wangzhenxiong

可以新增PPO continuous 版本么

6

可以新增PPO continuous 版本么 看了好多版本==但是都晕晕乎乎的，从你的tutorial开始学的RL代码，所以还是想看看你咋写的T.T

DQN code chose_action

1

我发现老师在训练DQN的时候使用的是self.eval_net.forward(input),为什么这里不用self.eval_net(input)？

add PPO-continuous code

1

可以新增PPO-continuous的demo code么

pubsub.get_message(blocking_timeout) don't work with async_timeout.timeout

2

### Describe the bug ``` await pubsub.subscribe(self.channel) try: start_time = datetime.datetime.now() async with async_timeout.timeout(blocking_timeout): while True: message = await pubsub.get_message(ignore_subscribe_messages=True, timeout=blocking_timeout) end_time = datetime.datetime.now() logger.info(f"{str(message)},{blocking_timeout}==={end_time-start_time}") if message is not None:...

warning when work with torch.compile

2

[2023-11-30 18:31:43,775] torch._dynamo.convert_frame: [WARNING] WON'T CONVERT /miniconda3/envs/venv/lib/python3.11/site-packages/torch_cluster/radius.py line 69 [2023-11-30 18:31:43,775] torch._dynamo.convert_frame: [WARNING] due to: [2023-11-30 18:31:43,775] torch._dynamo.convert_frame: [WARNING] Traceback (most recent call last): [2023-11-30 18:31:43,775] torch._dynamo.convert_frame: [WARNING] File "/miniconda3/envs/venv/lib/python3.11/site-packages/torch/_ops.py",...

can k >100 in knn?

2

when i use knn to calculate atom neighbour,it's easy count > 100,can i set k > 100? in your cuda code u set k