zhangvia

using align and post-process simultaneously Error occured during execution of the processing block! See the log for more info

15

---------------------------------------------------------------------------------------------------- | Required Info | | |---------------------------------|------------------------------------------- | | Camera Model | { D435i } | | Firmware Version | (Open RealSense Viewer --> Click info) | | Operating System...

mutexlock类的析构函数

1

不应该先释放锁再destroy吗，怎么mutexlock.h里面的析构函数先lock（mutex）再destroy

[BUG]: ddp training in diffusion

1

### 🐛 Describe the bug how can i use the ddp train in diffusion? i saw the train_ddp.yaml，but there is nothing different with the train_colossalai.yaml. how do i set the...

<title>the results of text3d are bad

5

### Description when i use the text3d ,the results looks badly.such as i use the text "a highly detailed stone bust of Aphrodite", there always are two face of the...

[BUG]: assert len(grad_args) > 0 in training the animate anyone

2

### 🐛 Describe the bug i'm using the colossalai to train the [animate anyone](https://github.com/guoqincode/Open-AnimateAnyone). but there is an error in unet forward. ```python Traceback (most recent call last): File "core/train_stage2_colo.py",...

[BUG]: bug in using HybridAdam optimizer

1

### 🐛 Describe the bug i'm using the colossalai to train animate anyone here is the code where bug happened: ```python for name, param in unet.named_parameters(): if "motion_modules" in name:...

[BUG]: does gemini plugin only support fp16 and bf16?

4

### 🐛 Describe the bug what if i want to train model with full precision not mixed precision? the gemini plugin can only be used to mixed precision training? ###...

[BUG]: PIDM model is not compatible with gemini ddp

2

### 🐛 Describe the bug when i use booster api and gemini plugin to train the PIDM, this error happens: ```python File "train.py", line 167, in train booster.backward(loss, optimizer) File...

[BUG]: model not compatible with GeminiDDP

3

### 🐛 Describe the bug i'm using the colossalai to train the tryon diffusion.but there is a comptatible error. the paper link is [try on diffusion](https://tryondiffusion.github.io/) ("ZERO DDP error: the...

i have a training script written by myself (stage 1). it worked in single gpu training. but it will random stuck in backward at some step. the nccl communication has...