Improve plots and README

phiyodr · Jan 3, 2023 · 0d11c5b · 0d11c5b
1 parent 81a8589
commit 0d11c5b
Show file tree

Hide file tree

Showing 7 changed files with 83 additions and 74 deletions.
diff --git a/README.md b/README.md
@@ -1,4 +1,4 @@
-# Multilabel Oversampling
+# Multilabel Oversampling :sunflower:
 
 **Many algorithms for imbalanced data support binary and multiclass classification only.**
 **This approach is made for multi-label classification (aka multi-target classification).**
@@ -10,7 +10,7 @@
 * Multilabel dataset (as `pandas.DataFrame`) with imbalanced data
 * Calculate counts per class and then calculate the standard deviation (std) of the count values
 * Do for `number_of_adds` times the following: 
-* Randomly draw a sample from your data and calculate new std 
+ * Randomly draw a sample from your data and calculate new std 
  * If new std reduces, add sample to your dataset
  * If not, draw another sample (to this up to `number_of_tries` times)
 * A new df is returned. 
@@ -19,15 +19,24 @@
 ## :arrow_right: Usage
 
 ```python
-from multilabel_oversampling import multilabel_oversampling as mo
+import multilabel_oversampling as mo
 
-df = mo.create_fake_data(size=1, seed=3)
+mo.seed_everything(20)
+df = mo.create_fake_data(size=1) # difficult fake dataset with very high dependency of y1 and y2
 ml_oversampler = mo.MultilabelOversampler(number_of_adds=100, number_of_tries=100)
-df_new, plot_at = ml_oversampler.fit(df)
-#> Iteration: 20%|██████ | 20/100 [00:00<00:00, 111.68it/s]
-#> No improvement after 100 tries in iter 20.
+df_new = ml_oversampler.fit(df)
+#>Start the upsampling process.
+#>Iteration: 11%|████████████████ | 11/100 [00:00<00:01, 48.43it/s]
+#>Iter 11: No improvement after 100 tries.
+#>Sampling done.
+#>
+#>Dataset size original: 20; Upsampled dataset size: 31
+#>Original target distribution: {'y1': 16, 'y2': 12, 'y3': 4, 'y4': 4}
+#>Upsampled target distribution: {'y1': 19, 'y2': 12, 'y3': 15, 'y4': 15}
+
+ml_oversampler.plot_all_tries()
 ```
-![Plot from df_new = ml_oversampler.fit(df)](assets/plot.png)
+![Plot from ml_oversampler.plot_all_tries()](assets/plot_all_tries.png)
 
 ```python
 ml_oversampler.plot_results()
@@ -41,72 +50,63 @@ ml_oversampler.plot_results()
 
 # Original DataFrame
 print(df)
-#> y1 y2 y3 y4 x
-#> 0 1 1 0 0 img_0.jpg
-#> 1 1 1 1 0 img_1.jpg
-#> 2 1 1 1 0 img_2.jpg
-#> 3 1 1 0 0 img_3.jpg
-#> 4 1 1 0 0 img_4.jpg
-#> 5 1 1 0 1 img_5.jpg
-#> 6 1 1 0 0 img_6.jpg
-#> 7 1 1 0 0 img_7.jpg
-#> 8 1 1 0 0 img_8.jpg
-#> 9 1 1 0 0 img_9.jpg
-#> 10 1 1 0 0 img_10.jpg
-#> 11 1 1 0 0 img_11.jpg
-#> 12 1 0 0 1 img_12.jpg
-#> 13 1 0 0 1 img_13.jpg
-#> 14 1 0 0 0 img_14.jpg
-#> 15 1 0 0 0 img_15.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 18 0 0 1 0 img_18.jpg
-#> 19 0 0 0 0 img_19.jpg
+#> y1 y2 y3 y4 x
+#>0 1 1 0 0 img_0.jpg
+#>1 1 1 0 0 img_1.jpg
+#>2 1 1 0 1 img_2.jpg
+#>3 1 1 0 0 img_3.jpg
+#>4 1 1 1 0 img_4.jpg
+#>5 1 1 0 0 img_5.jpg
+#>6 1 1 0 0 img_6.jpg
+#>7 1 1 0 0 img_7.jpg
+#>8 1 1 0 1 img_8.jpg
+#>9 1 1 0 0 img_9.jpg
+#>10 1 1 0 0 img_10.jpg
+#>11 1 1 0 0 img_11.jpg
+#>12 1 0 1 0 img_12.jpg
+#>13 1 0 1 1 img_13.jpg
+#>14 1 0 0 0 img_14.jpg
+#>15 1 0 0 0 img_15.jpg
+#>16 0 0 0 0 img_16.jpg
+#>17 0 0 0 0 img_17.jpg
+#>18 0 0 0 0 img_18.jpg
+#>19 0 0 1 1 img_19.jpg
+
 
 # New DataFrame after upsampling
 print(df_new)
-#> y1 y2 y3 y4 x
-#> 0 1 1 0 0 img_0.jpg
-#> 1 1 1 1 0 img_1.jpg
-#> 2 1 1 1 0 img_2.jpg
-#> 3 1 1 0 0 img_3.jpg
-#> 4 1 1 0 0 img_4.jpg
-#> 5 1 1 0 1 img_5.jpg
-#> 6 1 1 0 0 img_6.jpg
-#> 7 1 1 0 0 img_7.jpg
-#> 8 1 1 0 0 img_8.jpg
-#> 9 1 1 0 0 img_9.jpg
-#> 10 1 1 0 0 img_10.jpg
-#> 11 1 1 0 0 img_11.jpg
-#> 12 1 0 0 1 img_12.jpg
-#> 13 1 0 0 1 img_13.jpg
-#> 14 1 0 0 0 img_14.jpg
-#> 15 1 0 0 0 img_15.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 18 0 0 1 0 img_18.jpg
-#> 19 0 0 0 0 img_19.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 18 0 0 1 0 img_18.jpg
-#> 13 1 0 0 1 img_13.jpg
-#> 18 0 0 1 0 img_18.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 16 0 0 1 0 img_16.jpg
-#> 17 0 0 0 1 img_17.jpg
-#> 17 0 0 0 1 img_17.jpg
-
+#> y1 y2 y3 y4 x
+#>0 1 1 0 0 img_0.jpg
+#>1 1 1 0 0 img_1.jpg
+#>2 1 1 0 1 img_2.jpg
+#>3 1 1 0 0 img_3.jpg
+#>4 1 1 1 0 img_4.jpg
+#>5 1 1 0 0 img_5.jpg
+#>6 1 1 0 0 img_6.jpg
+#>7 1 1 0 0 img_7.jpg
+#>8 1 1 0 1 img_8.jpg
+#>9 1 1 0 0 img_9.jpg
+#>10 1 1 0 0 img_10.jpg
+#>11 1 1 0 0 img_11.jpg
+#>12 1 0 1 0 img_12.jpg
+#>13 1 0 1 1 img_13.jpg
+#>14 1 0 0 0 img_14.jpg
+#>15 1 0 0 0 img_15.jpg
+#>16 0 0 0 0 img_16.jpg
+#>17 0 0 0 0 img_17.jpg
+#>18 0 0 0 0 img_18.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>13 1 0 1 1 img_13.jpg
+#>13 1 0 1 1 img_13.jpg
+#>13 1 0 1 1 img_13.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
+#>19 0 0 1 1 img_19.jpg
 ```
 
