functions_celia

Author: celiacnavarro

toolkit/plot.py

@@ -2,7 +2,11 @@
 import pandas as pd
 import numpy as np
 from typing import Union
-
+import plotly.graph_objs as go
+from plotly import tools
+from collections import defaultdict
+import plotly.offline as py
+from wordcloud import STOPWORDS
 
 def plot_multiclass_prediction_image(df, row_index: int, X_test: Union[pd.DataFrame, np.ndarray], prediction_col: str = 'Top Prediction', label_col: str = 'Label'):
     '''
@@ -39,4 +43,69 @@ def plot_multiclass_prediction_image(df, row_index: int, X_test: Union[pd.DataFr
     df.loc[row_index, df.dtypes != object].plot.bar(ax = ax[1])
     plt.xticks(rotation = 45, ha = 'right')
     plt.title('Probabilities of Each Class')
-    plt.show()
+    plt.show()
+
+
+
+def plot_ngrams(df, target:str, text:str, n_gram:int):
+    ''' 
+    This function allows visualizing the frequency of occurrence of words and n-grams (sets of a number of consecutive words) based on the target variable.
+    
+    Parameters:
+    - df: Dataframe with a text variable.
+    - target: Column of the dataframe where the target variable is located.
+    - text: Column of the dataframe where the text is located.
+    - n_gram: Number of consecutive words whose frequency we want to visualize.
+    
+    Returns:
+    - Bar chart representing the frequency of words, sorted from highest to lowest, divided by target variable.
+    '''
+
+    df1 = df[df[target] ==1]
+    df0 = df[df[target] ==0]
+
+    # Custom function for ngram generation 
+    def generate_ngrams(text, n_gram):
+        token = [token for token in text.lower().split(" ") if token != "" if token not in STOPWORDS]
+        ngrams = zip(*[token[i:] for i in range(n_gram)])
+        return [" ".join(ngram) for ngram in ngrams]
+
+    # Custom function for horizontal bar chart 
+    def horizontal_bar_chart(df, color):
+        trace = go.Bar(
+            y=df["word"].values[::-1],
+            x=df["wordcount"].values[::-1],
+            showlegend=False,
+            orientation = 'h',
+            marker=dict(
+                color=color,
+            ),
+        )
+        return trace
+
+    # Get the bar chart from text with label 0 
+    freq_dict = defaultdict(int)
+    for sent in df0[text]:
+        for word in generate_ngrams(sent, n_gram):
+            freq_dict[word] += 1
+    fd_sorted = pd.DataFrame(sorted(freq_dict.items(), key=lambda x: x[1])[::-1])
+    fd_sorted.columns = ["word", "wordcount"]
+    trace0 = horizontal_bar_chart(fd_sorted.head(50), 'blue')
+
+    # Get the bar chart from text with label 1 
+    freq_dict = defaultdict(int)
+    for sent in df1[text]:
+        for word in generate_ngrams(sent, n_gram):
+            freq_dict[word] += 1
+    fd_sorted = pd.DataFrame(sorted(freq_dict.items(), key=lambda x: x[1])[::-1])
+    fd_sorted.columns = ["word", "wordcount"]
+    trace1 = horizontal_bar_chart(fd_sorted.head(50), 'blue')
+
+    # Creating two subplots
+    fig = tools.make_subplots(rows=1, cols=2, vertical_spacing=0.04,
+                          subplot_titles=["Frequent words from text with label 0", 
+                                          "Frequent words from text with label 1"])
+    fig.append_trace(trace0, 1, 1)
+    fig.append_trace(trace1, 1, 2)
+    fig['layout'].update(height=1200, width=900, paper_bgcolor='rgb(233,233,233)', title="Word Count Plots")
+    py.iplot(fig, filename='word-plots')