crawftv

syntax enable                                                                                                      
set nu                                                                                                             
highlight BadWhitespace ctermbg=red guibg=red                                                                      
au BufNewFile,BufRead *.py                                                                                         
    \ set fileformat=unix |                                                                                        
    \ set encoding=utf-8 |                                                                                         
    \ match BadWhitespace /\s\+$/                                                                                  
set tabstop=4                                                                                                      
set shiftwidth=4                                                                                

crawftv

from keras.datasets import mnist
(X_train, y_train), (X_test, y_test) = mnist.load_data()

#Scale the data to between 0 and 1
X_train = X_train/ 255
X_test = X_test/ 255

#Flatten arrays from (28x28) to (784x1)
X_train = X_train.reshape(60000,784)
X_test = X_test.reshape(10000,784)

crawftv

# the usual imports for a vanilla nueral net
import keras
from keras.models import Sequential
from keras.layers import Dense

model = Sequential()
model.add(Dense(16, input_shape=input_shape, activation='relu',name = 'input_layer'))
model.add(Dense(16, activation='relu', name="hidden_layer"))

model.add(Dense(10,activation='softmax',name="output_layer"))

crawftv

#imports we know we'll need
import skopt
# !pip install scikit-optimize if  necessary
from skopt import gbrt_minimize, gp_minimize
from skopt.utils import use_named_args
from skopt.space import Real, Categorical, Integer  

import keras
from keras.models import Sequential
from keras.layers import Dense

crawftv

dim_learning_rate = Real(low=1e-4, high=1e-2, prior='log-uniform',
                         name='learning_rate')
dim_num_dense_layers = Integer(low=1, high=5, name='num_dense_layers')
dim_num_input_nodes = Integer(low=1, high=512, name='num_input_nodes')
dim_num_dense_nodes = Integer(low=1, high=28, name='num_dense_nodes')
dim_activation = Categorical(categories=['relu', 'sigmoid'],
                             name='activation')
dim_batch_size = Integer(low=1, high=128, name='batch_size')
dim_adam_decay = Real(low=1e-6,high=1e-2,name="adam_decay")

crawftv

from keras.optimizers import Adam
def create_model(learning_rate, num_dense_layers,num_input_nodes,
                 num_dense_nodes, activation, adam_decay):
    #start the model making process and create our first layer
    model = Sequential()
    model.add(Dense(num_input_nodes, input_shape= input_shape, activation=activation
                   ))
    #create a loop making a new dense layer for the amount passed to this model.
    #naming the layers helps avoid tensorflow error deep in the stack trace.
    for i in range(num_dense_layers):

crawftv

@use_named_args(dimensions=dimensions)
def fitness(learning_rate, num_dense_layers, num_input_nodes, 
            num_dense_nodes,activation, batch_size,adam_decay):

    model = create_model(learning_rate=learning_rate,
                         num_dense_layers=num_dense_layers,
                         num_input_nodes=num_input_nodes,
                         num_dense_nodes=num_dense_nodes,
                         activation=activation,
                         adam_decay=adam_decay

crawftv

K.clear_session()
tensorflow.reset_default_graph()

crawftv

gp_result = gp_minimize(func=fitness,
                            dimensions=dimensions,
                            n_calls=12,
                            noise= 0.01,
                            n_jobs=-1,
                            kappa = 5,
                            x0=default_parameters)

crawftv

gbrt_result = gbrt_minimize(func=fitness,
                            dimensions=dimensions,
                            n_calls=12,
                            n_jobs=-1,
                            x0=default_parameters)