Keras model with 3 input images giving wrong output

Question

I have created a keras model that takes 3 images as input, passes them to individual CNN backbone(mobilenet_v2) and fuse the results from 3 individual streams. These fused outputs further goes through a FCN and gives probability values for 10 classes. Now when i pass 3 images to my model using model.predict(), I am getting an output of 3x10 (list of 3 outputs with 10 values in each).

Here is the network snapshot

and here is the output

*[[0.04718336 0.07464679 0.1329775  0.09312231 0.12029872 0.10404643 0.08732469 0.03571845 0.16900443 0.13567731]
 [0.0726063  0.0712122  0.12180576 0.07443767 0.14696348 0.10402806 0.09013776 0.03013403 0.17304562 0.11562922]
 [0.06313297 0.06455057 0.11175603 0.06945134 0.16163406 0.12042907 0.10410044 0.04393855 0.12963305 0.13137391]]*

Any idea what is happening here?

Answer 1

It returns the probabilities of the given sample to be in some class. So you have 10 classes. You can see which class your model predicts by adding;

y_pred = y_pred.argmax(axis=1)