Exemplo n.º 1
0
def test_non_square_precomputed_distances():
    # Precomputed distance matrices must be square matrices.
    tsne = TSNE(metric="precomputed")
    assert_raises_regexp(ValueError, ".* square distance matrix",
                         tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 2
0
def test_early_exaggeration_too_small():
    # Early exaggeration factor must be >= 1.
    tsne = TSNE(early_exaggeration=0.99)
    assert_raises_regexp(ValueError, "early_exaggeration .*",
                         tsne.fit_transform, np.array([[0.0], [0.0]]))
Exemplo n.º 3
0
def test_too_few_iterations():
    # Number of gradient descent iterations must be at least 200.
    tsne = TSNE(n_iter=199)
    assert_raises_regexp(ValueError, "n_iter .*", tsne.fit_transform,
                         np.array([[0.0], [0.0]]))
Exemplo n.º 4
0
def test_n_components_range():
    # barnes_hut method should only be used with n_components <= 3
    tsne = TSNE(n_components=4, method="barnes_hut")
    assert_raises_regexp(ValueError, "'n_components' should be .*",
                         tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 5
0
import pandas as pd

from scipy.io import loadmat
from sklearn.manifold.t_sne import TSNE

chris_data = loadmat('/home/itskov/Downloads/dataForEyal.mat')
tnse_manifold = TSNE().fit_transform(chris_data['PCs'])
pass
Exemplo n.º 6
0
def test_init_ndarray_precomputed():
    # Initialize TSNE with ndarray and metric 'precomputed'
    # Make sure no FutureWarning is thrown from _fit
    tsne = TSNE(init=np.zeros((100, 2)), metric="precomputed")
    tsne.fit(np.zeros((100, 100)))
Exemplo n.º 7
0
def test_angle_out_of_range_checks():
    # check the angle parameter range
    for angle in [-1, -1e-6, 1 + 1e-6, 2]:
        tsne = TSNE(angle=angle)
        assert_raises_regexp(ValueError, "'angle' must be between 0.0 - 1.0",
                             tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 8
0
def test_too_few_iterations():
    # Number of gradient descent iterations must be at least 200.
    tsne = TSNE(n_iter=199)
    with pytest.raises(ValueError, match="n_iter .*"):
        tsne.fit_transform(np.array([[0.0], [0.0]]))
Exemplo n.º 9
0
def test_non_square_precomputed_distances():
    # Precomputed distance matrices must be square matrices.
    tsne = TSNE(metric="precomputed")
    with pytest.raises(ValueError, match=".* square distance matrix"):
        tsne.fit_transform(np.array([[0.0], [1.0]]))
Exemplo n.º 10
0
    img_path_dict = dict(zip(letters, paths))

    images = []
    for letter in letters:
        img = Image.open(img_path_dict[letter])
        img = img.crop((80, 50, 400, 365))
        img.thumbnail((16, 16), Image.ANTIALIAS)
        images.append(img)

    images_dict = dict(zip(letters, images))

    X, Y = get_input_data(
        '../letter_classification/letter_classification_train.data')
    images = []
    for letter in Y:
        images.append(images_dict[letter])

    tsne = TSNE(n_components=2,
                init='pca',
                verbose=1,
                learning_rate=600,
                n_iter=5000)
    t0 = time.time()
    X_tsne = tsne.fit_transform(X)

    plot_embedding(
        X_tsne, Y, X, images,
        "t-SNE embedding of the letters. Time: %.3f seconds" %
        (time.time() - t0))
    plt.show()
Exemplo n.º 11
0
def test_early_exaggeration_too_small():
    # Early exaggeration factor must be >= 1.
    tsne = TSNE(early_exaggeration=0.99)
    with pytest.raises(ValueError, match="early_exaggeration .*"):
        tsne.fit_transform(np.array([[0.0], [0.0]]))
Exemplo n.º 12
0
#create reduced feature matrix
reader=csv.reader(open("reduced_features.csv","r"),delimiter=",")
X=list(reader)
X=np.array(X)
X=X.astype(np.float)


#create result vector
reader=csv.reader(open("target_output.csv","r"),delimiter=",")
y=list(reader)
y=np.array(y)
y=y.astype(np.int)
y=y.ravel() 

X_Train_embedded = TSNE(n_components=2).fit_transform(X)
print X_Train_embedded.shape
model =  BernoulliNB().fit(X,y)
y_predicted = model.predict(X)
# replace the above by your data and model

# create meshgrid
resolution = 1024 # 100x100 background pixels
X2d_xmin, X2d_xmax = np.min(X_Train_embedded[:,0]), np.max(X_Train_embedded[:,0])
X2d_ymin, X2d_ymax = np.min(X_Train_embedded[:,1]), np.max(X_Train_embedded[:,1])
xx, yy = np.meshgrid(np.linspace(X2d_xmin, X2d_xmax, resolution), np.linspace(X2d_ymin, X2d_ymax, resolution))

