-
start-zertoclonevpg
Is there a way to call multiple VPG's using this function? I have tried a few different options, like comma, brackets, quotes, etc. I figured maybe I would have to build an …
-
Tested on rpcs3-v0.0.5-6498-7233640c_win64
**Graphical issues**
- [x] Hud and menu colors are off.
- [x] Some particles are purple/blue.
- [x] Some objects are only lit up on one half.
- [x] Gr…
-
I cloned your repo and ran the vpg algo and compared the perf with the tensorflow version. I did an average of 5 runs to take care of the random seed and I saw some interesting results
Tensorflow: …
-
Given the number of science experiments now included in bdb some custom contracts for BDB could be a nice addition. They can be a compatibility patch for Contract Configurator.
I think the lowest f…
-
When I run the test for ThreadedRunner (https://github.com/reinforceio/tensorforce/blob/master/tensorforce/tests/test_vpg_multithreaded.py#L74-L75) it passes when it shouldn't
In the test, the runn…
-
## Expected Behavior
A ZertoFailoverTest begins.
## Current Behavior
After using 'Connect-ZertoServer' I am able to run other functions, but cannot start a failover test. Specifying vpgName and/…
-
Hi @AlexKuhnle,
I have some trouble training a _ppo agent_. Basically, I'm able to train it for only **very few episodes** (e.g. 4, 8). If I increase the number of episodes, my laptop will crash or …
-
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？
-
### Terraform Version
**Terraform Version**
```
Terraform 10.8
provider.aws 0.1.4
provider.random 0.1.0
provider.template 0.1.1
```
### Terraform Configuration Files
```hcl
/*
ter…
-
This is final epoch values from trpo_swimmer_ray_sampler.py code:
--------------------------------------- ------------
AverageDiscountedReturn 21.5863
AverageReturn …