Duplicate from SRankChatGpt/Presentation-Assistant

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+pandoc-2.14.2/bin/pandoc filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: PA(Presentation Assistant)
+emoji: 🧑‍🏫
+colorFrom: indigo
+colorTo: green
+sdk: streamlit
+sdk_version: 1.17.0
+app_file: app.py
+pinned: false
+license: apache-2.0
+duplicated_from: SRankChatGpt/Presentation-Assistant
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,127 @@
+import os
+import presentation_assistant.env_set as env
+env.env_set()
+print(os.getcwd())
+
+import streamlit as st
+import PyPDF2
+import openai
+import subprocess
+from io import BytesIO
+from pptx import Presentation
+
+
+import presentation_assistant.presentation_assistant as pa
+
+tab1, tab2, tab3 = st.tabs(['What is PA!?', 'Text2PPT', 'PPT2Script'])
+
+with tab1:
+    st.header('Introduction')
+    st.title('PA!(Presentation Assistant):sparkles:')
+    contents = """
+    ▶ Based on the content entered by the user, it :blue[automatically creates] PPT and
+      provides a presentation :red[script] to improve presentation skills!"""
+    st.markdown(contents)
+    st.markdown('-------------------------')
+    st.header('How to use')
+    st.subheader('Text2PPT')
+    contents = """
+    ▶ If the user provides a link or file, we will :blue[create a presentation material] for you!
+      The user only needs to select the desired theme (template) type and number of pages!"""
+    st.markdown(contents)
+    st.subheader('PPT2Script')
+    contents = """
+    ▶ If the user provides PPT or PDF presentation materials, we will automatically create a :blue[presentation script] for you!"""
+    st.markdown(contents)
+
+    # Test
+    # test_ppt_theme = "--reference-doc="+"/home/user/app/template/blue"+".pptx"
+    # subprocess.run(["/home/user/app/pandoc-2.14.2/bin/pandoc", "text2ppt_test.md", "-t", "pptx", test_ppt_theme, "-o", "output.pptx"], capture_output=True)
+    # print(os.listdir(os.getcwd()))
+    # prs = Presentation("output.pptx")
+    # binary_output = BytesIO()
+    # prs.save(binary_output)
+    # st.download_button(label="Download PPT",
+    #                  data = binary_output.getvalue(),
+    #                 file_name="export_output.pptx",
+    #                 mime='application/octet-stream', key = "<Text2PPT_test_download>")
+
+
+with tab2:
+    st.header('Text2PPT')
+    gpt_token = st.text_input('Please enter your ChatGPT API token.', key="<Text2PPT_token>")
+    st.markdown('-------------------------')
+
+    st.subheader(':computer: PPT Auto Generator :computer:')
+
+    thema_select = st.selectbox(
+        'Please select the template you want.',
+        ['default', 'yellow', 'gradation_green', 'blue', 'green', 'custom'])
+    
+    if thema_select == "custom":
+        uploaded_template_file = st.file_uploader('Choose File!', type='pptx', key="<template_uploader>")
+
+    st.markdown('-------------------------')
+
+    page_choice = st.slider('Number of PPT pages', min_value=2, max_value=10, step=1, value=5)
+
+    st.markdown('-------------------------')
+
+    my_order = ['Text', 'Link', 'PDF']
+    status = st.radio('Please select the file type and enter the content! :smile: ', my_order)
+
+    # First method
+    if status == my_order[0]:
+        input_text = st.text_area('Enter TEXT', height=5)
+
+    elif status == my_order[1]:
+        input_text = st.text_area('Enter URL', height=5)
+
+    elif status == my_order[2]:
+        input_text = st.file_uploader('Upload PDF', type=['pdf'])
+
+    input_text_check = st.button('Confirm', key="<Text2PPT_start>")
+
+    st.markdown('-------------------------')
+
+    if input_text_check == True:
+        with st.spinner('Wait for it...'):
+            pa.text2ppt(gpt_token, pa.generate_text2ppt_input_prompt(status, input_text, page_choice), thema_select)
+            prs = Presentation("text2ppt_output.pptx")
+            binary_output = BytesIO()
+            prs.save(binary_output)
+        st.success('Done!')
+        st.download_button(label="Download PPT",
+                            data = binary_output.getvalue(),
+                            file_name="export_output.pptx",
+                            mime='application/octet-stream', key = "<Text2PPT_download>")
+
+with tab3:
+    st.header('PPT2Script')
+    st.subheader(':computer: Script Auto Generator :computer:')
+    gpt_token = st.text_input('Please enter your ChatGPT API token.', key="<PPT2Script_token>")
+    st.markdown('-------------------------')
+
+    st.subheader(':bookmark_tabs:Presentation Script Generator')
+
+    file_order = ['PDF', 'PPT']
+    choose = st.radio('Please select the file format of the presentation material', file_order)
+
+    if choose == file_order[0]:
+        uploaded_file = st.file_uploader('Choose File!', type='pdf', key="<PPT2Script_pdf_uploader>")
+    elif choose == file_order[1]:
+        uploaded_file = st.file_uploader('Choose File!', type='pptx', key="<PPT2Script_ppt_uploader>")
+
+    input_file_check = st.button('Confirm', key="<PPT2Script_start>")  # When this button is pressed, the input file should be passed
+    st.markdown('-------------------------')
+
+    if input_file_check == True:
+        with st.spinner('Wait for it...'):
+            with open(uploaded_file.name, mode='wb') as w:
+                w.write(uploaded_file.getvalue())
+
+            script = pa.ppt2script(gpt_token, uploaded_file.name, choose)
+
+        st.success('Done!')
+        st.download_button('Download Script',
+                           data=script, file_name="script_output.txt", key="<PPT2Script_download>")

