import Utility.parameter_generator as Pg
import Utility.bachelor_utilities as Bu
import Utility.network_training as Tr
from tensorflow import set_random_seed
from numpy.random import seed as set_numpy_seed
from random import shuffle, randint
import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
filename = r'../../../Logs/results.csv'
n_cv = 5
pg = Pg.ParameterGenerator()
pg.add_value('dense_layers', default_value=[139, 486, 152, 79, 61, 0, 0, 0, 0, 0])
pg.add_value('learning_rate', default_value=0.005)
pg.add_value('optimizer', default_value='adam')
pg.add_value('activation', default_value='relu')
pg.add_value('dropout', default_value=0.1)
pg.add_value('rnn_type', default_value='lstm')
pg.add_value('rnn_size', default_value=230)
pg.add_value('rnn_activation', default_value='tanh')
pg.add_value('rnn_dropout', default_value=0.1)
pg.add_value('last_activation', default_value='linear')
og_param = pg.sample(1, unique=True)[0]
parameters = []
for output_act in ['linear', 'relu', 'leaky_relu']:
seed = None
n_cv = 5
lr_grid = []
a = 7
lr_exp_range = range(-8, 0)
lr_cof_range = range(1, a * 10)
for e in lr_exp_range:
for c in lr_cof_range:
lr_grid.append(c / a * pow(10, e))
pg = Pg.ParameterGenerator(seed=seed)
pg.add_value('dense_layers',
default_value=[139, 486, 152, 79, 61, 0, 0, 0, 0, 0])
pg.add_value('learning_rate', default_value=0)
pg.add_value('optimizer', default_value='adam')
pg.add_value('activation', default_value='relu')
pg.add_value('dropout', default_value=0.1)
pg.add_value('rnn_type', default_value='lstm')
pg.add_value('rnn_size', default_value=311)
pg.add_value('rnn_activation', default_value='tanh')
pg.add_value('rnn_dropout', default_value=0.1)
pg.add_value('last_activation', default_value='linear')
og_param = pg.sample(1, unique=True)
parameters = []