GoingMyWay
GoingMyWay
@IbrahimSobh , How is your result now? Have you tried combinations of actions in the example code of ViZdoom? ``` In [1]: import itertools as it In [2]: actions =...
@wolleraudude, I ran the code with GPU ``` with tf.device('/gpu:1'): spawning multiple agents ``` After 10 hours training, the performance collapsed, my scenario is `deadly_corridor` not `healthpack gathering`
0.5 is the coefficient of loss of value, there is another [post](https://jaromiru.com/2017/03/26/lets-make-an-a3c-implementation/) which introduced this. However, why there is a minus sign in `- self.entropy * 0.01`, since the loss...
@wwxFromTju , if you have successfully installed vizdoom under Python3, I think the code is compatible under Python3 if you change print to print() in helper.py. I can run the...
@DontGiveUpEasily I trained it with 10k episode with the code, but the result is  Looks like it did not converge. Does the code work?
@hamzahamzii Could you share the code and the details on the settings? For example, the environment, the code, etc.
> Hi, > I run some scenarios like simple tag and I can only observed a partial window(sometimes the agent run out of the window,I can only see a part...
> @GoingMyWay Could you share the directory structures (in`megaverse/megaverse` and `megaverse/megaverse/extension` specifically) and the commit id for megaverse? Hi, inside the Docker container, it is ``` (sample-factory) I have no...
Output of `python setup.py develop` https://drive.google.com/file/d/13zosZBq9Kl254Rqnw22x9DmMgHq6rp1L/view?usp=sharing and output of ``` python -m megaverse_rl.train --train_for_seconds=360000000 --train_for_env_steps=2000000000 --algo=APPO --gamma=0.997 --use_rnn=True --rnn_num_layers=2 --num_workers=12 --num_envs_per_worker=2 --ppo_epochs=1 --rollout=32 --recurrence=32 --batch_size=2048 --actor_worker_gpus 0 --num_policies=1 --with_pbt=False --max_grad_norm=0.0...
> @GoingMyWay It seems that `setup.py` in `pip install -e .` didn't run properly. > > In `/workspace/megaverse`, could you run `git submodule update --init --recursive` (if you haven't init...