Spaces:
Paused
Paused
Commit ·
17cb555
1
Parent(s): d6a3f78
Update app.py
Browse files
app.py
CHANGED
|
@@ -6,6 +6,24 @@ import random
|
|
| 6 |
import nltk
|
| 7 |
nltk.download('punkt')
|
| 8 |
nltk.download('wordnet')
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
def paraphrase_text(text, synonyms_num=5, random_synonym=True):
|
| 10 |
# Tokenize the text
|
| 11 |
tokens = word_tokenize(text)
|
|
@@ -38,6 +56,7 @@ def paraphrase_text(text, synonyms_num=5, random_synonym=True):
|
|
| 38 |
paraphrased_text = ' '.join(paraphrased_tokens)
|
| 39 |
return paraphrased_text
|
| 40 |
|
|
|
|
| 41 |
def paraphrase_html(html_text, synonyms_num, random_synonym):
|
| 42 |
# Parse the HTML using BeautifulSoup
|
| 43 |
soup = BeautifulSoup(html_text, 'html.parser')
|
|
@@ -50,14 +69,101 @@ def paraphrase_html(html_text, synonyms_num, random_synonym):
|
|
| 50 |
paraphrased_html = str(soup)
|
| 51 |
return paraphrased_html
|
| 52 |
|
| 53 |
-
st.set_page_config(page_title="HTML Paraphraser", page_icon=":pencil2:")
|
| 54 |
-
st.title("HTML Paraphraser")
|
| 55 |
|
| 56 |
-
|
| 57 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
|
| 59 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
st.write(paraphrased_html, unsafe_allow_html=True)
|
|
|
|
| 6 |
import nltk
|
| 7 |
nltk.download('punkt')
|
| 8 |
nltk.download('wordnet')
|
| 9 |
+
|
| 10 |
+
import streamlit as st
|
| 11 |
+
from bs4 import BeautifulSoup
|
| 12 |
+
from nltk.tokenize import word_tokenize
|
| 13 |
+
from nltk.corpus import wordnet
|
| 14 |
+
from transformers import pipeline
|
| 15 |
+
|
| 16 |
+
# Text analytics
|
| 17 |
+
from textblob import TextBlob
|
| 18 |
+
from textatistic import Textatistic
|
| 19 |
+
|
| 20 |
+
# Integration with other tools
|
| 21 |
+
from grammarbot import GrammarBotClient
|
| 22 |
+
|
| 23 |
+
# Error handling and logging
|
| 24 |
+
import logging
|
| 25 |
+
|
| 26 |
+
|
| 27 |
def paraphrase_text(text, synonyms_num=5, random_synonym=True):
|
| 28 |
# Tokenize the text
|
| 29 |
tokens = word_tokenize(text)
|
|
|
|
| 56 |
paraphrased_text = ' '.join(paraphrased_tokens)
|
| 57 |
return paraphrased_text
|
| 58 |
|
| 59 |
+
|
| 60 |
def paraphrase_html(html_text, synonyms_num, random_synonym):
|
| 61 |
# Parse the HTML using BeautifulSoup
|
| 62 |
soup = BeautifulSoup(html_text, 'html.parser')
|
|
|
|
| 69 |
paraphrased_html = str(soup)
|
| 70 |
return paraphrased_html
|
| 71 |
|
|
|
|
|
|
|
| 72 |
|
| 73 |
+
def analyze_text(text):
|
| 74 |
+
# Text analytics using textblob and textatistic
|
| 75 |
+
blob = TextBlob(text)
|
| 76 |
+
sentiment = blob.sentiment.polarity
|
| 77 |
+
subjectivity = blob.sentiment.subjectivity
|
| 78 |
+
text_stats = Textatistic(text)
|
| 79 |
+
readability_score = text_stats.flesch_score
|
| 80 |
+
|
| 81 |
+
return sentiment, subjectivity, readability_score
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def grammar_check(text):
|
| 85 |
+
# Integration with grammarbot tool
|
| 86 |
+
client = GrammarBotClient()
|
| 87 |
+
response = client.check(text)
|
| 88 |
+
|
| 89 |
+
matches = response['matches']
|
| 90 |
+
if not matches:
|
| 91 |
+
return "No grammar errors found."
