Expected shape of ndim=4, got ndim=3, test_model.py
ayaanp-dev opened this issue · comments
Hi, I tried using v0.03 of this. Since this project is technically using tf 1, I build an xception model in keras that works with tf 2, it is basically just an advanced CNN. The train_model.py code is at https://mystb.in/wobemagohi.py. My xception model code is at https://mystb.in/ezorodifek.py. The training works perfectly fine, it's just my test_model that is having issues. The error is ValueError: Input 0 of layer sequential is incompatible with the layer: expected ndim=4, found ndim=3. Full shape received: [32, 270, 3]
The test_model code is at https://mystb.in/levodosubu.py. Thanks!
Solved with a np.reshape
Solved with a np.reshape
Hi, can you share your Xception model? I am having hard time to run sentdex's model.
Solved with a np.reshape
Hi, can you share your Xception model? I am having hard time to run sentdex's model.