# approximate Voronoi tesselation on resolution x resolution grid using 1-NN
background_model = KNeighborsClassifier(n_neighbors=1).fit(X_Train_embedded, y_predicted) 
voronoiBackground = background_model.predict(np.c_[xx.ravel(), yy.ravel()])
voronoiBackground = voronoiBackground.reshape((resolution, resolution))
Exemplo n.º 13
0
def test_exact_no_precomputed_sparse():
    tsne = TSNE(metric='precomputed', method='exact')
    with pytest.raises(TypeError, match='sparse'):
        tsne.fit_transform(sp.csr_matrix([[0, 5], [5, 0]]))
Exemplo n.º 14
0
def test_bad_precomputed_distances(method, D, retype, message_regex):
    tsne = TSNE(metric="precomputed", method=method)
    with pytest.raises(ValueError, match=message_regex):
        tsne.fit_transform(retype(D))
Exemplo n.º 15
0
def test_init_not_available():
    # 'init' must be 'pca', 'random', or numpy array.
    tsne = TSNE(init="not available")
    m = "'init' must be 'pca', 'random', or a numpy array"
    assert_raises_regexp(ValueError, m, tsne.fit_transform,
                         np.array([[0.0], [1.0]]))
Exemplo n.º 16
0
def test_init_not_available():
    # 'init' must be 'pca', 'random', or numpy array.
    tsne = TSNE(init="not available")
    m = "'init' must be 'pca', 'random', or a numpy array"
    with pytest.raises(ValueError, match=m):
        tsne.fit_transform(np.array([[0.0], [1.0]]))
Exemplo n.º 17
0
def test_init_ndarray():
    # Initialize TSNE with ndarray and test fit
    tsne = TSNE(init=np.zeros((100, 2)))
    X_embedded = tsne.fit_transform(np.ones((100, 5)))
    assert_array_equal(np.zeros((100, 2)), X_embedded)
Exemplo n.º 18
0
def test_method_not_available():
    # 'nethod' must be 'barnes_hut' or 'exact'
    tsne = TSNE(method='not available')
    with pytest.raises(ValueError, match="'method' must be 'barnes_hut' or "):
        tsne.fit_transform(np.array([[0.0], [1.0]]))
Exemplo n.º 19
0
def test_method_not_available():
    # 'nethod' must be 'barnes_hut' or 'exact'
    tsne = TSNE(method='not available')
    assert_raises_regexp(ValueError, "'method' must be 'barnes_hut' or ",
                         tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 20
0
def test_n_components_range():
    # barnes_hut method should only be used with n_components <= 3
    tsne = TSNE(n_components=4, method="barnes_hut")
    with pytest.raises(ValueError, match="'n_components' should be .*"):
        tsne.fit_transform(np.array([[0.0], [1.0]]))
Exemplo n.º 21
0
def test_pca_initialization_not_compatible_with_precomputed_kernel():
    # Precomputed distance matrices must be square matrices.
    tsne = TSNE(metric="precomputed", init="pca")
    assert_raises_regexp(ValueError, "The parameter init=\"pca\" cannot be "
                         "used with metric=\"precomputed\".",
                         tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 22
0
def test_distance_not_available():
    """'metric' must be valid."""
    tsne = TSNE(metric="not available")
    assert_raises_regexp(ValueError, "Unknown metric not available.*",
                         tsne.fit_transform, np.array([[0.0], [1.0]]))
Exemplo n.º 23
0
def test_chebyshev_metric():
    # t-SNE should allow metrics that cannot be squared (issue #3526).
    random_state = check_random_state(0)
    tsne = TSNE(metric="chebyshev")
    X = random_state.randn(5, 2)
    tsne.fit_transform(X)
Exemplo n.º 24
0
            name = image.split('/')[-1]
            if name in valid_names:
                img = open_image(image)
                pred = learn.predict(img)
                embedding = pred[-1].detach().numpy()

                x.append(embedding)
                y.append(n_class)

        n_class += 1

    x, y = np.asarray(x), np.asarray(y)
    print(x.shape, y.shape)
    print(np.unique(y))

    x_tsne = TSNE(n_components=2).fit_transform(x)
    print(x_tsne.shape)

    color = ['darkcyan', 'g', 'y', 'magenta', 'lightgreen', 'mediumslateblue', 'yellow', 'brown', 'k', 'b']
    labels = [str(i) for i in range(10)]
    times = [0 for i in range(10)]
    for i in range(len(x_tsne)):
        if times[y[i]] == 0:
            plt.scatter(x_tsne[i, 0], x_tsne[i, 1], color=color[y[i]], label=labels[y[i]])
            times[y[i]] += 1
        else:
            plt.scatter(x_tsne[i, 0], x_tsne[i, 1], color=color[y[i]])
    plt.legend()
    plt.title('TSNE validation')
    plt.savefig('tsne_val.png')