pandoc-2.14.2/bin/pandoc

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a2cb859c04eddd0f36e18dca1466d2957e9686bdbe113bc96e5476b59aa87af
+size 76884208

pandoc-2.14.2/share/man/man1/pandoc.1.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67bd56ab5fa90753f6af440b6cf29e39525f517c2d9a55183e3c7f89d0540b9c
+size 70192

presentation_assistant/env_set.py

@@ -0,0 +1,39 @@
+import subprocess
+import os
+import sys
+
+sys.path.append("/home/user/app") 
+
+def env_set():
+    # 패키지가 이미 설치되어 있는지 확인
+    installed_packages = subprocess.check_output("pip list", shell=True).decode()
+
+    if "PyPDF2" not in installed_packages:
+        subprocess.run(f"pip install PyPDF2", shell=True)
+
+    if "openai" not in installed_packages:
+        subprocess.run(f"pip install -q openai", shell=True)
+
+    if "python-pptx" not in installed_packages:
+        subprocess.run(f"pip install python-pptx", shell=True)
+
+    # pandoc 실행 권한 설정
+    if not os.access("./pandoc-2.14.2/bin/pandoc", os.X_OK):
+        subprocess.run(f"chmod +x ./pandoc-2.14.2/bin/pandoc", shell=True)
+
+    # 디렉토리 권한 설정
+    if not os.access(".", os.W_OK):
+        subprocess.run(f"chmod -R 777 .", shell=True)
+
+    # subprocess.run(f"wget https://github.com/jgm/pandoc/releases/download/2.14.2/pandoc-2.14.2-linux-amd64.tar.gz", shell=True)
+    # subprocess.run(f"tar -xvf pandoc-2.14.2-linux-amd64.tar.gz", shell=True)
+    # subprocess.run(f"wget https://nodejs.org/dist/v17.0.1/node-v17.0.1-linux-x64.tar.xz", shell=True)
+    # subprocess.run(f"tar -xf node-v17.0.1-linux-x64.tar.xz", shell=True)
+    # subprocess.run(f"mv node-v17.0.1-linux-x64 node", shell=True)
+    # subprocess.run(f"export PATH=/home/user/app/node/bin:$PATH", shell=True)
+    # subprocess.run(f"/home/user/app/node/ install -g npm", shell=True)
+    # subprocess.run(f"apt-get install -f", shell=True)
+    # subprocess.run(f"apt-get install -y fonts-nanum", shell=True)
+    # subprocess.run(f"fc-cache -fv", shell=True)
+    # subprocess.run(f"rm ~/.cache/matplotlib -rf", shell=True)
+

