Frank (Haofan) Wang

Result divergence of deployed ConvNeXt model

3

I used the latest MMDeploy tool to convert ConvNeXt into tensorRT backend. It converted successfully. However, when I visualized the results, I found the output from tensorRT backend was totally...

Inconsistency with web result

7

I found the result from ts.google() is different from the website (https://translate.google.com/), why this happen?

import error

1

When I run ``` import onnx import onnx_caffe2.backend ``` I got ``` [libprotobuf FATAL google/protobuf/stubs/common.cc:61] This program requires version 3.4.0 of the Protocol Buffer runtime library, but the installed version...

This line is a wired bug. Its bounding box is not right. I have noticed that many of recent works (PARE, MPS-Net) adopt this repo to do multi-person tracking, this...

Usage of motion infiller

1

I'm using another 3D pose estimation method instead of HybrIK, I have several questions. 1. Does GLAMR depend on [HybrIK](https://github.com/Jeff-sjtu/HybrIK)? Can it be replaced by other recent pose estimation methods?...

Remove duplicates in training scripts

2

As title.

accelerator.num_processes -> gpc.get_world_size(ParallelMode.DATA)

[FEATURE]: Integration to diffusers framework

1

### Describe the feature Diffusers has become a common framework for stable diffusion. It seems that you already have a [plan](https://github.com/huggingface/diffusers/issues/1212) to integrate to diffusers. What's the progress? I'm looking...

[BUG]: AssertionError: you are calculating the l2 norm twice

4

### 🐛 Describe the bug **Describe the bug** I meet AssertionError: you are calculating the l2 norm twice, which looks similar to another [issue](https://github.com/hpcaitech/ColossalAI/issues/2380). I guess it is related to...

### 🐛 Describe the bug **Describe the bug** I'm working on the [dreambooth example](train_dreambooth_colossalai.py). gpc.set_seed(args.seed) will cause following error. `AssertionError: The seed for ParallelMode.DATA has been added ` if I...