mm/mud-games
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Compare commits

base: mm:feature/notebook
Branches Tags
mm:main
mm:feature/notebook
..
compare: mm:main
Branches Tags
mm:main
mm:feature/notebook

3 Commits
feature/no ... main

Author SHA1 Message Date
mm
1a2259827f Update 'README.md' 2022-05-26 15:28:47 +00:00
mm
0949c9bda0 Update 'README.md' 2022-05-26 15:23:08 +00:00
mm
cdee785166 add notebook, support interactive output to enable visualizing results (#1) 
needed to install some virtual display software.
`gcc` may be required to add to `./headless.sh`

Co-authored-by: Michael Pilosov <consistentbayes@gmail.com>
Reviewed-on: #1
Co-authored-by: mm <mm@clfx.cc>
Co-committed-by: mm <mm@clfx.cc>
 2022-05-26 15:12:56 +00:00
1 changed files with 8 additions and 2 deletions
Show Stats Expand all files Collapse all files

10
README.md
View File

@ -2,6 +2,7 @@
control systems with MUD points control systems with MUD points
# installation # installation
```bash ```bash
@ -17,6 +18,9 @@ A `data.pkl` file is provided for your convenience with input / output samples.
python main.py python main.py
``` ```
You can also instead use the included [jupyter notebook](./DemoMUD.ipynb).
# info # info
The inputs are the parameters to a `1x4` matrix which is multiplied against the observations of the state in order to make a decision for the next action (push left or right). The output of the vector inner-product is binarized by comparing it to zero as a threshold value. The inputs are the parameters to a `1x4` matrix which is multiplied against the observations of the state in order to make a decision for the next action (push left or right). The output of the vector inner-product is binarized by comparing it to zero as a threshold value.
@ -29,6 +33,7 @@ There is no assumed error in observations; the "data variance" is designed to re
Therefore, since our objective is to stabilize the cart, the target "time series signal" is zero for all four dimensions of the observation space. The presumed "data variance" should actually correspond to the acceptable bands of signal (WIP). Therefore, since our objective is to stabilize the cart, the target "time series signal" is zero for all four dimensions of the observation space. The presumed "data variance" should actually correspond to the acceptable bands of signal (WIP).
# generate data # generate data
You can generate your own data with: You can generate your own data with:
@ -45,7 +50,8 @@ Using the following presumptions, we can establish better values for the "data v
> The angular momentum of the pole is the most important thing to stabilize. > The angular momentum of the pole is the most important thing to stabilize.
# headless mode
Run `./headless.sh` (requires `sudo`) to install virtual displays so you can view results in a Jupyter notebook. # headless mode / notebook demos
Run `./headless.sh` (requires `sudo`) to install virtual displays so you can use the included Jupyter notebooks.