sp
|
d7e7a2411b
|
use shield in evaluation when full shielding
|
1 year ago |
|
sp
|
028c942625
|
evaluate sb3 training
WIP: Not a 100% sure whether the masking will be used in the evaluation
|
1 year ago |
|
sp
|
a69dab422c
|
log to stdout
|
1 year ago |
|
sp
|
b696dac5f6
|
configure logger manually to change csv filename
|
1 year ago |
|
sp
|
703b213248
|
log to csv and tensorboard only
|
1 year ago |
|
sp
|
95a06bd0f0
|
pass nocleanup flag correctly
|
1 year ago |
|
sp
|
5ab83b7460
|
only create ShieldHandler when necessary
this also renames camelcase variable logDir
|
1 year ago |
|
sp
|
7c39ab3b87
|
pass cleanup flag to handler
|
1 year ago |
|
sp
|
2bcb38f6af
|
refactored training without shield
|
1 year ago |
|
sp
|
36c04f1b81
|
set sb3 device to auto
This automatically detects whether a GPU can be used for training.
|
1 year ago |
|
sp
|
71854bae01
|
init evalCallback for training with sb3
|
1 year 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
|
1 year ago |
|
sp
|
7ccbe8f9bc
|
changes according to refactoring of utils
|
1 year ago |
 Thomas Knoll
|
5dcabef8e0
|
added utils classes
|
1 year ago |
Thomas Knoll
|
1528173f58
|
changed iterations to evaluations
|
2 years ago |
|
Thomas Knoll
|
3dee543e24
|
renaming and notebooks
|
2 years ago |
|
Thomas Knoll
|
138d917fd6
|
added tune example
refactored and evaluation logging
|
2 years ago |
|
Thomas Knoll
|
f3747a1479
|
renaming / shield handling changes
|
2 years ago |
|
Thomas Knoll
|
757fbbcc0d
|
fixed shield generation
worker handling
|
2 years ago |
|
Thomas Knoll
|
1c2dbf706e
|
changed shield creation to create shield on reset
|
2 years ago |
|
Thomas Knoll
|
97f7d23cda
|
added rudimental key / door masking
|
2 years ago |
|
Thomas Knoll
|
b1b014dbd6
|
some refactoring as preparation for sb3 example
added sb3 example
|
2 years ago |