presentation_assistant/presentation_assistant.py

@@ -0,0 +1,159 @@
+import os
+import PyPDF2
+from pptx import Presentation
+import openai
+import subprocess
+from io import BytesIO
+import sys
+
+sys.path.append("/home/user/app") 
+
+# Function to generate text2ppt input prompt
+def generate_text2ppt_input_prompt(input_type, input_value, input_pages):
+    header = """
+    Assume you are a designer creating a PPT using markdown syntax, and write a PPT of %s pages.
+    +++ Summarize the content or link below in markdown language, adhering to the rules in ===, and refer to the slide examples in ~~~.
+    +++
+    """ % input_pages
+
+    summary_value = ""
+
+    if input_type == "Link":
+        summary_value += input_value
+        summary_value += "\n"
+    elif input_type == "Text":
+        summary_value += input_value
+        summary_value += "\n"
+    elif input_type == "PDF":
+        with open(input_value, 'rb') as pdf_file:
+            pdf_reader = PyPDF2.PdfReader(pdf_file)
+            num_pages = len(pdf_reader.pages)
+
+        # Convert the content of each page to a string.
+        text = ""
+        for page_num in range(num_pages):
+            page = pdf_reader.pages[page_num]
+            text += page.extract_text()
+        summary_value += text
+        summary_value += "\n"
+    else:
+        print("ERROR: Invalid input")
+
+    rule_value = """
+    ===
+    - Write factually only about the content or link provided.
+    - Always use --- as a slide divider.
+    - Design and arrange the slides diversely with appropriate shapes, images(![Image](Image link), https://unsplash.com/ko/images/stock/non-copyrighted for actual use), tables(|-|), quotes(>), emphasis(bold, ``), emojis(https://kr.piliapp.com/twitter-symbols/), icons (https://kr.piliapp.com/symbol/#popular).
+    - Use emojis only once in every two pages, and use various other designs.
+    - When using images and tables, specify the size considering the page size so that all the text content appears.
+    - Make Slide 1 the title, for a total of %s pages.
+    - Write the content of the PPT richly in markdown.
+    - Don't explain slide by slide, just write the code.
+    - Don't write using the content of the example, just refer to the format.
+    ~~~
+    <!-- Slide 0. Slide Topic -->
+    # Slide Title
+    ![Image link](https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo-with-title.png)
+    - This is 🤗**TEXT2PPT service PA!** using ChatGPT.
+    - Converts `link`,`text`, `PDF` input or upload into PPT. 
+    """ % input_pages
+
+    return header + summary_value + rule_value
+
+
+# Function to execute text2ppt
+def text2ppt(token_key, input_prompt, input_theme):
+    openai.api_key = token_key
+
+    messages = [
+        {"role": "system", "content": "You are a kind helpful PPT designer."},
+    ]
+
+    message = input_prompt
+
+    if message:
+        messages.append(
+            {"role": "user", "content": message},
+        )
+        chat = openai.ChatCompletion.create(
+            model="gpt-3.5-turbo-0301", messages=messages
+        )
+
+    reply = chat.choices[0].message.content
+    messages.append({"role": "assistant", "content": reply})
+
+    md_text = reply[4:] if reply[:3] == "---" else reply
+    md_text_list = md_text.split('\n')
+
+    f = open("text2ppt_input.md", 'w')
+    for i in range(0, len(md_text_list)):
+        data = md_text_list[i] + "\n"
+        f.write(data)
+    f.close()
+
+    if input_theme == 'default':
+        subprocess.run(["/home/user/app/pandoc-2.14.2/bin/pandoc", "/home/user/app/text2ppt_input.md", "-t", "pptx", "-o", "/home/user/app/text2ppt_output.pptx"], capture_output=True)
+    else:
+        ppt_theme = "--reference-doc=/home/user/app/template/"+input_theme+".pptx"
