Issue241 rcm param and ducktyping for models

tests/test_topic_model.py

@@ -62,6 +62,22 @@ def test_n_topics():
         assert TopicModel(model, n_topics=20).n_topics == 20
 
 
+def test_duck_typing():
+    class TrainedDummyModel():
+        def __init__(self):
+            self.n_topics = 5
+            self.components_ = np.array([[0,0,0,1], [1,0,0,0]])
+
+        def transform(self, text):
+            return text
+    dummy = TrainedDummyModel()
+    tmodel = TopicModel(dummy)
+
+    assert tmodel.n_topics == dummy.n_topics
+    assert tmodel.model.transform == dummy.transform
+    np.testing.assert_array_equal(tmodel.model.components_, dummy.components_)
+
+
 def test_init_model():
     expecteds = (NMF, LatentDirichletAllocation, TruncatedSVD)
     models = ["nmf", "lda", "lsa"]

textacy/tm/topic_model.py

@@ -21,7 +21,8 @@
 class TopicModel(object):
     """
     Train and apply a topic model to vectorized texts using scikit-learn's
-    implementations of LSA, LDA, and NMF models. Inspect and visualize results.
+    implementations of LSA, LDA, and NMF models. Also any other topic model implementations that have
+    `component_`, `n_topics` and `transform` attributes. Inspect and visualize results.
     Save and load trained models to and from disk.
 
     Prepare a vectorized corpus (i.e. document-term matrix) and corresponding
@@ -103,9 +104,13 @@ class TopicModel(object):
         - http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.TruncatedSVD.html
     """
 
+    _required_trained_model_attr = {"transform", "components_", "n_topics"}
+
     def __init__(self, model, n_topics=10, **kwargs):
         if isinstance(model, (NMF, LatentDirichletAllocation, TruncatedSVD)):
             self.model = model
+        elif all(hasattr(model, required_attr) for required_attr in self._required_trained_model_attr):
+            self.model = model
         else:
             self.init_model(model, n_topics=n_topics, **kwargs)
 
@@ -365,6 +370,7 @@ def termite_plot(
         rank_terms_by="topic_weight",
         sort_terms_by="seriation",
         save=False,
+        rc_params=None,
     ):
         """
         Make a "termite" plot for assessing topic models using a tabular layout
@@ -392,6 +398,8 @@ def termite_plot(
                 the default ("seriation") groups similar terms together, which
                 facilitates cross-topic assessment
             save (str): give the full /path/to/fname on disk to save figure
+                rc_params (dict, optional): allow passing parameters to rc_context in matplotlib.plyplot,
+                details in https://matplotlib.org/3.1.0/api/_as_gen/matplotlib.pyplot.rc_context.html
 
         Returns:
             ``matplotlib.axes.Axes.axis``: Axis on which termite plot is plotted.
@@ -522,4 +530,5 @@ def termite_plot(
             term_labels,
             highlight_cols=highlight_cols,
             save=save,
+            rc_params=rc_params,
         )

textacy/viz/termite.py

@@ -71,6 +71,7 @@ def draw_termite_plot(
     highlight_cols=None,
     highlight_colors=None,
     save=False,
+    rc_params=None,
 ):
     """
     Make a "termite" plot, typically used for assessing topic models with a tabular
@@ -87,6 +88,8 @@ def draw_termite_plot(
             of (light/dark) matplotlib-friendly colors used to highlight a single
             column; if not specified (default), a good set of 6 pairs are used
         save (str, optional): give the full /path/to/fname on disk to save figure
+        rc_params (dict, optional): allow passing parameters to rc_context in matplotlib.plyplot,
+            details in https://matplotlib.org/3.1.0/api/_as_gen/matplotlib.pyplot.rc_context.html
 
     Returns:
         :obj:`matplotlib.axes.Axes.axis`: Axis on which termite plot is plotted.
@@ -138,6 +141,10 @@ def draw_termite_plot(
             raise ValueError(msg)
         highlight_colors = {hc: COLOR_PAIRS[i] for i, hc in enumerate(highlight_cols)}
 
+    _rc_params = RC_PARAMS.copy()
+    if rc_params:
+        _rc_params.update(rc_params)
+
     with plt.rc_context(RC_PARAMS):
         fig, ax = plt.subplots(figsize=(pow(n_cols, 0.8), pow(n_rows, 0.66)))