Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
|
@@ -74,7 +74,7 @@ def correct_singular_plural_errors(text):
|
|
| 74 |
if token.pos_ == "NOUN":
|
| 75 |
# Check if the noun is singular or plural
|
| 76 |
if token.tag_ == "NN": # Singular noun
|
| 77 |
-
# Look for determiners like "many" to correct to plural
|
| 78 |
if any(child.text.lower() in ['many', 'several', 'few'] for child in token.head.children):
|
| 79 |
corrected_text.append(token.lemma_ + 's')
|
| 80 |
else:
|
|
@@ -85,8 +85,6 @@ def correct_singular_plural_errors(text):
|
|
| 85 |
corrected_text.append(token.lemma_)
|
| 86 |
else:
|
| 87 |
corrected_text.append(token.text)
|
| 88 |
-
else:
|
| 89 |
-
corrected_text.append(token.text)
|
| 90 |
else:
|
| 91 |
corrected_text.append(token.text)
|
| 92 |
|
|
@@ -109,52 +107,16 @@ def correct_article_errors(text):
|
|
| 109 |
corrected_text.append(token.text)
|
| 110 |
return ' '.join(corrected_text)
|
| 111 |
|
| 112 |
-
#
|
| 113 |
-
def paraphrase_with_spacy_nltk(text):
|
| 114 |
-
doc = nlp(text)
|
| 115 |
-
paraphrased_words = []
|
| 116 |
-
|
| 117 |
-
for token in doc:
|
| 118 |
-
# Map SpaCy POS tags to WordNet POS tags
|
| 119 |
-
pos = None
|
| 120 |
-
if token.pos_ in {"NOUN"}:
|
| 121 |
-
pos = wordnet.NOUN
|
| 122 |
-
elif token.pos_ in {"VERB"}:
|
| 123 |
-
pos = wordnet.VERB
|
| 124 |
-
elif token.pos_ in {"ADJ"}:
|
| 125 |
-
pos = wordnet.ADJ
|
| 126 |
-
elif token.pos_ in {"ADV"}:
|
| 127 |
-
pos = wordnet.ADV
|
| 128 |
-
|
| 129 |
-
synonyms = get_synonyms_nltk(token.text.lower(), pos) if pos else []
|
| 130 |
-
|
| 131 |
-
# Replace with a synonym only if it makes sense
|
| 132 |
-
if synonyms and token.pos_ in {"NOUN", "VERB", "ADJ", "ADV"} and synonyms[0] != token.text.lower():
|
| 133 |
-
paraphrased_words.append(synonyms[0])
|
| 134 |
-
else:
|
| 135 |
-
paraphrased_words.append(token.text)
|
| 136 |
-
|
| 137 |
-
# Join the words back into a sentence
|
| 138 |
-
paraphrased_sentence = ' '.join(paraphrased_words)
|
| 139 |
-
|
| 140 |
-
return paraphrased_sentence
|
| 141 |
-
|
| 142 |
-
# Combined function: Paraphrase -> Grammar Correction -> Capitalization (Humanifier)
|
| 143 |
def paraphrase_and_correct(text):
|
| 144 |
-
#
|
| 145 |
-
paraphrased_text = paraphrase_with_spacy_nltk(text)
|
| 146 |
-
|
| 147 |
-
# Step 2: Apply grammatical corrections on the paraphrased text
|
| 148 |
-
corrected_text = correct_article_errors(paraphrased_text)
|
| 149 |
-
|
| 150 |
-
corrected_text = capitalize_sentences_and_nouns(corrected_text)
|
| 151 |
|
| 152 |
-
|
| 153 |
-
|
| 154 |
-
|
| 155 |
-
|
| 156 |
-
|
| 157 |
-
return
|
| 158 |
|
| 159 |
# Gradio app setup with two tabs
|
| 160 |
with gr.Blocks() as demo:
|
|
@@ -176,4 +138,4 @@ with gr.Blocks() as demo:
|
|
| 176 |
paraphrase_button.click(paraphrase_and_correct, inputs=text_input, outputs=output_text)
|
| 177 |
|
| 178 |
# Launch the app with the remaining functionalities
|
| 179 |
-
demo.launch()
|
|
|
|
| 74 |
if token.pos_ == "NOUN":
|
| 75 |
# Check if the noun is singular or plural
|
| 76 |
if token.tag_ == "NN": # Singular noun
|
| 77 |
+
# Look for determiners like "many", "several", "few" to correct to plural
|
| 78 |
if any(child.text.lower() in ['many', 'several', 'few'] for child in token.head.children):
|
| 79 |
corrected_text.append(token.lemma_ + 's')
|
| 80 |
else:
|
|
|
|
| 85 |
corrected_text.append(token.lemma_)
|
| 86 |
else:
|
| 87 |
corrected_text.append(token.text)
|
|
|
|
|
|
|
| 88 |
else:
|
| 89 |
corrected_text.append(token.text)
|
| 90 |
|
|
|
|
| 107 |
corrected_text.append(token.text)
|
| 108 |
return ' '.join(corrected_text)
|
| 109 |
|
| 110 |
+
# Function to paraphrase and correct grammar
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
def paraphrase_and_correct(text):
|
| 112 |
+
paraphrased_text = capitalize_sentences_and_nouns(text) # Capitalize first to ensure proper noun capitalization
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
|
| 114 |
+
# Apply grammatical corrections
|
| 115 |
+
paraphrased_text = correct_article_errors(paraphrased_text)
|
| 116 |
+
paraphrased_text = correct_singular_plural_errors(paraphrased_text)
|
| 117 |
+
paraphrased_text = correct_tense_errors(paraphrased_text)
|
| 118 |
+
|
| 119 |
+
return paraphrased_text
|
| 120 |
|
| 121 |
# Gradio app setup with two tabs
|
| 122 |
with gr.Blocks() as demo:
|
|
|
|
| 138 |
paraphrase_button.click(paraphrase_and_correct, inputs=text_input, outputs=output_text)
|
| 139 |
|
| 140 |
# Launch the app with the remaining functionalities
|
| 141 |
+
demo.launch()
|