Googlenet for regression type problem

I am trying to use transfer learning to retrain Googlenet for a regression type problem (replacing the last 3 layers with a fc layer and a regression layer—143 layers total)—similar to this tutorial (https://www.mathworks.com/help/deeplearning/examples/train-a-convolutional-neural-network-for-regression.html).

I’m getting an error for an ‘Input Size Mismatch’...but it’s for one of the interior layers and not layer 1? I have built a table with Var1 as the relative paths to my files and Var2 as the response value for each image. I have also created an augmented datastore to automatically resize my images to the required 224x224. The augmented datastore seems OK...although the .Files property is empty (not sure if this is contributing to my problem). Any thoughts on what I’m doing wrong?

[Andreas Götz]

I'm trying the same. I get this error message:

Caused by:

Layer 'inception_3a-3x3_reduce': Invalid input data. The number of channels of the input data (64) must match the layer's expected number of channels (192).

Layer 'inception_3a-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_3b-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_4a-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_4b-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_4c-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_4d-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_4e-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_5a-output': Unconnected input. Each layer input must be connected to the output of another layer.

Layer 'inception_5b-output': Unconnected input. Each layer input must be connected to the output of another layer.

What went wrong?

[Cris LaPierre]

Please share your code.