公告

大中华汽车电子生态圈社区并入开发者社区- 更多资讯点击此

Tip / 登入 to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
lock attach
Attachments are accessible only for community members.
gaohailong
Level 3
Level 3
50 sign-ins 25 replies posted First like received

使用ml configurator进行tflite网络模型的转换不成功,提示错误,请帮忙协助  谢谢

[ERROR] Failed to run QEMU with tflm-qemu-model-mem-calc-cm4

[ERROR] ****************** TFL-micro Inference Arena Measure ******************,详细的错误日志及待转换模型已经挂在附件上   请求大神指导

0 点赞
1 解答
Ida_Wang
Moderator
Moderator
Moderator
25 solutions authored 10 likes received 50 sign-ins

Ida_Wang_1-1692857256155.png

您看一下这个报错信息,模型的输入张量超过了最大值。kMaxInputNum相关的信息可以查看下面这个链接

https://github.com/tensorflow/tflite-micro/blob/34691a66f27f4f7c2207ba78e630e57d881d2a4d/tensorflow/...

是否在生成模型的时候不满足这个限制条件?

在原帖中查看解决方案

0 点赞
5 回复数
gaohailong
Level 3
Level 3
50 sign-ins 25 replies posted First like received

ml configurator版本号是使用的2.0

0 点赞
lock attach
Attachments are accessible only for community members.
Ida_Wang
Moderator
Moderator
Moderator
25 solutions authored 10 likes received 50 sign-ins

@gaohailong 

附件的这个模型是示例工程中的,您可以先用这个模型试一下是否可以用ML Configurator 2.0转换。如果可以转换,说明当前配置的环境没有问题;如果转换失败,我们再看一下您配置的环境哪个地方有问题。期待您的回复~

Ida

0 点赞
lock attach
Attachments are accessible only for community members.
gaohailong
Level 3
Level 3
50 sign-ins 25 replies posted First like received

您好 您给的这个实例是可以转换成功的 ,您可否把我的模型在您的电脑上转换看看  帮忙给排查一下是什么原因 转换不成功跟模型大小有关系吗?我的模型比实例模型大且复杂  期待您的回复 谢谢

0 点赞
Ida_Wang
Moderator
Moderator
Moderator
25 solutions authored 10 likes received 50 sign-ins

Ida_Wang_1-1692857256155.png

您看一下这个报错信息,模型的输入张量超过了最大值。kMaxInputNum相关的信息可以查看下面这个链接

https://github.com/tensorflow/tflite-micro/blob/34691a66f27f4f7c2207ba78e630e57d881d2a4d/tensorflow/...

是否在生成模型的时候不满足这个限制条件?

0 点赞
gaohailong
Level 3
Level 3
50 sign-ins 25 replies posted First like received

好的 谢谢 我们排查一下我们的模型  有需要再联系您  谢谢

0 点赞