f8a3c52b9c
|
fixed shielding
|
11 months ago |
|
Thomas Knoll
|
1cbaac75cb
|
cleanups
|
11 months ago |
|
Thomas Knoll
|
5dcabef8e0
|
added utils classes
|
11 months ago |
|
Thomas Knoll
|
ae94b57876
|
changed iteration handling
|
11 months ago |
|
Stefan Pranger
|
f3b12f4caa
|
removed callbacks for shield info
|
11 months ago |
|
sp
|
dd8809b517
|
trying add_text
|
11 months ago |
|
sp
|
8d7546cfc7
|
trying subdirectory
|
11 months ago |
|
sp
|
d4cc2aa3b9
|
debugging
|
11 months ago |
|
sp
|
7cc0f5984a
|
debugging
|
11 months ago |
|
sp
|
ac1bddc1ec
|
flush after writing text
|
11 months ago |
|
sp
|
1d6b12ba5a
|
fixed typo
|
11 months ago |
|
sp
|
e8dc44673d
|
another try
|
11 months ago |
|
sp
|
8599241c01
|
set file_writer as field
|
11 months ago |
|
sp
|
24badd36ac
|
trying to log on every episode start
|
11 months ago |
