arduino / arduino/ArduinoTensorFlowLiteTutorials

xx bytes lost due to alignment. To avoid this loss, please make sure the tensor_arena is 16 bytes aligned

未关闭
#21 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Jupyter Notebook
星标
270
派生
122
PR 合并指标
30 天内没有已合并 PR

描述

I get this error message I get when I basically replicate the gesture recognition tutorial in [Arduino ML guide](https://blog.arduino.cc/2019/10/15/get-started-with-machine-learning-on-arduino/).

More importantly, though the arduino sketch builds, the model doesn't appear to be working, since the serial output doesn't report any gestures.

The workaround that worked for me was to change the header in the model.h encoding in the Google Colab ML notebook:

Original: !echo "const unsigned char model[] = {"

What worked for me: !echo "alignas(8)const unsigned char model[] = {"

The addition of alignas(8) fixed the issue in that the arduino sketch build is now working, and I am able to classify gestures.

贡献指南

这个仓库没有索引到贡献指南

调研方向

从生成 model.h 的 Google Colab ML notebook 命令以及报告 tensor_arena 对齐警告的 Arduino sketch 开始。重现手势识别构建和串行输出,然后验证生成的模型头文件已正确对齐,并且手势能够在没有该警告的情况下完成分类。

由索引模型根据 Issue 内容生成。

评估

技术栈
arduino, jupyter-notebook
领域
embedded-iot, machine-learning
Issue 类型
缺陷
难度
3/5
预计耗时
1-2 天
活跃度
停滞
描述清晰度
基本清楚
新手友好度
35/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。