Tejan Karmali

Decoupled Neural Interface on MLP model. This model is able to produce results close to MLP, without standard backpropagation of gradients. Ref.: [Understanding Synthetic Gradients and Decoupled Neural Interfaces](https://arxiv.org/pdf/1703.00522.pdf)

I have added the following GAN models: - LSGAN - DCGAN - WGAN Here are some results:  WGAN samples  DCGAN samples  LSGAN samples

Prioritized Experience Replay DQN

3

Implemented DQN with PER. Kindly review the code. Looking forward towards the feedback. Ref: - https://arxiv.org/pdf/1511.05952.pdf - https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow/blob/master/contents/5.2_Prioritized_Replay_DQN/RL_brain.py

Added Advantage Actor-Critic

Does PyTorch3D support autocast?

4

Hi, I was wondering if current release of PyTorch3D intends to support the latest amp and autocast features of PyTorch 1.6. I tried rendering a mesh with `autocast` enabled, but...

[fix] Update `upgrade` command for logging repo

1

`aim upgrade 2to3` command is now `aim storage upgrade 2to3`[1]. Updated the `repo_utils.py` to reflect the same. References: [1] https://aimstack.readthedocs.io/en/latest/refs/cli.html#storage

Code review

9

Hey @MikeInnes, if you are back could you please review the code? New models which I have added are Dueling DQN, Advantage Actor-Critic, and DDPG. Also, all the previous work...

Fixed vt for face textures

1

Corrected `face_textures` in places where `vt` was optional for face