Update app.py

app.py

@@ -1,139 +1,36 @@
-import os
 import gradio as gr
-from transformers import pipeline
-import spacy
-import subprocess
-import nltk
-from nltk.corpus import wordnet
-from nltk.corpus import stopwords
-from nltk.tokenize import word_tokenize
-from spellchecker import SpellChecker
-import re
-import string
-import random
-
-# Download necessary NLTK data
-nltk.download('punkt')
-nltk.download('stopwords')
-nltk.download('averaged_perceptron_tagger')
-nltk.download('averaged_perceptron_tagger_eng')
-nltk.download('wordnet')
-nltk.download('omw-1.4')
-nltk.download('punkt_tab')
-
-# Initialize stopwords
-stop_words = set(stopwords.words("english"))
-
-# Words we don't want to replace
-exclude_tags = {'PRP', 'PRP$', 'MD', 'VBZ', 'VBP', 'VBD', 'VBG', 'VBN', 'TO', 'IN', 'DT', 'CC'}
-exclude_words = {'is', 'am', 'are', 'was', 'were', 'have', 'has', 'do', 'does', 'did', 'will', 'shall', 'should', 'would', 'could', 'can', 'may', 'might'}
-
-# Initialize the English text classification pipeline for AI detection
-pipeline_en = pipeline(task="text-classification", model="Hello-SimpleAI/chatgpt-detector-roberta")
-
-# Initialize the spell checker
-spell = SpellChecker()
-
-# Ensure the SpaCy model is installed
-try:
-    nlp = spacy.load("en_core_web_sm")
-except OSError:
-    subprocess.run(["python", "-m", "spacy", "download", "en_core_web_sm"])
-    nlp = spacy.load("en_core_web_sm")
-
-# Load SpaCy model
-nlp = spacy.load("en_core_web_sm")
-
-# Exclude tags and words (adjusted for better precision)
-exclude_tags = {'PRP', 'PRP$', 'MD', 'VBZ', 'VBP', 'VBD', 'VBG', 'VBN', 'TO', 'IN', 'DT', 'CC'}
-exclude_words = {'is', 'am', 'are', 'was', 'were', 'have', 'has', 'do', 'does', 'did', 'will', 'shall', 'should', 'would', 'could', 'can', 'may', 'might'}
-
-def get_synonyms(word):
-    """Find synonyms for a given word considering the context."""
-    synonyms = set()
-    for syn in wordnet.synsets(word):
-        for lemma in syn.lemmas():
-            if "_" not in lemma.name() and lemma.name().isalpha() and lemma.name().lower() != word.lower():
-                synonyms.add(lemma.name())
-    return synonyms
-
-def replace_with_synonyms(word, pos_tag):
-    """Replace words with synonyms, keeping the original POS tag."""
-    synonyms = get_synonyms(word)
-    # Filter by POS tag
-    filtered_synonyms = [syn for syn in synonyms if nltk.pos_tag([syn])[0][1] == pos_tag]
-    if filtered_synonyms:
-        return random.choice(filtered_synonyms)
-    return word
-
-def improve_paraphrasing_and_grammar(text):
-    """Paraphrase and correct grammatical errors in the text."""
-    doc = nlp(text)
-    corrected_text = []
-
-    for sent in doc.sents:
-        sentence = []
-        for token in sent:
-            # Replace words with synonyms, excluding special POS tags
-            if token.tag_ not in exclude_tags and token.text.lower() not in exclude_words and token.text not in string.punctuation:
-                synonym = replace_with_synonyms(token.text, token.tag_)
-                sentence.append(synonym if synonym else token.text)
-            else:
-                sentence.append(token.text)
-
-        corrected_text.append(' '.join(sentence))
-    
-    # Ensure proper punctuation and capitalization
-    final_text = ' '.join(corrected_text)
-    final_text = fix_possessives(final_text)
-    final_text = fix_punctuation_spacing(final_text)
-    final_text = capitalize_sentences(final_text)
-    final_text = fix_article_errors(final_text)
-    
-    return final_text
-
-def fix_punctuation_spacing(text):
-    """Fix spaces before punctuation marks."""
-    text = re.sub(r'\s+([,.!?])', r'\1', text)
-    return text
-
-def fix_possessives(text):
-    """Correct possessives like 'John ' s' -> 'John's'."""
-    return re.sub(r"(\w)\s?'\s?s", r"\1's", text)
-
-def capitalize_sentences(text):
-    """Capitalize the first letter of each sentence."""
-    return '. '.join([s.capitalize() for s in re.split(r'(?<=\w[.!?])\s+', text)])
-
-def fix_article_errors(text):
-    """Correct 'a' and 'an' usage based on following word's sound."""
-    doc = nlp(text)
-    corrected = []
-    for token in doc:
-        if token.text in ('a', 'an'):
-            next_token = token.nbor(1)
-            if token.text == "a" and next_token.text[0].lower() in "aeiou":
-                corrected.append("an")
-            elif token.text == "an" and next_token.text[0].lower() not in "aeiou":
-                corrected.append("a")
-            else:
-                corrected.append(token.text)
-        else:
-            corrected.append(token.text)
-    return ' '.join(corrected)
-
-# Gradio app setup
-def gradio_interface(text):
-    """Gradio interface function to process the input text."""
-    return improve_paraphrasing_and_grammar(text)
-
-with gr.Blocks() as demo:
-    gr.Markdown("## Text Paraphrasing and Grammar Correction")
-    text_input = gr.Textbox(lines=10, label='Enter text for paraphrasing and grammar correction')
-    text_output = gr.Textbox(lines=10, label='Corrected Text', interactive=False)
-    submit_button = gr.Button("🔄 Paraphrase and Correct")
-    
-    submit_button.click(fn=gradio_interface, inputs=text_input, outputs=text_output)
+from gector.gec_model import GecBERTModel
+
+# Load the GECToR model
+def load_model():
+    model = GecBERTModel(
+        vocab_path='data/output_vocabulary',
+        model_paths=['data/model_files/xlnet_0_gector.th'],
+        max_len=128, min_len=3
+    )
+    return model
+
+# Function to correct grammar using GECToR model
+def correct_grammar(text):
+    # Initialize the model (you can load it once and use globally to avoid reloading each time)
+    model = load_model()
+    # Correct the input text
+    corrected_text = model.handle_batch([text])
+    return corrected_text[0]  # Since the result is a list, return the first (and only) item
+
+# Define Gradio interface
+def create_gradio_interface():
+    # Input and output in Gradio
+    interface = gr.Interface(
+        fn=correct_grammar,  # Function to run
+        inputs="text",  # Input is plain text
+        outputs="text",  # Output is plain text
+        title="Grammar Correction App using GECToR",
+        description="Enter your text, and this app will correct its grammar using GECToR."
+    )
+    return interface
 
 # Launch the Gradio app
-demo.launch(share=True)
+if __name__ == "__main__":
+    gradio_interface = create_gradio_interface()
+    gradio_interface.launch()