def test_plot_1d_cluster_custom_1():
data_reduced_1d = reducer(data, ndims=1)
NUM_CLUSTERS = 4
geo = plot.plot(data_reduced_1d,
cluster='KMeans',
n_clusters=NUM_CLUSTERS,
show=False)
assert all([i.shape[1] == 1
for i in geo.data]) and len(geo.ax.get_lines()) == NUM_CLUSTERS
def test_plot_2d_cluster():
data_reduced_2d = reducer(data, ndims=2)
NUM_CLUSTERS = 4
geo = plot.plot(data_reduced_2d,
cluster={
'model': 'KMeans',
'params': {
'n_clusters': NUM_CLUSTERS
}
},
show=False)
assert all([i.shape[1] == 2
for i in geo.data]) and len(geo.ax.get_lines()) == NUM_CLUSTERS
def test_plot_3d_animate_cluster():
NUM_CLUSTERS = 4
data_reduced_3d = reducer(data, ndims=3)
geo = plot.plot(data_reduced_3d,
cluster={
'model': 'KMeans',
'params': {
'n_clusters': NUM_CLUSTERS
}
},
animate=True,
show=False)
assert all([i.shape[1] == 3 for i in geo.data])
def test_plot_2d_animate_cluster():
NUM_CLUSTERS = 4
data_reduced_2d = reducer(data, ndims=2)
with pytest.raises(Exception) as e_info:
plot.plot(data_reduced_2d,
cluster={
'model': 'KMeans',
'params': {
'n_clusters': NUM_CLUSTERS
}
},
animate=True,
show=False)
def test_reduce_params_UMAP():
from umap import UMAP
data1 = np.random.rand(20, 10)
params = {
'n_neighbors': 5,
'n_components': 2,
'metric': 'correlation',
'random_state': 1234
}
# testing override of n_dims by n_components. Should raise UserWarning due to conflict
hyp_data = reducer(data1,
reduce={
'model': 'UMAP',
'params': params
},
ndims=3)
umap_data = UMAP(**params).fit_transform(data1)
np.testing.assert_array_equal(hyp_data, umap_data)
def test_reduce_geo():
geo = plot(data, show=False)
reduced_data_3d = reducer(geo, ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_IncrementalPCA():
reduced_data_3d = reducer(data, reduce='IncrementalPCA', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_dims_2d():
reduced_data_2d = reducer(data, ndims=2)
assert reduced_data_2d[0].shape == (10, 2)
def test_reduce_dims_1d():
reduced_data_1d = reducer(data, ndims=1)
assert reduced_data_1d[0].shape == (10, 1)
def test_reduce_TSNE():
reduced_data_3d = reducer(data, reduce='TSNE', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_SparsePCA():
reduced_data_3d = reducer(data, model='SparsePCA', ndims=3)
assert reduced_data_3d[0].shape == (100, 3)
def test_reduce_TruncatedSVD():
reduced_data_3d = reducer(data, reduce='TruncatedSVD', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_FastICA():
reduced_data_3d = reducer(data, reduce='FastICA', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_MDS():
reduced_data_3d = reducer(data, reduce='MDS', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_UMAP():
reduced_data_3d = reducer(data, reduce='UMAP', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_LocallyLinearEmbedding():
reduced_data_3d = reducer(data, reduce='LocallyLinearEmbedding', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_SpectralEmbedding():
reduced_data_3d = reducer(data, reduce='SpectralEmbedding', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_Isomap():
reduced_data_3d = reducer(data, reduce='Isomap', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_MiniBatchSparsePCA():
reduced_data_3d = reducer(data, reduce='MiniBatchSparsePCA', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_plot_2d_animate():
data_reduced_2d = reducer(data, ndims=2)
with pytest.raises(Exception) as e_info:
plot.plot(data_reduced_2d, animate=True, show=False)
def test_reduce_KernelPCA():
reduced_data_3d = reducer(data, reduce='KernelPCA', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_plot_2d():
data_reduced_2d = reducer(data, ndims=2)
geo = plot.plot(data_reduced_2d, show=False)
assert all([i.shape[1] == 2 for i in geo.data])
def test_reduce_FactorAnalysis():
reduced_data_3d = reducer(data, reduce='FactorAnalysis', ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
# -*- coding: utf-8 -*-
from builtins import range
import numpy as np
from hypertools.tools.reduce import reduce as reducer
from hypertools.plot.plot import plot
data = [
np.random.multivariate_normal(np.zeros(4), np.eye(4), size=10)
for i in range(2)
]
reduced_data_2d = reducer(data, ndims=2)
reduced_data_1d = reducer(data, ndims=1)
def test_reduce_is_list():
reduced_data_3d = reducer(data)
assert type(reduced_data_3d) is list
def test_reduce_is_array():
reduced_data_3d = reducer(data, ndims=3)
assert isinstance(reduced_data_3d[0], np.ndarray)
def test_reduce_dims_3d():
reduced_data_3d = reducer(data, ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_MiniBatchDictionaryLearning():
reduced_data_3d = reducer(data,
reduce='MiniBatchDictionaryLearning',
ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
def test_reduce_is_list():
reduced_data_3d = reducer(data)
assert type(reduced_data_3d) is list
def test_plot_1d():
data_reduced_1d = reducer(data, ndims=1)
geo = plot.plot(data_reduced_1d, show=False)
assert all([i.shape[1] == 1 for i in geo.data])
def test_reduce_is_array():
reduced_data_3d = reducer(data, ndims=3)
assert isinstance(reduced_data_3d[0], np.ndarray)
def test_plot_3d():
data_reduced_3d = reducer(data, ndims=3)
geo = plot.plot(data_reduced_3d, show=False)
assert all([i.shape[1] == 3 for i in geo.data])
def test_reduce_dims_3d():
reduced_data_3d = reducer(data, ndims=3)
assert reduced_data_3d[0].shape == (10, 3)