+        subprocess.run(["/home/user/app/pandoc-2.14.2/bin/pandoc", "/home/user/app/text2ppt_input.md", "-t", "pptx", ppt_theme, "-o", "/home/user/app/text2ppt_output.pptx"], capture_output=True)
+    
+
+def ppt2script(token_key, input_file, input_type):
+    openai.api_key = token_key
+
+    if input_type=="PDF":
+        with open(input_file, 'rb') as pdf_file:
+            pdf_reader = PyPDF2.PdfReader(pdf_file)
+            num_pages = len(pdf_reader.pages)
+    
+            # Convert the content of each page to a string.
+            text = ""
+            for page_num in range(num_pages):
+                page = pdf_reader.pages[page_num]
+                text += "[PAGE_NUM " + str(page_num + 1) + "]"
+                text += page.extract_text()
+    else:
+        prs = Presentation(input_file)
+
+        text = ""
+        page_num = 0
+        for slide in prs.slides:
+            text += "[PAGE_NUM " + str(page_num + 1) + "]"
+            page_num += 1
+            for shape in slide.shapes:
+                if not shape.has_text_frame:
+                    continue
+                for paragraph in shape.text_frame.paragraphs:
+                    for run in paragraph.runs:
+                        text += run.text
+
+    header = """
+    You are an assistant helping with PPT presentations. 
+    ~~~Follow the rules below and write a presentation script for the PPT content below.
+    ~~~
+    - When [PAGE_NUM 1], where 1 is the page number, write a presentation script for each page number.
+    - Write only in text without using markdown language.
+    - Add additional explanations or examples to the PPT content.
+    ---
+    """
+
+    input_prompt = header + text
+
+    messages = [
+        {"role": "system", "content": "You are a kind helpful PPT Assistant."},
+    ]
+
+    message = input_prompt
+
+    if message:
+        messages.append(
+            {"role": "user", "content": message},
+        )
+        chat = openai.ChatCompletion.create(
+            model="gpt-3.5-turbo-0301", messages=messages
+        )
+
+    reply = chat.choices[0].message.content
+    messages.append({"role": "assistant", "content": reply})
+
+    return reply

requirements.txt

@@ -0,0 +1,2 @@
+streamlit==0.88.0
+altair==4.1.0

text2ppt_test.md

@@ -0,0 +1,87 @@
+<!-- Slide 1. 인공지능 개요 -->
+# 인공지능 개요
+> 인공지능(Artificial Intelligence)은 인간의 지능을 모방하여 만든 컴퓨터 시스템입니다. 
+
+---
+
+<!-- Slide 2. 인공지능 분류 -->
+# 인공지능 분류
+<table>
+    <thead>
+        <tr>
+            <th>분류</th>
+            <th>설명</th>
+        </tr>
+    </thead>
+    <tbody>
+        <tr>
+            <td>강한 인공지능</td>
+            <td>사람과 같이 모든 인간의 작업을 수행</td>
+        </tr>
+        <tr>
+            <td>약한 인공지능</td>
+            <td>일부분의 작업만 수행</td>
+        </tr>
+    </tbody>
+</table>
+
+---
+
+<!-- Slide 3. 인공지능 적용 분야 -->
+# 인공지능 적용 분야
+- 자율주행 자동차
+- 음성인식 기술
+- 인공지능 영상 인식 기술
+- 자동 번역
+---
+
+<!-- Slide 4. 인공신경망 -->
+# 인공신경망
+- 생물학적 퍼셉트론 구조 참고
+- 입력 계층, 출력 계층, 은닉 계층으로 구성
+![인공신경망 이미지](https://upload.wikimedia.org/wikipedia/commons/4/46/Colored_neural_network.svg)
+
+---
+
+<!-- Slide 5. 심층학습 -->
+# 심층학습
+- 다층퍼셉트론(MLP)과 같은 네트워크 위에 다양한 레이어를 쌓아 올림
+- 이미지/음성/텍스트 분석 분야에서 광범위하게 활용
+
+---
+
+<!-- Slide 6. 강화학습 -->
+# 강화학습
+- 상호작용을 통해 주어진 환경을 학습하는 방식
+- 에이전트가 최적의 행동을 취하도록 보상하는 방식
+- 슈퍼마리오 게임/바둑/장기/알파고 등에서 활용
+
+---
+
+<!-- Slide 7. 자연어 처리 -->
+# 자연어 처리(NLP)
+- 인간의 언어(음성 또는 문자)를 컴퓨터가 이해하고 처리
+- 딥러닝 이용한 자동번역, 챗봇, 텍스트 감정분석 등
+
+---
+
+<!-- Slide 8. 딥러닝 -->
+# 딥러닝
+- 심층신경망을 이용한 머신러닝 방법론
+- 컴퓨터가 인간의 학습 능력을 모방하여 자동으로 패턴 인식 달성
+- 이미지/음성/자연어 처리 등에서 활용
+
+---
+
+<!-- Slide 9. 인공지능의 미래 -->
+# 인공지능의 미래
+- 기술의 발전과 함께 인간의 여러 분야에서 고도화
+- 더욱 편리하고 효율적인 인간의 일상생활에 보다 많은 기여
+
+---
+
+<!-- Slide 10. # PA! -->
+# PA! 🎉
+- 🤗**TEXT2PPT** 서비스 PA!를 이용해주셔서 감사합니다.
+- 리뷰나 건의사항은 언제든지 환영합니다!
+- 📧문의: pa@pa.com