tree

Visualize a sklearn Decision Tree Classifier

• simple tree
  
• sankey tree
  

How to use?

Take simple_tree as an example:

1. Download the folder tree
2. The folder structure:

   tree
     - src: folder for all codes regarding the visualization
     - data: folder for the test data
     - simple_tree_output: folder for simple_tree outputs (will be re-generated if it was deleted)
     - sankey_tree_output: folder for sankey_tree outputs (will be re-generated if it was deleted)
     - generate_tree_test.ipynb: instructions and examples in jupyter notebook
     - ...
   

3. Install jinja2

   pip install jinja2
   

4. Use simple_tree or sankey_tree visualization: (visualization depends on D3.js, so you need to connect to the network)

   import sys
   sys.path.insert(0, 'src/')
   import generate_tree
   
   # simple tree
   generate_tree.generate_simple_tree(tree_title='Titanic_Tree', tree_model=dt, X=titanic[features], 
                                  target_names=['Not Survived', 'Survived'], target_colors = None,
                                  color_map=None, width=1500, height=1000)
   			   
   # sankey tree
   generate_tree.generate_sankey_tree(tree_title='Titanic_Tree', tree_model=dt, X=titanic[features], 
                                  target_names=['Not Survived', 'Survived'], target_colors = None,
                                  color_map=None, width=1500, height=1200)

5. A html file would be generated in simple_tree_output or sankey_tree_output folder. Open it using any browser you like (I like Chrome anyway).

Parameters

def generate_simple_tree(tree_title, tree_model, X, target_names, 
                         target_colors=None, color_map=None, width=None, height=None):
	'''
	visualize a sklearn Decision Tree Classifier

	:param tree_title: string
		name of the tree
	:param tree_model: a fitted sklearn Decision Tree Classifier
	:param X: pandas DataFrame
		dataset model was fitted on
	:param target_names: list
		list of names for targets
	:param target_colors: list, default=None
		list of colors for targets
	:param color_map: string, default=None
		matplotlib color map name, like 'Vega20'
	:param width: int
		width of the html page
	:param height: int
		height of the html page
	'''
	
def generate_sankey_tree(tree_title, tree_model, X, target_names,
			 target_colors=None, color_map=None, width=None, height=None):
	'''
	visualize a sklearn Decision Tree Classifier

	:param tree_title: string
		name of the tree
	:param tree_model: a fitted sklearn Decision Tree Classifier
	:param X: pandas DataFrame
		dataset model was fitted on
	:param target_names: list
		list of names for targets
	:param target_colors: list, default=None
		list of colors for targets
	:param color_map: string, default=None
		matplotlib color map name, like 'Vega20'
	:param width: int
		width of the html page
	:param height: int
		height of the html page
	'''