Training on age & custom factor - generated model dimension issue

[taewookim]

Hey Yusuke

This is a continuation from issue 17.

After i ran train.py, I generated checkpoints, along with these files:

• history_16_8.h5
• WRN_16_8.h5
• WRN_16_8.json (you can see here)

In you predict.ipynb, I changed the

model.load_weights(os.path.join("pretrained_models", "weights.18-4.06.hdf5"))

to

model.load_weights("WRN_16_8.h5")

And I'm getting this odd error:

ValueError: Dimension 0 in both shapes must be equal, but are 131072 and 32768 for 'Assign_137' (op: 'Assign') with input shapes: [131072,2], [32768,2].

Any ideas where I might've gone wrong? I made sure I am using same widenet_res.py in both training and in prediction

BTW, I changed the nb_epoch from 30 to 100 during training but i doubt this is the cause...?

[yu4u]

Hi, you seem to have trained the network using images with 32x32 size.
Therefore, img_size = 32 will fix the issue.

[taewookim]

Oh wow. Once again, thank you SO much for your help. That worked!

Out of curiosity.. how were u able to figure that out w/o looking at my code?

[bsonnenschein]

It is evident because 131072/32768 = 4 just as 64x64/(32x32) = 4.

[taewookim]

wow.. you are sharp indeed. never even occurred to me to calculate that.
once again, thank you @bsonnenschein and @yu4u

[yu4u]

@bsonnenschein is right. And I saw "batch_input_shape": [null, 32, 32, 3] in WRN_16_8.json =D