Spaces:
Runtime error
Runtime error
Upload app.py
Browse files
app.py
ADDED
@@ -0,0 +1,236 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import PyPDF2
|
3 |
+
import os
|
4 |
+
import subprocess
|
5 |
+
import tempfile
|
6 |
+
import google.generativeai as genai
|
7 |
+
from nltk.corpus import stopwords
|
8 |
+
from nltk.tokenize import word_tokenize, sent_tokenize
|
9 |
+
def summarize_text(text):
|
10 |
+
# Tokenizing the text
|
11 |
+
stopWords = set(stopwords.words("english"))
|
12 |
+
words = word_tokenize(text)
|
13 |
+
|
14 |
+
# Creating a frequency table to keep the score of each word
|
15 |
+
freqTable = dict()
|
16 |
+
for word in words:
|
17 |
+
word = word.lower()
|
18 |
+
if word in stopWords:
|
19 |
+
continue
|
20 |
+
if word in freqTable:
|
21 |
+
freqTable[word] += 1
|
22 |
+
else:
|
23 |
+
freqTable[word] = 1
|
24 |
+
|
25 |
+
# Creating a dictionary to keep the score of each sentence
|
26 |
+
sentences = sent_tokenize(text)
|
27 |
+
sentenceValue = dict()
|
28 |
+
|
29 |
+
for sentence in sentences:
|
30 |
+
for word, freq in freqTable.items():
|
31 |
+
if word in sentence.lower():
|
32 |
+
if sentence in sentenceValue:
|
33 |
+
sentenceValue[sentence] += freq
|
34 |
+
else:
|
35 |
+
sentenceValue[sentence] = freq
|
36 |
+
|
37 |
+
sumValues = 0
|
38 |
+
for sentence in sentenceValue:
|
39 |
+
sumValues += sentenceValue[sentence]
|
40 |
+
|
41 |
+
# Average value of a sentence from the original text
|
42 |
+
average = int(sumValues / len(sentenceValue))
|
43 |
+
|
44 |
+
# Storing sentences into our summary.
|
45 |
+
summary = ''
|
46 |
+
for sentence in sentences:
|
47 |
+
if (sentence in sentenceValue) and (sentenceValue[sentence] > (1.2 * average)):
|
48 |
+
summary += " " + sentence
|
49 |
+
|
50 |
+
return summary
|
51 |
+
|
52 |
+
|
53 |
+
|
54 |
+
def Notes_data(API,prompt):
|
55 |
+
genai.configure(api_key=API)
|
56 |
+
model = genai.GenerativeModel('gemini-pro')
|
57 |
+
response = model.generate_content(prompt)
|
58 |
+
return response.text
|
59 |
+
def markdown_to_pdf(markdown_content):
|
60 |
+
# Create a temporary file to store the markdown content
|
61 |
+
with tempfile.NamedTemporaryFile(mode="w", delete=False, suffix=".md", encoding="utf-8") as temp_md_file:
|
62 |
+
temp_md_file.write(markdown_content)
|
63 |
+
temp_md_filename = temp_md_file.name
|
64 |
+
|
65 |
+
# Define output PDF filename
|
66 |
+
output_pdf_filename = tempfile.NamedTemporaryFile(suffix=".pdf", delete=False).name
|
67 |
+
|
68 |
+
# Execute mdpdf command
|
69 |
+
command = f"mdpdf -o {str(output_pdf_filename)} {str(temp_md_filename)}"
|
70 |
+
subprocess.run(command, shell=True)
|
71 |
+
|
72 |
+
# Delete temporary markdown file
|
73 |
+
os.remove(temp_md_filename)
|
74 |
+
|
75 |
+
return output_pdf_filename
|
76 |
+
# Authenticate using the API key
|
77 |
+
# Set the API key as an environment variable
|
78 |
+
def extract_text_from_pdf(pdf_file):
|
79 |
+
text = ""
|
80 |
+
with open(pdf_file, "rb") as file:
|
81 |
+
pdf_reader = PyPDF2.PdfReader(file)
|
82 |
+
for page_num in range(len(pdf_reader.pages)):
|
83 |
+
page = pdf_reader.pages[page_num]
|
84 |
+
text += page.extract_text()
|
85 |
+
return text
|
86 |
+
def get_prompt(text,subject,topic):
|
87 |
+
prompt = f"""
|
88 |
+
you are given a {subject} topic
|
89 |
+
{text}
|
90 |
+
Create the well Structured detailed Notes as a html code in the following templat. Consider the following points while creating the template.
