import random import numpy as np import matplotlib.pyplot as plt from matplotlib.offsetbox import OffsetImage, AnnotationBbox import tensorflow from tensorflow.python.framework.ops import disable_eager_execution import pandas as pd import math disable_eager_execution() load_data = np.load('data/train_test_split_data.npz') # Data saved by the VAE # Convert Data to Tuples and Assign to respective variables box_matrix_train, box_density_train, additional_pixels_train, box_shape_train = tuple(load_data['box_matrix_train']), tuple(load_data['box_density_train']), tuple(load_data['additional_pixels_train']), tuple(load_data['box_shape_train']) box_matrix_test, box_density_test, additional_pixels_test, box_shape_test = tuple(load_data['box_matrix_test']), tuple(load_data['box_density_test']), tuple(load_data['additional_pixels_test']), tuple(load_data['box_shape_test']) testX = box_matrix_test # Shows the relationship to the MNIST Dataset vs the Shape Dataset image_size = np.shape(testX)[-1] # Determines the size of the images test_data = np.reshape(testX, (len(testX), image_size, image_size, 1)) # Creates tuples that contain all of the data generated allX = np.append(box_matrix_train,box_matrix_test, axis=0) all_box_density = np.append(box_density_train, box_density_test, axis=0) all_additional_pixels = np.append(additional_pixels_train, additional_pixels_test,axis=0) all_box_shape = np.append(box_shape_train, box_shape_test,axis=0) all_data = np.reshape(allX, (len(allX), image_size, image_size, 1)) # train_latent_points = [] # train_data = np.reshape(box_matrix_train, (len(box_matrix_train), image_size, image_size, 1)) # for i in range(len(box_shape_train)): # predicted_train = encoder_model_boxes.predict(np.array([train_data[i]])) # train_latent_points.append(predicted_train[0]) # train_latent_points = np.array(train_latent_points) shapes = ("basic_box", "diagonal_box_split", "horizontal_vertical_box_split", "back_slash_box", "forward_slash_box", "back_slash_plus_box", "forward_slash_plus_box", "hot_dog_box", "hamburger_box", "x_hamburger_box", "x_hot_dog_box", "x_plus_box") import math def basic_box_array(image_size): A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values # Creates the outside edges of the box for i in range(image_size): for j in range(image_size): if i == 0 or j == 0 or i == image_size - 1 or j == image_size - 1: A[i][j] = 1 return A def back_slash_array(image_size): A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for i in range(image_size): for j in range(image_size): if i == j: A[i][j] = 1 return A def forward_slash_array(image_size): A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for i in range(image_size): for j in range(image_size): if i == (image_size-1)-j: A[i][j] = 1 return A def hot_dog_array(image_size): # Places pixels down the vertical axis to split the box A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for i in range(image_size): for j in range(image_size): if j == math.floor((image_size - 1) / 2) or j == math.ceil((image_size - 1) / 2): A[i][j] = 1 return A def hamburger_array(image_size): # Places pixels across the horizontal axis to split the box A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for i in range(image_size): for j in range(image_size): if i == math.floor((image_size - 1) / 2) or i == math.ceil((image_size - 1) / 2): A[i][j] = 1 return A def center_array(image_size): A = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for i in range(image_size): for j in range(image_size): if i == math.floor((image_size-1)/2) and j == math.ceil((image_size-1)/2): A[i][j] = 1 if i == math.floor((image_size-1)/2) and j == math.floor((image_size-1)/2): A[i][j] = 1 if j == math.ceil((image_size-1)/2) and i == math.ceil((image_size-1)/2): A[i][j] = 1 if j == math.floor((image_size-1)/2) and i == math.ceil((image_size-1)/2): A[i][j] = 1 return A def update_array(array_original, array_new, image_size): A = array_original for i in range(image_size): for j in range(image_size): if array_new[i][j] == 1: A[i][j] = 1 return A def add_pixels(array_original, additional_pixels, image_size): # Adds pixels to the thickness of each