Added "open_browser" option for show_html(), under-the-hood correlati…

…on error flagging

sweetviz/dataframe_report.py

@@ -10,6 +10,7 @@
 import sweetviz.series_analyzer as sa
 import sweetviz.utils as su
 from sweetviz.graph_associations import GraphAssoc
+from sweetviz.graph_associations import CORRELATION_ERROR
 from sweetviz.graph_legend import GraphLegend
 from sweetviz.config import config
 import sweetviz.sv_html as sv_html
@@ -268,24 +269,26 @@ def __getitem__(self, key):
  def __setitem__(self, key, value):
  self._features[key] = value
 
- def show_html(self, filepath='SWEETVIZ_REPORT.html', layout='widescreen'):
+ # OUTPUT
+ # ----------------------------------------------------------------------------------------------
+ def show_html(self, filepath='SWEETVIZ_REPORT.html', open_browser=True, layout='widescreen'):
  sv_html.load_layout_globals_from_config()
  self.page_layout = layout
  sv_html.set_summary_positions(self)
  sv_html.generate_html_detail(self)
  self._page_html = sv_html.generate_html_dataframe_page(self)
 
- # self.temp_folder = config["Files"].get("temp_folder")
- # os.makedirs(os.path.normpath(self.temp_folder), exist_ok=True)
-
  f = open(filepath, 'w', encoding="utf-8")
  f.write(self._page_html)
  f.close()
 
- print(f"Report {filepath} was generated! NOTEBOOK/COLAB USERS: no browser will pop up, the report is saved in your notebook/colab files.")
- # Not sure how to work around this: not fatal but annoying...Notebook/colab
- # https://bugs.python.org/issue5993
- webbrowser.open('file://' + os.path.realpath(filepath))
+ if open_browser:
+ print(f"Report {filepath} was generated! NOTEBOOK/COLAB USERS: no browser will pop up, the report is saved in your notebook/colab files.")
+ # Not sure how to work around this: not fatal but annoying...Notebook/colab
+ # https://bugs.python.org/issue5993
+ webbrowser.open('file://' + os.path.realpath(filepath))
+ else:
+ print(f"Report {filepath} was generated!")
 
  @staticmethod
  def get_predetermined_type(name: str,
@@ -425,15 +428,15 @@ def mirror_association(association_dict, feature_name, other_name, value):
  feature.source.corr(other.source, method='pearson')
  # TODO: display correlation error better in graph!
  if isnan(cur_associations[other.source.name]):
- cur_associations[other.source.name] = 0.0
+ cur_associations[other.source.name] = CORRELATION_ERROR
  mirror_association(self._associations, feature_name, other.source.name, \
  cur_associations[other.source.name])
  if process_compare:
  cur_associations_compare[other.source.name] = \
  feature.compare.corr(other.compare, method='pearson')
  # TODO: display correlation error better in graph!
  if isnan(cur_associations_compare[other.source.name]):
- cur_associations_compare[other.source.name] = 0.0
+ cur_associations_compare[other.source.name] = CORRELATION_ERROR
  mirror_association(self._associations_compare, feature_name, other.source.name, \
  cur_associations_compare[other.source.name])
  self.progress_bar.update(1)

sweetviz/graph_associations.py

@@ -46,6 +46,10 @@
 # A name for a custom index column that likely will not be used by users
 UNIQUE_INDEX_NAME = 'indexZZ8vr$#RVwadfaFASDFSA'
 
+# Something to detect correlation errors to display
+# TODO: Better/more intuitive display of correlation errors (right now just show up as empty)
+CORRELATION_ERROR = 83572398457329.0
+
 class GraphAssoc(sweetviz.graph.Graph):
  def __init__(self, dataframe_report, which_graph: str, association_data):
  self.set_style(["graph_base.mplstyle"])
@@ -207,6 +211,8 @@ def value_to_color(val):
  if color_min == color_max:
  return palette[-1]
  else:
+ if val == CORRELATION_ERROR:
+ return palette[(n_colors - 1)]
  val_position = float((val - color_min)) / (color_max - color_min) # position of value in the input range, relative to the length of the input range
  val_position = min(max(val_position, 0), 1) # bound the position betwen 0 and 1
  # LOG IT
@@ -234,6 +240,9 @@ def value_to_size(val):
  else:
  if val == 0:
  return 0.0
+ # TODO: Better/more intuitive display of correlation errors
+ if val == CORRELATION_ERROR:
+ return 0.0
  val_position = (val - size_min) * 0.999 / (size_max - size_min) + 0.001 # position of value in the input range, relative to the length of the input range
  val_position = min(max(val_position, 0), 1) # bound the position betwen 0 and 1
  # LOG IT