FEAT: 作业提交V4版本

selected_homework/openai-translator_v4/README.md

@@ -0,0 +1,19 @@
+### 作业需求
+基于 ChatGLM2-6B 实现带图形化界面的 openai-translator
+
+### 作业总结
++ [openai_api_demo](openai_api_demo)
+
+利用的chatGLM中的api_demo进行调整为server项，故没有采用ChatGLM2-6b，而是ChatGLM3-6b
+
+运行起来需要：
+1. git clone https://www.modelscope.cn/ZhipuAI/chatglm3-6b.git
+2. 确保机器有足够的资源【俺没有。。。故暂未实现。。。】
+
++ [ai_translator](ai_translator)
+
+将历史的 
+from langchain_openai import ChatOpenAI 
+替换为
+from langchain.llms import ChatGLM
+并针对ChatGLM做参数匹配

selected_homework/openai-translator_v4/ai_translator/book/__init__.py

@@ -0,0 +1,3 @@
+from .book import Book
+from .page import Page
+from .content import ContentType, Content, TableContent

selected_homework/openai-translator_v4/ai_translator/book/book.py

@@ -0,0 +1,9 @@
+from .page import Page
+
+class Book:
+    def __init__(self, pdf_file_path):
+        self.pdf_file_path = pdf_file_path
+        self.pages = []
+
+    def add_page(self, page: Page):
+        self.pages.append(page)

selected_homework/openai-translator_v4/ai_translator/book/content.py

@@ -0,0 +1,85 @@
+import pandas as pd
+
+from enum import Enum, auto
+from PIL import Image as PILImage
+from utils import LOG
+from io import StringIO
+
+class ContentType(Enum):
+    TEXT = auto()
+    TABLE = auto()
+    IMAGE = auto()
+
+class Content:
+    def __init__(self, content_type, original, translation=None):
+        self.content_type = content_type
+        self.original = original
+        self.translation = translation
+        self.status = False
+
+    def set_translation(self, translation, status):
+        if not self.check_translation_type(translation):
+            raise ValueError(f"Invalid translation type. Expected {self.content_type}, but got {type(translation)}")
+        self.translation = translation
+        self.status = status
+
+    def check_translation_type(self, translation):
+        if self.content_type == ContentType.TEXT and isinstance(translation, str):
+            return True
+        elif self.content_type == ContentType.TABLE and isinstance(translation, list):
+            return True
+        elif self.content_type == ContentType.IMAGE and isinstance(translation, PILImage.Image):
+            return True
+        return False
+
+    def __str__(self):
+        return self.original
+
+
+class TableContent(Content):
+    def __init__(self, data, translation=None):
+        df = pd.DataFrame(data)
+
+        # Verify if the number of rows and columns in the data and DataFrame object match
+        if len(data) != len(df) or len(data[0]) != len(df.columns):
+            raise ValueError("The number of rows and columns in the extracted table data and DataFrame object do not match.")
+
+        super().__init__(ContentType.TABLE, df)
+
+    def set_translation(self, translation, status):
+        try:
+            if not isinstance(translation, str):
+                raise ValueError(f"Invalid translation type. Expected str, but got {type(translation)}")
+
+            LOG.debug(f"[translation]\n{translation}")
+            # Extract column names from the first set of brackets
+            header = translation.split(']')[0][1:].split(', ')
+            # Extract data rows from the remaining brackets
+            data_rows = translation.split('] ')[1:]
+            # Replace Chinese punctuation and split each row into a list of values
+            data_rows = [row[1:-1].split(', ') for row in data_rows]
+            # Create a DataFrame using the extracted header and data
+            translated_df = pd.DataFrame(data_rows, columns=header)
+            LOG.debug(f"[translated_df]\n{translated_df}")
+            self.translation = translated_df
+            self.status = status
+        except Exception as e:
+            LOG.error(f"An error occurred during table translation: {e}")
+            self.translation = None
+            self.status = False
+
+    def __str__(self):
+        return self.original.to_string(header=False, index=False)
+
+    def iter_items(self, translated=False):
+        target_df = self.translation if translated else self.original
+        for row_idx, row in target_df.iterrows():
+            for col_idx, item in enumerate(row):
+                yield (row_idx, col_idx, item)
+
+    def update_item(self, row_idx, col_idx, new_value, translated=False):
+        target_df = self.translation if translated else self.original
+        target_df.at[row_idx, col_idx] = new_value
+
+    def get_original_as_str(self):
+        return self.original.to_string(header=False, index=False)

