Megvii-BaseDetection / Megvii-BaseDetection/YOLOX
关于转换TensorRT后显存增加的问题
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 10.6k
- Forks
- 2.5k
- PR merge metrics
- No merged PRs in 30d
Description
大佬你好， 感谢您的工作。
我目前在使用yolox-l，经过torchtrt转换以后的模型，在推理(python)的时候，我发现显存占用会很大，大约在4.9G左右。然而使用C++推理程序，显存只有1.8G。检测效果都是正常，非常nice。
请问一下，这是什么原因呢，这个问题有解决办法吗？如果能将python推理的显存占用降低到与C++一样就非常好了。
这个问题不管是使用您的官方权重，还是使用我自己的数据集训练出来的模型，都是有这个问题。
我使用的是3080显卡。
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the Python inference path for the torchtrt-converted yolox-l model and compare its GPU memory use with the C++ inference program. Reproduce the reported 4.9G versus 1.8G usage on an RTX 3080 using both official and custom-trained weights, then identify and document the cause or a verified way to reduce the Python allocation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100