Deep Learning with TensorFlow in Python: Convolution Neural Nets

The following problems appeared in the assignments in the Udacity course Deep Learning (by Google). The descriptions of the problems are taken from the assignments (continued from the last post). Let's try to get the best performance using a multi-layer model! The following figure recapitulates the neural network with a 3 hidden layers, the first one with 2048 nodes, the second one with 512 nodes and the third one with with 128nodes, each one with Relu intermediate outputs. The L2 regularizations applied on the lossfunction for the weights learnt at the input and the hidden layers are λ1, λ2, λ3 and λ4, respectively.