component of the box A = array_original A_updated = np.zeros((int(image_size), int(image_size))) # Initializes A matrix with 0 values for dens in range(additional_pixels): for i in range(1, image_size - 1): for j in range(1, image_size - 1): if A[i - 1][j] + A[i + 1][j] + A[i][j - 1] + A[i][j + 1] > 0: A_updated[i][j] = 1 A = update_array(A, A_updated,image_size) return A def basic_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Creates the outside edges of the box # Increase the thickness of each part of the box A = add_pixels(A, additional_pixels, image_size) return A*density def horizontal_vertical_box_split(additional_pixels, density, image_size): A = basic_box_array(image_size) # Creates the outside edges of the box # Place pixels across the horizontal and vertical axes to split the box A = update_array(A, hot_dog_array(image_size), image_size) A = update_array(A, hamburger_array(image_size), image_size) # Increase the thickness of each part of the box A = add_pixels(A, additional_pixels, image_size) return A*density def diagonal_box_split(additional_pixels, density, image_size): A = basic_box_array(image_size) # Creates the outside edges of the box # Add pixels along the diagonals of the box A = update_array(A, back_slash_array(image_size), image_size) A = update_array(A, forward_slash_array(image_size), image_size) # Adds pixels to the thickness of each component of the box # Increase the thickness of each part of the box A = add_pixels(A, additional_pixels, image_size) return A*density def back_slash_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def forward_slash_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, forward_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def hot_dog_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hot_dog_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def hamburger_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hamburger_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def x_plus_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hot_dog_array(image_size), image_size) A = update_array(A, hamburger_array(image_size), image_size) A = update_array(A, forward_slash_array(image_size), image_size) A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def forward_slash_plus_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hot_dog_array(image_size), image_size) A = update_array(A, hamburger_array(image_size), image_size) A = update_array(A, forward_slash_array(image_size), image_size) # A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def back_slash_plus_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hot_dog_array(image_size), image_size) A = update_array(A, hamburger_array(image_size), image_size) # A = update_array(A, forward_slash_array(image_size), image_size) A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def x_hot_dog_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, hot_dog_array(image_size), image_size) # A = update_array(A, hamburger_array(image_size), image_size) A = update_array(A, forward_slash_array(image_size), image_size) A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def x_hamburger_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values # A = update_array(A, hot_dog_array(image_size), image_size) A = update_array(A, hamburger_array(image_size), image_size) A = update_array(A, forward_slash_array(image_size), image_size) A = update_array(A, back_slash_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density def center_box(additional_pixels, density, image_size): A = basic_box_array(image_size) # Initializes A matrix with 0 values A = update_array(A, center_array(image_size), image_size) A = add_pixels(A, additional_pixels, image_size) return A * density import tensorflow as tf sess = tf.compat.v1.Session() from keras import backend as K K.set_session(sess) # Gradio Interface import gradio import numpy endpoint_types = shapes density_options = ["{:.2f}".format(x) for x in numpy.linspace(0.1, 1, 10)] thickness_options = [str(int(x)) for x in numpy.linspace(0, 10, 11)] interpolation_options = [str(int(x)) for x in [3, 5, 10, 20]] def