Neural network training combine data store 2018

Question

tim pearce 2024-1-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2070986-neural-network-training-combine-data-store-2018

回答： Ramtej 2024-1-18

Hi,

I'm stuck with 2018b do to work restrictions, I'm trying to build a image classification network, i have all my images segmented and also bounding boxes with a gTruth file.

I need to train the network following a tutorial it

trainingData = combine(imds,pxds)

net = trainingNetwork(trainingData,layers,options)

I run into a problem ive got my image data store - imds and PixelDatastore pxds but the command combione is not supported in 2018b

any help how to procced

kind regards

Tim

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Ramtej 2024-1-18

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2070986-neural-network-training-combine-data-store-2018#answer_1392186

在 MATLAB Online 中打开

Hi Tim,

I assume you are referring to pixel label datastore by the term "PixelDatastore" and image datastore, pixel label datastore are stored in the variables "imds" and "pxds" respectively.

You can leverage the "pixelLabelImageDatastore" function to create a datastore based on the input image datastore and the pixel label datastore as shown below:

trainingData = pixelLabelImageDatastore(imds,pxds);

Refer to the below documentation for the detailed instructions and limitations of the "pixelLableImageDatastore" function.

pixelLabelImageDatastore: www.mathworks.com/help/vision/ref/pixellabelimagedatastore.html

Hope this resolves your query!