-
First of all, thanks for sharing this code.
I'm trying to run the notebook `keras-ddpg` and I ran into this error:
During the `agent.fit()` method there is a callback to `TrainEpisodeLoggerPortfol…
-
I saw that there is a structure for the gym stuff that I will follow for the other gym implementations. train_* for the training and enjoy_* for the testing. I will add an optional param for continuin…
-
```
3.09s$ $CI_SOURCE_PATH/costar_task_plan/tests/task_test.py
pybullet build time: Aug 17 2017 04:18:10
[2017-08-17 04:27:03,449] You have 'numpy' version 1.8.2 installed, but 'gym' requires at le…
-
Pat,
Love your code and algo here. But I'd really like to see it running in parallel.
You might want to take Schulmans trpo_mpi code from openai baselines and put your algo in it. I'm trying th…
-
When following directions in **Training Your First Model** and running python 2.7.13 (conda 4.3.22), I get the following error. Please let me know any further info that I can provide.
```
Training…
-
Hi,
I tried to set the render flag to True but I received the following error:
**DDPG-master$** python3 ddpg_main.py
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA li…
-
What is a canonical way to tell the Keras-RL agent (DDPG for example) to save weights after epochs with the best reward? And to regularly save weights, for example after each 5 training epochs? Is it …
-
Hi Lukasz,
Currently there are 2 DDPG training scripts, in osim-rl/osim/sandbox/train.ddpg.py and osim-rl/scripts/keras-rl/train.ddpg.py
Is this intended? The osim-rl/scripts/keras-rl/train.ddpg…
-
I'd like to adapt the ddpg agent for environments other than Pendulum-v0. One problem is that the environment action is Discrete(6) for the Atari environment but a numpy array for the Pendulum-v0 envi…
-
Hi,
I am trying to run your code with tf 1.0.0, beside some changes like mul to multiply I have the following error:
**ddpg$** python pendulum_ddpg.py --device=cpu --episodes=300
Namespace(de…