Chen Junyang

icon indicating a website link https://colorfulandcjy0806.github.io/

icon company Southeast University @SEU icon location Jiangsu, China icon location Deep learning, multimodal perception, and 3D vision.

Results 6 comments of Chen Junyang

Import Error for spconv-cu117

Perhaps everyone can try this. My environment is ubuntu20.04 and cuda is 11.4. Although I tried many methods but found them ineffective, I successfully solved the above problem through pip...

import spconv.core_cc as _ext ImportError: arg(): could not convert default argument 'workspace: tv::Tensor' in method '<class 'spconv.core_cc.csrc.sparse.convops.gemmops.GemmTunerSimple'>.run_with_tuned_result' into a Python object (type not registered yet?

Perhaps everyone can try this. My environment is ubuntu20.04 and cuda is 11.4. Although I tried many methods but found them ineffective, I successfully solved the above problem through pip...

error ： raise AttributeError(name) from None

It is possible to change _2D to TWO_D, which is effective in my environment.

AttributeError: 'LiDARPoints' object has no attribute 'numpy'

It should be an environmental issue. Today, I reinstalled the MMDetection3D environment (as described in the document) and ran it again `python tools/create_data.py nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag nuscenes...

💡 [REQUEST] - <MiniCPM-V-2_6的int4量化>

这个问题有点神奇 应该是依赖库的原因 我重新pip install -r requirements.txt后成功了 应该是之前配vllm的时候 环境改动了

[BUG] <title>MiniCPM_V_2_6_INT4 版本执行提示 tensor 异常

这个问题有点神奇 应该是依赖库的原因 我重新pip install -r requirements.txt后成功了 应该是之前配vllm的时候，环境改动了。