from helpers.DataLoader import DataLoader from recsom.RecSom import RecSom from plotting_helpers.plot_utils import * from helpers.norms import * dim = 26 rows = 30 cols = 30 metric = euclidean_distance top_left = np.array((0, 0)) bottom_right = np.array((rows - 1, cols - 1)) lambda_s = metric(top_left, bottom_right) * 0.5 train_data = DataLoader.load_data('random_strings') for i in range(1): model = MergeSom(dim, rows, cols) model.train(train_data, metric=metric, alpha_s=1.0, alpha_f=0.05, lambda_s=lambda_s, lambda_f=1, eps=20, in3d=False, trace=True, trace_interval=5, sliding_window_size=10, log=True,
from helpers.DataLoader import DataLoader from recsom.RecSom import RecSom from helpers.norms import * dimension = 26 number_of_rows = 30 number_of_columns = 30 metric = euclidean_distance top_left = np.array((0, 0)) bottom_right = np.array((number_of_rows - 1, number_of_columns - 1)) lambda_s = metric(top_left, bottom_right) * 0.5 train_data = DataLoader.load_data('abcd_short') alpha_values = [x*.01 for x in range(0, 101)] for alpha in alpha_values: log_file_name = 'abs.csv' model = RecSom(input_dimension=dimension, rows_count=number_of_rows, columns_count=number_of_columns) model.train(train_data, metric=metric, alpha_s=1.0, alpha_f=0.05, lambda_s=lambda_s, lambda_f=1, eps=20, in3d=False, trace=False, trace_interval=5, sliding_window_size=10, log=True, log_file_name=log_file_name, alpha=alpha) # print(model.distances_between_adjacent_neurons_horizontal()) # print(model.distances_between_adjacent_neurons_vertical())