-
Hi, I recently **upgraded to pytorch 0.2.0 from 0.1.12** and noticed that it drastically slowed down an a3c implementation I tested. I investigated what could be the problem and found out that the **.…
-
-
I have implemented A3C with multiprocessing (+ pytorch) as opposed to using threads, however `bazel run` seems to break silently and clearly without any visible trace. This is what I do:
```bash
$…
edran updated
4 years ago
-
Hi,
I am trying to follow the ray [tutorials](https://github.com/ray-project/tutorial). Nevertheless, I found several bug with save/load functional in ray 0.4 and decide to move to the master brach…
-
I use same optimizer, same model architecture and same initialization, but the performance is bad.
-
### System information
- **OS Platform and Distribution (e.g., Linux Ubuntu 16.04)**: Linux Ubuntu 16.04
- **Ray installed from (source or binary)**: source
- **Ray version**: 0.5.3
- **Python…
-
Thank you for sharing your work. It helps a lot.
It seems the code does not include training part? (please correct me if I'm wrong) Would you kindly include it as well, thanks so much.
-
在论文中给出的算法伪码,我理解的是,不论是task encoder的更新还是策略的更新都是直接对公式(10)进行SGD,这是一个meta-update,应该可以简单地理解成一个梯度更新。但是在源码中,我看到了ppo和a3c以及它们的loss,请问tesp需要借助ppo或者a3c的策略更新方式么?
-
I'm training an A3C trainer with Pytorch and tune.run().
tune.run(a3c.A3CTrainer, local_dir=".", stop={"episode_reward_mean": 0.5},
resources_per_trial={"cpu":1, "gpu":1},
…
-
Hello, I have an error when I run test.py.
`python3 test.py
Traceback (most recent call last):
File "test.py", line 68, in
test(opt)
File "test.py", line 58, in test
state, rewar…