if (path.exists('training_progress.csv')): progress = np.loadtxt('training_progress.csv', delimiter=',').tolist() else: progress = [] if (path.exists('models/min_model.h5')): model = load_model(location='models/min_model.h5') else: model = load_model() model.compile(optimizer='adam', loss='mse', metrics=['mae']) X_train, Y_train, X_train_gender = load_training_data() X_dev, Y_dev, X_dev_gender = load_development_data() if (path.exists('learner_params.txt')): learner_params = np.loadtxt('learner_params.txt') min_loss_dev = learner_params[0] min_mae = learner_params[1] prev_loss_dev = learner_params[4] loss_dev = [learner_params[2], learner_params[3]] increase_count = int(learner_params[5]) current_epoch_number = int(learner_params[6]) total_epoch_count = int(learner_params[7]) + 1 else: min_loss_dev = 10000 min_mae = 10000
import os from os import path import random #os.environ["CUDA_VISIBLE_DEVICES"]="3,4,5,6" training_progress = [] development_progress = [] test_progress = [] model = load_model() model.compile(optimizer='adagrad', loss='mse', metrics=['mae']) X_train, Y_train = load_training_data() X_dev, Y_dev = load_development_data() X_test, Y_test = load_test_data() min_mse_dev = 10000 min_mae_dev = 10000 min_mse_test = 10000 min_mae_test = 10000 current_epoch_number = 1 total_epoch_count = 100 m = X_train.shape[0] batch_size_list = list(range(1, m)) print("\n\n")
import numpy as np import sklearn.metrics from load_data import load_development_data from load_model import load_model import os os.environ["CUDA_VISIBLE_DEVICES"] = "1" if __name__ == "__main__": model = load_model() model.load_weights('optimal_weights.h5') dev_COVAREP_X_FORMANT, dev_facial_X_pose, dev_gaze_X_action, dev_transcript, dev_Y, dev_X_gender = load_development_data( ) model.compile(loss='mse', optimizer='adam', metrics=['mean_absolute_error']) dev_Y_hat = model.predict( [dev_facial_X_pose, dev_gaze_X_action, dev_transcript, dev_X_gender]) dev_Y = np.array(dev_Y) dev_Y_hat = dev_Y_hat.reshape((dev_Y.shape[0], )) RMSE = np.sqrt(sklearn.metrics.mean_squared_error(dev_Y, dev_Y_hat)) MAE = sklearn.metrics.mean_absolute_error(dev_Y, dev_Y_hat) EVS = sklearn.metrics.explained_variance_score(dev_Y, dev_Y_hat)