interpolate(t1, t2, d1, d2, th1, th2, steps): # Load the decoder model decoder_model_boxes = tensorflow.keras.models.load_model('data/decoder_model_boxes', compile=False) # # import the encoder model architecture json_file_loaded = open('data/model.json', 'r') loaded_model_json = json_file_loaded.read() # load model using the saved json file encoder_model_boxes = tensorflow.keras.models.model_from_json(loaded_model_json) # load weights into newly loaded_model encoder_model_boxes.load_weights('data/model_tf') num_internal = int(steps) number_1 = globals()[t1](int(th1), float(d1), 28) number_2 = globals()[t2](int(th2), float(d2), 28) # resize the array to match the prediction size requirement number_1_expand = np.expand_dims(np.expand_dims(number_1, axis=2), axis=0) number_2_expand = np.expand_dims(np.expand_dims(number_2, axis=2), axis=0) # Determine the latent point that will represent our desired number latent_point_1 = encoder_model_boxes.predict(number_1_expand)[0] latent_point_2 = encoder_model_boxes.predict(number_2_expand)[0] latent_dimensionality = len(latent_point_1) # define the dimensionality of the latent space num_interp = num_internal # the number of images to be pictured latent_matrix = [] # This will contain the latent points of the interpolation for column in range(latent_dimensionality): new_column = np.linspace(latent_point_1[column], latent_point_2[column], num_interp) latent_matrix.append(new_column) latent_matrix = np.array(latent_matrix).T # Transposes the matrix so that each row can be easily indexed plot_rows = 2 plot_columns = num_interp + 2 predicted_interps = [number_1_expand[0, :, :, 0]] for latent_point in range(2, num_interp + 2): # cycles the latent points through the decoder model to create images generated_image = decoder_model_boxes.predict(np.array([latent_matrix[latent_point - 2]]))[0] # generates an interpolated image based on the latent point predicted_interps.append(generated_image[:, :, -1]) predicted_interps.append(number_2_expand[0, :, :, 0]) transition_region = predicted_interps[0] for i in range(len(predicted_interps)-1): transition_region = numpy.hstack((transition_region, predicted_interps[i+1])) return transition_region def generate_unit_cell(t, d, th): number_1 = globals()[t](int(th), float(d), 28) # resize the array to match the prediction size requirement number_1_expand = np.expand_dims(np.expand_dims(number_1, axis=2), axis=0) return number_1_expand[0, :, :, 0] with gradio.Blocks() as demo: with gradio.Row(): with gradio.Column(): t1 = gradio.Dropdown(endpoint_types, label="Type 1", value=random.choice(endpoint_types)) d1 = gradio.Dropdown(density_options, label="Density 1", value=random.choice(density_options)) th1 = gradio.Dropdown(thickness_options, label="Thickness 1", value=random.choice(thickness_options)) img1 = gradio.Image(label="Endpoint 1") t1.change(fn=generate_unit_cell, inputs=[t1, d1, th1], outputs=[img1]) d1.change(fn=generate_unit_cell, inputs=[t1, d1, th1], outputs=[img1]) th1.change(fn=generate_unit_cell, inputs=[t1, d1, th1], outputs=[img1]) with gradio.Column(): t2 = gradio.Dropdown(endpoint_types, label="Type 2", value=random.choice(endpoint_types)) d2 = gradio.Dropdown(density_options, label="Density 2", value=random.choice(density_options)) th2 = gradio.Dropdown(thickness_options, label="Thickness 2", value=random.choice(thickness_options)) img2 = gradio.Image(label="Endpoint 2") t2.change(fn=generate_unit_cell, inputs=[t2, d2, th2], outputs=[img1]) d2.change(fn=generate_unit_cell, inputs=[t2, d2, th2], outputs=[img1]) th2.change(fn=generate_unit_cell, inputs=[t2, d2, th2], outputs=[img1]) steps = gradio.Dropdown(interpolation_options, label="Interpolation Length", value=random.choice(interpolation_options)) btn = gradio.Button("Run") img = gradio.Image(label="Transition") btn.click(fn=interpolate, inputs=[t1, t2, d1, d2, th1, th2, steps], outputs=[img]) examples = gradio.Examples(examples=[["hamburger_box", "hot_dog_box", "1.00", "1.00", "2", "2", "20"], ["hamburger_box", "hot_dog_box", "0.10", "1.00", "10", "10", "5"]], fn=interpolate, inputs=[t1, t2, d1, d2, th1, th2, steps], outputs=[img], cache_examples = True) demo.launch()