|
91 |
+
1. Introduction part must have minimum 500 words
|
92 |
+
2. In the notes part please describe each topic in detail with word count 1000
|
93 |
+
3. Key points part must have minimum 250 words.
|
94 |
+
4. The notes should be on topic.
|
95 |
+
<!DOCTYPE html>
|
96 |
+
<html>
|
97 |
+
<head>
|
98 |
+
<meta charset="UTF-8">
|
99 |
+
<title>Chapter Study Material </title>
|
100 |
+
</head>
|
101 |
+
<body>
|
102 |
+
<h1>Study Material</h1>
|
103 |
+
<h3>1. Introduction</h3>
|
104 |
+
<p>
|
105 |
+
<!-- Replace with your introduction content-->
|
106 |
+
</p>
|
107 |
+
|
108 |
+
<h3>2.Important Topics in detail from the chapter. </h3>
|
109 |
+
<!--DEcribe the Imortant topics don't forget the quations formulas and other things replated to the topic.Describe and elaborate each of the topic and formula and equation-->
|
110 |
+
<h4><!-- Heading of topic 1>r</h4>
|
111 |
+
<p>
|
112 |
+
<!--Desribe the topice in 100 words and bullet out the subtopic and describe if any-->
|
113 |
+
<!-- Euations and formulas if any-->
|
114 |
+
</P>
|
115 |
+
<h4><!-- Heading of topic 2>r</h4>
|
116 |
+
<p>
|
117 |
+
<!--Desribe the topice in 100 words bullet out the subtopic and describe if any-->
|
118 |
+
<!-- Euations and formulas if any-->
|
119 |
+
</P>
|
120 |
+
<h4><!-- Heading of topic 3>r</h4>
|
121 |
+
<p>
|
122 |
+
<!--Desribe the topice in 100 words bullet out the subtopic and describe if any-->
|
123 |
+
<!-- Euations and formulas if any-->
|
124 |
+
</P>
|
125 |
+
<!-- Like wise add all possible topics from the chapter-->
|
126 |
+
<h3> Key Points from the chapter </h3>
|
127 |
+
<p> <!--Replace with your key points content in bulletpoints-->
|
128 |
+
</body>
|
129 |
+
</html>
|
130 |
+
|
131 |
+
"""
|
132 |
+
prompt2 = f"""
|
133 |
+
Create detailed study material from the given given Chapter from a book by considering the following points in mind
|
134 |
+
1.Ensure a thorough understanding of the chapter's main ideas and themes.
|
135 |
+
2.Identify and emphasise essential concepts and arguments.
|
136 |
+
3.Create a well-organised outline reflecting the chapter's structure.
|
137 |
+
4.Provide concise, to-the-point summaries for each section.
|
138 |
+
5.Define and clarify key terms and concepts introduced in the chapter.
|
139 |
+
6.Pose engaging questions for self-assessment and discussion.
|
140 |
+
7.Ensure accurate citations for quotes and references.
|
141 |
+
8.use given text only for reference you can add as much as detail you can.
|
142 |
+
9.The response should be complete.
|
143 |
+
10.Generate response as a html page with all type of formatting.
|
144 |
+
11.Each point you take in the study material please describe it in detail.
