Nicer plot

multilabel_oversampling/multilabel_oversampling.py

@@ -92,16 +92,13 @@ def fit(self, df, target_list=["y1", "y2", "y3", "y4"]):
  print(f"No improvement after {self.number_of_tries} tries in iter {iter_}.")
  break
  res_bad.append(not_working)
- #plt.plot(res_std)
- #plt.show()
- #df_new.sum().plot.bar()
+
  self.df_new = df_new
  self.res_std = res_std
  self.res_bad = res_bad
  if (len(res_std) > 0) and self.plot:
- plot_at = self.plot_all_tries(self.res_std, self.res_bad)
+ plot_at = self.plot_all_tries()
  plt.title("All tries per iteration with \n corresponding standard deviation")
- plt.show()
  return df_new, plot_at
  return df_new
 
@@ -112,13 +109,13 @@ def reset(self):
  self.res_std = None
  self.res_bad = None
 
- @staticmethod
- def plot_all_tries(res_std, res_bad):
- y_max = max([x[1] for x in res_bad[0]]) * 1.1
- plt.plot(res_std)
- plt.scatter(range(len(res_std)), res_std)
+ def plot_all_tries(self):
+  """Plot for all iterations the returned std and the best value"""
+ y_max = max([x[1] for x in self.res_bad[0]]) * 1.1
+ plt.plot(self.res_std)
+ plt.scatter(range(len(self.res_std)), self.res_std)
  plt.ylim(0, y_max)
- for i, row_std in enumerate(res_bad):
+ for i, row_std in enumerate(self.res_bad):
  for idx, (j, s) in enumerate(row_std):
  #plt.text(i + idx*0.02, s, f"{j}", fontsize=8)
  plt.scatter(i + idx*0.01, s)
@@ -127,6 +124,9 @@ def plot_all_tries(res_std, res_bad):
  return plt
 
  def plot_results(self):
+ """Plot target distribution before and after upsampling.
+ Also plot the counts of each index-id.
+ """
  plt.subplot(2,2,1)
  self.plot_distr(self.df, "before")
  plt.subplot(2,2,2)
@@ -138,12 +138,15 @@ def plot_results(self):
  return plt
 
  def plot_distr(self, df, when):
+ """Plot target distribtion"""
  df[self.target_list].sum().plot.bar()
  plt.title(f"Label distribution \n{when} upsampling")
  return plt
 
  def plot_index_counts(self, df_new):
- """TODO make better xticks alignment"""
+ """Plot upsampling counts for each index. 
+
+ TODO make better xticks alignment"""
  idxs = list(df_new.index)
  lens = len(set(idxs))
  plt.hist(idxs, bins=lens, width=.1)#, edgecolor='k')