selected_homework/openai-translator_v4/ai_translator/book/page.py

@@ -0,0 +1,8 @@
+from .content import Content
+
+class Page:
+    def __init__(self):
+        self.contents = []
+
+    def add_content(self, content: Content):
+        self.contents.append(content)

selected_homework/openai-translator_v4/ai_translator/config.yaml

@@ -0,0 +1,5 @@
+model_name: "chatglm2-6b"
+input_file: "tests/test.pdf"
+output_file_format: "markdown"
+source_language: "English"
+target_language: "Chinese"

selected_homework/openai-translator_v4/ai_translator/flask_server.py

@@ -0,0 +1,71 @@
+import sys
+import os
+
+sys.path.append(os.path.dirname(os.path.abspath(__file__)))
+
+from flask import Flask, request, send_file, jsonify
+from translator import PDFTranslator, TranslationConfig
+from utils import ArgumentParser, LOG
+
+app = Flask(__name__)
+
+TEMP_FILE_DIR = "flask_temps/"
+
+@app.route('/translation', methods=['POST'])
+def translation():
+    try:
+        input_file = request.files['input_file']
+        source_language = request.form.get('source_language', 'English')
+        target_language = request.form.get('target_language', 'Chinese')
+
+        LOG.debug(f"[input_file]\n{input_file}")
+        LOG.debug(f"[input_file.filename]\n{input_file.filename}")
+
+        if input_file and input_file.filename:
+            # # 创建临时文件
+            input_file_path = TEMP_FILE_DIR+input_file.filename
+            LOG.debug(f"[input_file_path]\n{input_file_path}")
+
+            input_file.save(input_file_path)
+
+            # 调用翻译函数
+            output_file_path = Translator.translate_pdf(
+                input_file=input_file_path,
+                source_language=source_language,
+                target_language=target_language)
+
+            # 移除临时文件
+            # os.remove(input_file_path)
+
+            # 构造完整的文件路径
+            output_file_path = os.getcwd() + "/" + output_file_path
+            LOG.debug(output_file_path)
+
+            # 返回翻译后的文件
+            return send_file(output_file_path, as_attachment=True)
+    except Exception as e:
+        response = {
+            'status': 'error',
+            'message': str(e)
+        }
+        return jsonify(response), 400
+
+
+def initialize_translator():
+    # 解析命令行
+    argument_parser = ArgumentParser()
+    args = argument_parser.parse_arguments()
+
+    # 初始化配置单例
+    config = TranslationConfig()
+    config.initialize(args)    
+    # 实例化 PDFTranslator 类，并调用 translate_pdf() 方法
+    global Translator
+    Translator = PDFTranslator(config.model_name)
+
+
+if __name__ == "__main__":
+    # 初始化 translator
+    initialize_translator()
+    # 启动 Flask Web Server
+    app.run(host="0.0.0.0", port=5000, debug=True)