|
145 |
+
Use the following HTML template to structure your notes:
|
146 |
+
<!DOCTYPE html>
|
147 |
+
<html>
|
148 |
+
<head>
|
149 |
+
<meta charset="UTF-8">
|
150 |
+
<title>Chapter Study Material</title>
|
151 |
+
</head>
|
152 |
+
<body>
|
153 |
+
<h1>Chapter Study Material</h1>
|
154 |
+
<h3>1. Introduction</h3>
|
155 |
+
<p>
|
156 |
+
<!-- Replace with your introduction content -->
|
157 |
+
</p>
|
158 |
+
|
159 |
+
<h3>2. Describe important Points in detail from the chapter</h3>
|
160 |
+
<h4><!-- Heading of topic 1>r</h4>
|
161 |
+
<p>
|
162 |
+
<!--Desribe the topice in detail and elaborate it and bullet out the subtopic and describe if any-->
|
163 |
+
</P>
|
164 |
+
<h4><!-- Heading of topic 2>r</h4>
|
165 |
+
<p>
|
166 |
+
<!--Desribe the topice in detail and elaborate bullet out the subtopic and describe if any-->
|
167 |
+
|
168 |
+
</P>
|
169 |
+
<h4><!-- Heading of topic 3>r</h4>
|
170 |
+
<p>
|
171 |
+
<!--Desribe the topice 3 in detail and elaborate-->
|
172 |
+
</P>
|
173 |
+
<!-- Like wise add all possible topics from the chapter-->
|
174 |
+
<h3>3. Summary</h3>
|
175 |
+
<p>
|
176 |
+
<!-- Replace with your summary content -->
|
177 |
+
</p>
|
178 |
+
|
179 |
+
<!-- Additional content can be added as needed -->
|
180 |
+
</body>
|
181 |
+
</html>
|
182 |
+
Chapter Context:
|
183 |
+
{text}"""
|
184 |
+
|
185 |
+
prompt3 = f"""
|
186 |
+
Create detailed study material from the given given topic from a book chapter by considering the following points in mind
|
187 |
+
1.Notes will be based on topic specific.
|
188 |
+
2.Ensure that explanations are clear and concise, using simple language to make complex concepts understandable.
|
189 |
+
3.Highlight key concepts, theories, and important formulas related to each topic in detail.
|
190 |
+
4.For mathematical components like physics and chemistry, provide step-by-step solutions for numerical problems.
|
191 |
+
5.The notes should be in detail cover all the concepts related to the topic.
|
192 |
+
6.use given text only for reference you can add as much as detail you can.
|
193 |
+
7.The response should be complete.
|
194 |
+
8.Ignore unneccessory informstion in the context.
|
195 |
+
9.Generate response as a valid 'mardown formate' page with all type of formatting.
|
196 |
+
|
197 |
+
Topics: {topic}
|
198 |
+
|
199 |
+
Chapter Reference:-
|
200 |
+
{text}"""
|
201 |
+
return prompt2 if subject=="English" else prompt3
|
202 |
+
|
203 |
+
def pdf_generator(API,class_name, subject,chapter,query=None):
|
204 |
+
# Create a list to store the generated PDF file paths
|
205 |
+
pdf_file_path = f"Main_content/Books/{class_name}/{subject}/{chapter}.pdf"
|
206 |
+
pdf_text = summarize_text(extract_text_from_pdf(pdf_file_path))
|
207 |
+
study_material = generate_study_material(API,pdf_text,subject,query)
|
208 |
+
study_material = markdown_to_pdf(study_material)
|
209 |
+
return study_material
|
210 |
+
def generate_study_material(API,pdf_text,subject,topic):
|
211 |
+
response = Notes_data(API,get_prompt(pdf_text,subject,topic))
|
212 |
+
return response
|
213 |
+
|
214 |
+
# Define Gradio input and output components
|
215 |
+
input_components = [
|
216 |
+
gr.Textbox(label="Enter your Gemini API"),
|
217 |
+
gr.Dropdown(["Class 10", "Class 11", "Class 12"], label="Select Class"),
|
218 |
+
gr.Dropdown(["Math", "Science", "English","Chemistry"], label="Select Subject"),
|
219 |
+
gr.Dropdown(["Chapter 1", "Chapter 2", "Chapter 3"], label="Select Chapters"),
|
220 |
+
gr.Textbox(label="Write the topics name saperated by ','")
|
221 |
+
]
|
222 |
+
output_component = gr.File(label="Notes")
|
223 |
+
|
224 |
+
# Create the Gradio interface
|
225 |
+
iface = gr.Interface(
|
226 |
+
fn=pdf_generator,
|
227 |
+
inputs=input_components,
|
228 |
+
outputs=output_component,
|
229 |
+
title="PDF Generator",
|
230 |
+
description="Generate PDFs based on class, subject, and chapter selection.",
|
231 |
+
)
|
232 |
+
|
233 |
+
# Launch the Gradio app
|
234 |
+
iface.launch()
|
235 |
+
|
236 |
+
|