|
| 92 |
+
else:
|
| 93 |
+
message = "Grammar errors found:\n"
|
| 94 |
+
for match in matches:
|
| 95 |
+
message += f"{match['message']} at line {match['replacements'][0]['startLine']}\n"
|
| 96 |
+
return message
|
| 97 |
+
|
| 98 |
+
|
| 99 |
+
def paraphrase_with_model(text, model):
|
| 100 |
+
# Advanced paraphrasing using pre-trained models
|
| 101 |
+
paraphrase_pipeline = pipeline("text2text-generation", model=model, tokenizer=model, device=0 if st.config.experimental.get_query_params()['device'] == 'cpu' else -1)
|
| 102 |
+
paraphrased_text = paraphrase_pipeline(text, max_length=50, do_sample=True, temperature=0.9)[0]['generated_text']
|
| 103 |
+
return paraphrased_text
|
| 104 |
+
|
| 105 |
+
|
| 106 |
+
st.set_page_config(page_title="HTML Paraphraser and Analyzer", page_icon=":pencil2:")
|
| 107 |
+
|
| 108 |
+
st.sidebar.title("HTML Paraphraser and Analyzer")
|
| 109 |
+
|
| 110 |
+
st.title("HTML Paraphraser and Analyzer")
|
| 111 |
+
html_input = st.text_area("Enter HTML to paraphrase", height=250)
|
| 112 |
+
|
| 113 |
+
synonyms_num = st.slider("Number of synonyms per word", min_value=1, max_value=10, value=5, step=1)
|
| 114 |
+
|
| 115 |
+
random_synonym = st.checkbox("Use random synonym", value=True)
|
| 116 |
+
|
| 117 |
+
#Add a button to paraphrase the HTML
|
| 118 |
+
if st.button("Paraphrase HTML"):
|
| 119 |
+
try:
|
| 120 |
+
paraphrased_html = paraphrase_html(html_input, synonyms_num, random_synonym)
|
| 121 |
+
st.write(paraphrased_html, unsafe_allow_html=True)
|
| 122 |
+
except Exception as e:
|
| 123 |
+
logging.exception("Error occurred while paraphrasing HTML")
|
| 124 |
+
st.error("An error occurred while paraphrasing the HTML. Please try again.")
|
| 125 |
+
|
| 126 |
+
#Add a text area for the user to input the text to analyze
|
| 127 |
+
text_input = st.text_area("Enter text to analyze")
|
| 128 |
+
|
| 129 |
+
#Add a button to analyze the text
|
| 130 |
+
if st.button("Analyze Text"):
|
| 131 |
+
try:
|
| 132 |
+
sentiment, subjectivity, readability_score = analyze_text(text_input)
|
| 133 |
+
st.write(f"Sentiment: {sentiment:.2f}")
|
| 134 |
+
st.write(f"Subjectivity: {subjectivity:.2f}")
|
| 135 |
+
st.write(f"Readability Score: {readability_score:.2f}")
|
| 136 |
+
except Exception as e:
|
| 137 |
+
logging.exception("Error occurred while analyzing text")
|
| 138 |
+
st.error("An error occurred while analyzing the text. Please try again.")
|
| 139 |
+
|
| 140 |
+
#Add a button to check grammar using GrammarBot
|
| 141 |
+
if st.button("Check Grammar"):
|
| 142 |
+
try:
|
| 143 |
+
grammar_check_result = grammar_check(text_input)
|
| 144 |
+
st.write(grammar_check_result)
|
| 145 |
+
except Exception as e:
|
| 146 |
+
logging.exception("Error occurred while checking grammar")
|
| 147 |
+
st.error("An error occurred while checking grammar. Please try again.")
|
| 148 |
+
|
| 149 |
+
#Add a selectbox for choosing the pre-trained model for advanced paraphrasing
|
| 150 |
+
models = {
|
| 151 |
+
"GPT-2": "gpt2",
|
| 152 |
+
"GPT-Neo": "EleutherAI/gpt-neo-1.3B",
|
| 153 |
+
"T5": "t5-base",
|
| 154 |
+
"Pegasus": "google/pegasus-large",
|
| 155 |
+
}
|
| 156 |
+
model_name = st.selectbox("Choose a pre-trained model for advanced paraphrasing", options=list(models.keys()))
|
| 157 |
|
| 158 |
+
#Add a button to paraphrase the text using the selected model
|
| 159 |
+
if st.button("Advanced Paraphrasing"):
|
| 160 |
+
try:
|
| 161 |
+
model = models[model_name]
|
| 162 |
+
paraphrased_text = paraphrase_with_model(text_input, model)
|
| 163 |
+
st.write(paraphrased_text)
|
| 164 |
+
except Exception as e:
|
| 165 |
+
logging.exception("Error occurred while performing advanced paraphrasing")
|
| 166 |
+
st.error("An error occurred while performing advanced paraphrasing. Please try again.")
|
| 167 |
|
| 168 |
+
#Set up logging to capture any errors
|
| 169 |
+
logging.basicConfig(filename='error.log', level=logging.ERROR, format='%(asctime)s %(message)s')
|
|
|