Zihan Ding

modify RL examples to TF2 TL2

3

### Checklist - [x] I've tested that my changes are compatible with the latest version of Tensorflow. - [x] I've read the [Contribution Guidelines](https://github.com/tensorlayer/tensorlayer/blob/master/CONTRIBUTING.md) - [x] I've updated the documentation...

An issue in gaussian mixture

1

Hi, Should it be the inverse of covariance matrix here? https://github.com/ctgk/PRML/blob/47d9830b39e91690cedb27003934df85cfb8c5c7/prml/rv/multivariate_gaussian_mixture.py#L124 i.e. ` D_sq = np.sum(np.einsum('nki,kij->nkj', d, np.linalg.inv(self.cov)) * d, -1) `

Hi, Since this repo has been developed years ago and in an Archive status, it seems to be incompatible with the newest gym. I tested this repo with gym==0.16 and...

Inconsistent loss term with paper

8

Hi, Thanks for open-sourcing this wonderful project! However, I notice in CT training, the [loss term](https://github.com/openai/consistency_models/blob/6d26080c58244555c031dbc63080c0961af74200/cm/karras_diffusion.py#L199) has the target model denoising for $x_{t_{n+1}}$ instead of $x_{t_n}$, which is different from...

Hi, Thanks for open-sourcing this great project! I'm wondering if it's possible to support controlling both side of players in the game, the env.step() will take two actions per step,...

Segmentation fault

5

Hi, When I run experiments with `python train.py --task=ShadowHandOver --algo=ppo`, it generates the following error: ``` Algorithm: ppo Python Averaging factor: 0.01 Obs type: full_state Not connected to PVD +++...

samples.chunk

1

Hi, I may ask a dummy question, why in the inference code, the latent is first duplicate and then take the first chunk, see code [here](https://github.com/hpcaitech/Open-Sora/blob/a37a189482a4cd1c7892aa06881e539cbf8078ce/opensora/schedulers/iddpm/__init__.py#L66): ``` z = torch.cat([z,...

data loading typo

1

Hi, this [line](https://github.com/nicklashansen/tdmpc2/blob/5f6fadec0fec78304b4b53e8171d348b58cac486/tdmpc2/trainer/offline_trainer.py#L68) should be: ``` assert self.buffer.num_eps == (self.buffer.capacity // _cfg.episode_length), \ f'Buffer has {self.buffer.num_eps} episodes, expected {self.buffer.capacity // _cfg.episode_length} episodes.' ```

Hi, Thanks for providing such a geat code as a baseline for research purpose! However, when applying the code, I notice there may be a memory (RAM) leakage during training....

issue when using spawn

1

The code: ``` from decord import VideoReader, cpu, gpu class DecordInit(object): """Using Decord(https://github.com/dmlc/decord) to initialize the video_reader.""" def __init__(self, device_type='cpu', device_id=0, num_threads=1): self.num_threads = num_threads if device_type == 'cpu': self.ctx...