def print_confusion(sinal_predicted_matrix, labels_signals, labels_model, no_numbers=False, norm=True): print(sinal_predicted_matrix) plot_confusion_matrix(sinal_predicted_matrix, labels_signals, labels_model, no_numbers, norm=norm)
def print_confusion(sinal_predicted_matrix, labels_signals, labels_model, no_numbers=False): print(sinal_predicted_matrix) # cmap = make_cmap(get_color(), max_colors=1000) plot_confusion_matrix(sinal_predicted_matrix, labels_signals, labels_model, no_numbers, norm=True) # , cmap=cmap)
def print_mean_loss(Mod, Sig, loss_tensor, signals_models, signals_tests): labels_model = np.asarray(np.zeros(len(Mod) * 2, dtype=np.str), dtype=np.object) labels_signals = np.asarray(np.zeros(len(Sig) * 2, dtype=np.str), dtype=np.object) labels_model[list(range(1, len(Mod) * 2, 2))] = [signals_models[i]["s_name"] for i in Mod] labels_signals[list(range(1, len(Sig) * 2, 2))] = [signals_tests[i][-1] for i in Sig] mean_values_matrix = np.mean(loss_tensor, axis = 2) sinal_predicted_matrix = np.zeros(len(Sig)) # for i in range(np.shape(sinal_predicted_matrix)[0]): for j in range(np.shape(sinal_predicted_matrix)[0]): sinal_predicted_matrix[j] = mean_values_matrix[0, j] print(sinal_predicted_matrix) # cmap = make_cmap(get_color(), max_colors=1000) plot_confusion_matrix(sinal_predicted_matrix.T, labels_model, labels_signals) # , cmap=cmap)
def print_confusion(Mod, Sig, loss_tensor, signals_models, signals_tests): labels_model = np.asarray(np.zeros(len(Mod) * 2, dtype=np.str), dtype=np.object) labels_signals = np.asarray(np.zeros(len(Sig) * 2, dtype=np.str), dtype=np.object) labels_model[list(range(1, len(Mod) * 2, 2))] = [signals_models[i]["s_name"] for i in Mod] labels_signals[list(range(1, len(Sig) * 2, 2))] = [signals_tests[i][-1] for i in Sig] predicted_matrix = np.argmin(loss_tensor[Mod][:, Sig, :], axis=0) sinal_predicted_matrix = np.zeros((len(Sig), len(Mod))) for i in range(np.shape(sinal_predicted_matrix)[0]): for j in range(np.shape(sinal_predicted_matrix)[1]): sinal_predicted_matrix[i, j] = sum(predicted_matrix[i, :] == j) print(sinal_predicted_matrix) # cmap = make_cmap(get_color(), max_colors=1000) plot_confusion_matrix(sinal_predicted_matrix, labels_model, labels_signals) # , cmap=cmap)