Changes to run TRPO (#442)

Filter/new updates

final keras changes

some changes
This commit is contained in:
Richard Liaw
2016-09-26 20:20:45 -07:00
committed by Robert Nishihara
parent 9a6991116f
commit d22321da7a
+3 -3
View File
@@ -21,11 +21,11 @@ Install these dependencies:
pip install tabulate
```
Then run
Run this code to start an experiment.
```
git clone https://github.com/joschu/modular_rl
cd modular_rl
./run_pg.py --env Pong-ram-v0 --agent modular_rl.agentzoo.TrpoAgent --video 0 --n_iter 500 --filter 0
export KERAS_BACKEND=theano && ./run_pg.py --env Pong-ram-v0 --agent modular_rl.agentzoo.TrpoAgent --video 0 --n_iter 500 --filter 0
```
**Note: On some versions of Mac OS X, this produces NaNs.**
@@ -70,7 +70,7 @@ git checkout remote
You can run the code using
```
./run_pg.py --env Pong-ram-v0 --agent modular_rl.agentzoo.TrpoAgent --video 0 --n_iter 500 --filter 0 --remote 1 --n_rollouts 8
export KERAS_BACKEND=theano && ./run_pg.py --env Pong-ram-v0 --agent modular_rl.agentzoo.TrpoAgent --video 0 --n_iter 500 --filter 1 --remote 1 --n_rollouts 8
```
There are few [changes](https://github.com/joschu/modular_rl/compare/master...pcmoritz:23d3ebc).