Made extremely simple model in DeepLearningToolbox, it trained, but I can't run it
1 次查看(过去 30 天)
显示 更早的评论
Revised wording of question.
I trained a deep learning system to add 4 numbers and return the sum.
To make it perform on new data, I think I have gotten the 4 numbers to be added into a cell array using mat2cell.
I make an array:
>> vc = [ 1; 2; 3; 4]
vc =
1
2
3
4
I convert it to a cell.
>> vca = mat2cell(vc,[4],[1])
vca =
1×1 cell array
{4×1 double}
My attempts to "index" the cell in an acceptible way have all failed. Here are some sample attempts.
>> xxy = fourplaceadderNet(vca)
Unable to use a value of type cell as an index.
>> xxy = fourplaceadderNet(vca{1,1})
Index exceeds the number of array elements. Index must not exceed 1.
>> xxy = fourplaceadderNet(vca{1})
Index exceeds the number of array elements. Index must not exceed 1.
>> xxy = fourplaceadderNet(vca(1))
Unable to use a value of type cell as an index.
>> xxy = fourplaceadderNet(vca(1:1))
Unable to use a value of type cell as an index.
What is the magic format for indexing my cell so this trained model can be used?
I understand from the "sim" man page that simple arrays might also work. I have tried that too without success.
I am now on my 6th day of messing with this.
Thanks for looking.
0 个评论
回答(1 个)
David Willingham
2022-6-16
Hi William,
Can you provide the function fourplaceadder? What type of deep learning model did you train?
3 个评论
David Willingham
2022-6-16
Hi William,
It's unfortunate to hear you didn't resolve your issue with your in house expert. If you do want to investigate it further, you have my and the entire deep learning development teams support at MathWorks to troubleshoot your issue. (Feel free to send the full code to me personally (dwilling@mathworks.com).
Regards,
Prinicipal Deep Learning Product Manager
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Sequence and Numeric Feature Data Workflows 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!