Ejemplo n.º 1
0
    args=parser.parse_args()

    current_working_dir = os.getcwd()
    print('current_working_dir: ', current_working_dir)
    pre = Preprocessing('fer2013', root_dir=current_working_dir)

    pre.load_data('train_reduced_norm.csv.gz', name='train')
    pre.load_data('test_public_norm.csv.gz', name='val')

    X = pre.get('val').drop(columns=['emotion'])
    y = pre.get('val')['emotion']
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.20, random_state = 42)
    val = pd.DataFrame(X_test)
    val['emotion'] = y_test
    pre.set(name='val', value=val)

    print(pre.get(name='val').head())

    train_pixels = pre.get(name='train').drop(columns=['emotion'])
    val_pixels = pre.get(name='val').drop(columns=['emotion'])

    print('data loaded')

    img_conv = ImageConverter()

    train_pixel_np = train_pixels.values
    train_pixel_224_np = np.zeros(shape=[train_pixel_np.shape[0], 224 * 224])
    for i in range(train_pixel_np.shape[0]):
        image = train_pixel_np[i].reshape(48, 48)
        newimg = img_conv.upscale(image)
Ejemplo n.º 2
0
    X_train_encoded = trained_model.encoder(X_train)
    X_test_encoded = trained_model.encoder(X_test)
    X_test_decoded = trained_model.decoder(X_test_encoded)

    X_train_encoded_df = pd.DataFrame(X_train_encoded.detach().numpy())
    X_test_encoded_df = pd.DataFrame(X_test_encoded.detach().numpy())

    cols = list(range(1, n_features_encoded + 1))

    X_train_encoded_df.columns = cols
    X_test_encoded_df.columns = cols

    train_encoded_data = X_train_encoded_df.join(y_train_df)
    test_encoded_data = X_test_encoded_df.join(y_test_df)

    pre.set('train_encoded', train_encoded_data)
    pre.set('test_encoded', test_encoded_data)

    pre.save('train_encoded')
    pre.save('test_encoded')

    plt.figure(1, figsize=(30, 20))
    for idx in range(30):
        image = X_test[idx].detach().numpy().reshape(48, 48)
        image2 = X_test_encoded[idx].detach().numpy().reshape(
            int(math.sqrt(n_features_encoded)),
            int(math.sqrt(n_features_encoded)))
        image3 = X_test_decoded[idx].detach().numpy().reshape(48, 48)
        # Call signature: subplot(nrows, ncols, index, **kwargs)
        plt.subplot(10, 9, 1 + idx * 3)
        plt.imshow(image, cmap='hot', interpolation='none')