Cleanup naming in Python code

Author: jonnor

MANIFEST.in

@@ -1,3 +1,3 @@
 include LICENSE.md
 include requirements*.txt
-include emtrees/*.h

emlearn/__init__.py

@@ -1,5 +1,8 @@
 
-from .randomforest import convert
+from . import common
 
-includedir = randomforest.get_include_dir()
+
+includedir = common.get_include_dir()
 

emlearn/common.py

@@ -0,0 +1,8 @@
+
+import os
+import os.path
+
+def get_include_dir():
+    return os.path.join(os.path.dirname(__file__))
+
+

emlearn/convert.py

@@ -0,0 +1,11 @@
+
+from . import trees
+
+def convert(estimator, kind=None, method='pymodule'):
+    if kind is None:
+        kind = type(estimator).__name__
+
+    if kind in ['RandomForestClassifier', 'ExtraTreesClassifier']:
+        return trees.Wrapper(estimator, method) 
+    else:
+        raise ValueError("Unknown model type: '{}'".format(kind))

emlearn/randomforest.py renamed to emlearn/trees.py

@@ -8,13 +8,10 @@
 
 import numpy
 
+from . import common
 import emtreesc
 
 
-def get_include_dir():
-    return os.path.join(os.path.dirname(__file__))
-
-
 # Tree representation as 2d array
 # feature, value, left_child, right_child
 # Leaf node: -1, class, -1, -1
@@ -378,7 +375,7 @@ def run_classifier(bin_path, data):
 class CompiledClassifier():
     def __init__(self, cmodel, name, call=None, include_dir=None, temp_dir='tmp/'):
         if include_dir == None:
-            include_dir = get_include_dir()
         self.bin_path = build_classifier(cmodel, name, include_dir=include_dir, temp_dir=temp_dir, func=call) 
 
     def predict(self, X):
@@ -432,13 +429,3 @@ def to_dot(self, **kwargs):
         return forest_to_dot(self.forest_, **kwargs)
 
 
-def convert(estimator, kind=None, method='pymodule'):
-    if kind is None:
-        kind = type(estimator).__name__
-
-    if kind in ['RandomForestClassifier', 'ExtraTreesClassifier']:
-        return Wrapper(estimator, method) 
-    else:
-        raise ValueError("Unknown model type: '{}'".format(kind))
-
-

examples/digits.py

@@ -1,5 +1,6 @@
 
-import emtrees
+
 import numpy
 from sklearn.model_selection import train_test_split
 from sklearn.ensemble import RandomForestClassifier
@@ -28,7 +29,7 @@
 m = numpy.max(Xtrain), numpy.min(Xtrain)
 
 filename = 'digits.h'
-cmodel = emtrees.convert(model)
 code = cmodel.save(file=filename)
 
 print('Wrote C code to', filename)

test/test_trees.py

@@ -55,7 +55,7 @@ def test_deduplicate_single_tree():
     ]
     roots = [ 6 ]
 
-    de_nodes, de_roots = emlearn.randomforest.remove_duplicate_leaves((nodes, roots))
 
     duplicates = 1
     assert len(de_roots) == len(roots)