optiluca

# https://ax.dev/tutorials/multiobjective_optimization.html

import warnings

import ax

import torch
import numpy as np

from ax.metrics.noisy_function import NoisyFunctionMetric

optiluca

import numpy as np

import torch

if torch.cuda.is_available():
    print('Using CUDA...')
    torch.set_default_tensor_type(torch.cuda.FloatTensor)

# Ax Service API seems like the right level of abstraction?
from ax.service.ax_client import AxClient

optiluca

import pygmo as pg

class TestProblem:

    def fitness(self, x):
        param_1 = x[0]
        param_2 = x[1]

        metric_1 = 2 * param_1
        metric_2 = 2 * param_2

optiluca

import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
import tensorflow as tf
import numpy as np

BATCH_SIZE = 8
WINDOW_SIZE = 5

optiluca

import os

# os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
import tensorflow as tf

root_dir = os.path.dirname(os.path.abspath(__file__))

os.makedirs('test_files', exist_ok=True)

for i_dummy_file in range(10):

optiluca

optiluca

import numpy as np
import tensorflow as tf


# Write test tfrecord
def _serialise_element(value):
    elements = {'label_1': _float_feature(float(value))}
    elements_proto = tf.train.Example(features=tf.train.Features(feature=elements))
    return elements_proto.SerializeToString()

optiluca

import numpy as np
from keras.layers.core import Dense, Activation, Dropout
from keras.layers import CuDNNLSTM, LSTM
from keras.models import Sequential
from keras.models import load_model
from keras import optimizers
from keras.engine.saving import preprocess_weights_for_loading

sample_size = 100