FP32 FP16 INT8推理cpu gpu速度都一样
Open
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 23.8k
- Forks
- 4.5k
- Avg merge
- 2d 20h
- Merged PRs (30d)
- 37
Description
detail | 详细描述 | 詳細な説明
ncnn android yolov5s ,用FP32 FP16 INT8推理cpu gpu速度都一样怎么回事？量化后的int8模型确实有变小很多，但精度也变低了，唯独推理速度没变，这是为什么？测试手机小米12+雷电模拟器，ncnn版本[20221128]，ncnn-android-vulkan.zip
Contributor guide
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 linked ncnn-android-yolov5 example and compare the CPU and GPU inference paths for FP32, FP16, and INT8 using ncnn version 20221128. Reproduce the timings on the Xiaomi 12 and 雷电 emulator, then determine whether the identical speeds are expected and document the cause or the missing optimization.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- android, cpp
- Domain
- machine-learning, mobile-dev, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100