def test_textnet_clustering(corpus):
"""Test calculating clustering coefficients."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases, connected=True)
assert len(n_np.clustering) == n_np.graph.vcount()
def test_save(tmpdir, corpus):
"""Test Textnet graph saving."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases)
out = tmpdir.join("graph.graphml")
n_np.save_graph(str(out))
assert len(tmpdir.listdir()) == 1
def test_plot_backbone(tmpdir, corpus):
"""Test ProjectedTextnet plotting with alpha cut."""
n = tn.Textnet(corpus.tokenized())
papers = n.project(node_type="doc")
out = tmpdir.join("plot-3.png")
plot = papers.plot(alpha=0.4, label_nodes=True, target=str(out))
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_plot_projected(tmpdir, corpus):
"""Test ProjectedTextnet plotting."""
n = tn.Textnet(corpus.tokenized())
papers = n.project(node_type="doc")
out = tmpdir.join("plot-2.png")
plot = papers.plot(show_clusters=True, label_nodes=True, target=str(out))
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_plot_layout(tmpdir, corpus):
"""Test Textnet plotting with bipartite layout and node labels."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases)
out = tmpdir.join("plot-1.png")
plot = n_np.plot(target=str(out), bipartite_layout=True, label_nodes=True)
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_plot(tmpdir, corpus):
"""Test Textnet plotting."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases)
out = tmpdir.join("plot-0.png")
plot = n_np.plot(target=str(out))
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_plot_scaled(tmpdir, corpus):
"""Test ProjectedTextnet plotting with scaled nodes."""
n = tn.Textnet(corpus.tokenized())
papers = n.project(node_type="doc")
out = tmpdir.join("plot-4.png")
plot = papers.plot(scale_nodes_by="betweenness",
label_nodes=True,
target=str(out))
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_textnet_birank(corpus):
"""Test calculating BiRank."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases, connected=True)
assert len(n_np.birank) == n_np.graph.vcount()
assert len(n_np.cohits) == n_np.graph.vcount()
assert len(n_np.hits) == n_np.graph.vcount()
bgrm = tn.network.bipartite_rank(n_np, normalizer="BGRM", max_iter=200)
assert len(bgrm) == n_np.graph.vcount()
def test_textnet(corpus):
"""Test Textnet class using sample data."""
noun_phrases = corpus.noun_phrases()
n_np = tn.Textnet(noun_phrases)
assert n_np.graph.vcount() > 0
assert n_np.graph.ecount() > 0
g_np_groups = n_np.project(node_type="doc")
assert g_np_groups.vcount() > 0
assert g_np_groups.ecount() > 0
g_np_words = n_np.project(node_type="term")
assert g_np_words.vcount() > 0
assert g_np_words.ecount() > 0
def test_plot_filtered(tmpdir, corpus):
"""Test ProjectedTextnet plotting filtered labels."""
n = tn.Textnet(corpus.tokenized())
papers = n.project(node_type="doc")
out = tmpdir.join("plot-5.png")
plot = papers.plot(
label_nodes=True,
label_edges=True,
node_label_filter=lambda v: v.degree() > 2,
edge_label_filter=lambda e: e["weight"] > 0.1,
target=str(out),
)
assert len(plot._objects) > 0
assert len(tmpdir.listdir()) == 1
def test_textnet_save_and_load(corpus, tmpdir):
"""Test roundtrip of saving and loading a textnet from file."""
out = tmpdir.join("out.textnet")
net = tn.Textnet(
corpus.tokenized(),
connected=True,
doc_attrs={"test": {
"New York Times": 1,
"Los Angeles Times": 3
}},
)
net.save(out)
loaded = tn.load_textnet(out)
assert net.nodes["id"] == loaded.nodes["id"]
assert net.edges["weight"] == loaded.edges["weight"]
assert net.summary == loaded.summary
def test_html_repr(corpus):
assert tn._repr_html_() != ""
assert corpus._repr_html_() != ""
assert tn.Textnet(corpus.tokenized())._repr_html_() != ""
def test_context(corpus):
"""Test formal context creation from textnet."""
n = tn.Textnet(corpus.tokenized(sublinear=False))
ctx = n.context
assert len(ctx) == 3