 
@@ -118,6 +118,13 @@ print(df_new)
 pip install git+https://github.com/phiyodr/multilabel-oversampling
 ```
 
+* Install from PyPI
+
+```bash
+pip install multilabel-oversampling
+```
+
+
 
 ## :construction_worker: Future work
 

diff --git a/assets/plot.png b/assets/plot.png
diff --git a/assets/plot_all_tries.png b/assets/plot_all_tries.png
diff --git a/assets/plot_results.png b/assets/plot_results.png
diff --git a/multilabel_oversampling/__init__.py b/multilabel_oversampling/__init__.py
@@ -1,3 +1,3 @@
-__version__ = "0.1.2"
+__version__ = "0.1.3"
 
 from .multilabel_oversampling import *
diff --git a/multilabel_oversampling/multilabel_oversampling.py b/multilabel_oversampling/multilabel_oversampling.py
@@ -131,6 +131,7 @@ def plot_all_tries(self):
  plt.scatter(i + idx*0.01, s)
  plt.xlabel('Iters')#, fontsize=18)
  plt.ylabel('Std')#, fontsize=16)
+ plt.title("All standard deviations per iteration")
  return plt
 
  def plot_results(self):
@@ -172,8 +173,8 @@ def plot_index_counts(self, df_new=None):
  df_new = self.df_new
  x = list(collections.Counter(list(df_new.index)).values())
  plt.hist(x, bins=max(x)+1, rwidth=.9)
- plt.title("Frequency of indexes in df")
- plt.xlabel('Frequency in dataset')
+ plt.title("Frequency of\nindexes in df")
+ plt.xlabel('Occurances of indexes in dataset')
  plt.ylabel('Counts')
  return plt
 

diff --git a/setup.py b/setup.py
@@ -18,6 +18,7 @@
  long_description_content_type="text/markdown",
  url="https://github.com/phiyodr/multilabel-oversampling",
  classifiers=[
+ "Topic :: Scientific/Engineering :: Artificial Intelligence",
  "Programming Language :: Python :: 3",
  ],
  packages=setuptools.find_packages(),