more colors

app.py

@@ -1,4 +1,5 @@
 from io import StringIO
+import itertools
 
 import gradio as gr
 import pandas as pd
@@ -9,6 +10,7 @@ nlp = spacy.load('en_core_web_sm')
 
 HTML_RED = '<span style="background-color: rgba(255, 0, 0, 0.2)">{t}</span>'
 HTML_GRN = '<span style="background-color: rgba(0, 255, 0, 0.3)">{t}</span>'
+HTML_YLW = '<span style="background-color: rgba(255, 255, 0, 0.3)">{t}</span>'
 HTML_BLU = '<span style="background-color: rgba(0, 0, 255, 0.2)">{t}</span>'
 HTML_PLN = '<span>{t}</span>'
 TABLE_CSS = '''
@@ -22,6 +24,7 @@ table, th, td {
 }
 '''
 
+
 def colorize(file_obj):
     with open(file_obj.name, 'r') as f:
         raw = f.read()
@@ -37,9 +40,26 @@ def colorize(file_obj):
             row[1]['model summary A'],
             row[1]['model summary B']
         ))
-        tokens_gold = {token.lemma_.lower() for token in gold}
+        tokens_gold = {token.lemma_.lower(): 0 for token in gold}
+        for token in itertools.chain(genA, genB):
+            if token.lemma_.lower() in tokens_gold:
+                tokens_gold[token.lemma_.lower()] += 1
+
+        gold_text = ''.join([
+            (
+                HTML_PLN.format(t=token.text)
+                if token.pos_ not in {'NOUN', 'PROPN', 'VERB'}
+                else (
+                    (
+                        HTML_BLU if tokens_gold[token.lemma_.lower()] > 0
+                        else HTML_YLW
+                    ).format(t=token.text)
+                )
+            ) + token.whitespace_
+            for token in gold
+        ])
         table_content.append(
-            [id_, gold.text] +
+            [id_, gold_text] +
             [
                 ''.join(
                     (
@@ -88,6 +108,30 @@ def main():
                 )
                 run = gr.Button(label='Run')
             with gr.TabItem("Visualization"):
+                gr.HTML(
+                    ''.join(
+                        (
+                            "<b>Explanation of colors:</b>",
+                            "<br><ul>",
+                            "<li><b>",
+                            HTML_RED.format(t='Red'),
+                            "</b>: word is in generated, but not in gold.</li>",
+                            "<li><b>",
+                            HTML_GRN.format(t='Green'),
+                            "</b>: word is in generated summary and gold.</li>",
+                            "<li><b>",
+                            HTML_YLW.format(t='Yellow'),
+                            "</b>: word is in gold, but not in generated.</li>",
+                            "<li><b>",
+                            HTML_BLU.format(t='Blue'),
+                            "</b>: word is in gold and in generated.</li>",
+                            "</ul>",
+                            "<br>",
+                            "<b>Important</b>: Only nouns, verbs and proper ",
+                            "nouns are colored.</b>"
+                        )
+                    )
+                )
                 viz = gr.HTML(label='Upload a csv file to start.')
         run.click(colorize, data, viz)