selected_homework/openai-translator_v4/ai_translator/gradio_server.py

@@ -0,0 +1,59 @@
+import sys
+import os
+import gradio as gr
+
+
+sys.path.append(os.path.dirname(os.path.abspath(__file__)))
+
+from utils import ArgumentParser, LOG
+from translator import PDFTranslator, TranslationConfig
+
+
+def translation(input_file, source_language, target_language, translate_style):
+    LOG.debug(
+        f"[翻译任务]\n源文件: {input_file.name}\n源语言: {source_language}\n目标语言: {target_language}\n翻译风格: {translate_style}")
+
+    output_file_path = Translator.translate_pdf(
+        input_file.name, source_language=source_language,
+        target_language=target_language, translate_style=translate_style
+    )
+
+    return output_file_path
+
+def launch_gradio():
+
+    iface = gr.Interface(
+        fn=translation,
+        title="[Homework]OpenAI-Translator v4(接入GLM3-6b)",
+        inputs=[
+            gr.File(label="上传PDF文件"),
+            gr.Textbox(label="源语言（默认：英文）", placeholder="English", value="English"),
+            gr.Textbox(label="目标语言（默认：中文）", placeholder="Chinese", value="Chinese"),
+            gr.Radio(["Normal people", "Children", "Professor"]),
+        ],
+        outputs=[
+            gr.File(label="下载翻译文件")
+        ],
+        allow_flagging="never"
+    )
+
+    iface.launch(share=True, server_name="0.0.0.0")
+
+def initialize_translator():
+    # 解析命令行
+    argument_parser = ArgumentParser()
+    args = argument_parser.parse_arguments()
+
+    # 初始化配置单例
+    config = TranslationConfig()
+    config.initialize(args)    
+    # 实例化 PDFTranslator 类，并调用 translate_pdf() 方法
+    global Translator
+    Translator = PDFTranslator(config.model_name)
+
+
+if __name__ == "__main__":
+    # 初始化 translator
+    initialize_translator()
+    # 启动 Gradio 服务
+    launch_gradio()

selected_homework/openai-translator_v4/ai_translator/tests/test_translated.md

@@ -0,0 +1,19 @@
+测试数据
+这个数据集包含了由OpenAI的AI语言模型ChatGPT提供的两个测试样本。
+这些样本包括一个Markdown表格和一个英文文本段落，可以用来测试支持文本和表格格式的英译中翻译软件。
+文本测试
+快速的棕色狐狸跳过懒狗。这个句子包含了英语字母表中的每个字母至少一次。句子是经常用来测试字体、键盘和其他与文本相关的工具的。除了英语，其他许多语言也有句子。由于语言的独特特点，有些句子更难构造。
+
+| 水果 | 颜色 | 价格（美元） |
+| --- | --- | --- |
+| 苹果 | 红色 | 1.2 |
+| 香蕉 | 黄色 | 0.5 |
+| 橙子 | 橙色 | 0.8 |
+| 草莓 | 红色 | 2.5 |
+| 蓝莓 | 蓝色 | 3.0 |
+| 猕猴桃 | 绿色 | 1.0 |
+| 芒果 | 橙色 | 1.5 |
+| 葡萄 | 紫色 | 2.00 |
+
+---
+

selected_homework/openai-translator_v4/ai_translator/translator/__init__.py

@@ -0,0 +1,2 @@
+from .pdf_translator import PDFTranslator
+from .translation_config import TranslationConfig

selected_homework/openai-translator_v4/ai_translator/translator/exceptions.py

@@ -0,0 +1,5 @@
+class PageOutOfRangeException(Exception):
+    def __init__(self, book_pages, requested_pages):
+        self.book_pages = book_pages
+        self.requested_pages = requested_pages
+        super().__init__(f"Page out of range: Book has {book_pages} pages, but {requested_pages} pages were requested.")

selected_homework/openai-translator_v4/ai_translator/translator/pdf_parser.py

@@ -0,0 +1,58 @@
+import pdfplumber
+from typing import Optional
+from book import Book, Page, Content, ContentType, TableContent
+from translator.exceptions import PageOutOfRangeException
+from utils import LOG
+
+
+class PDFParser:
+    def __init__(self):
+        pass
+
+    def parse_pdf(self, pdf_file_path: str, pages: Optional[int] = None) -> Book:
+        book = Book(pdf_file_path)
+
+        with pdfplumber.open(pdf_file_path) as pdf:
+            if pages is not None and pages > len(pdf.pages):
+                raise PageOutOfRangeException(len(pdf.pages), pages)
+
+            if pages is None:
+                pages_to_parse = pdf.pages
+            else:
+                pages_to_parse = pdf.pages[:pages]
+
+            for pdf_page in pages_to_parse:
+                page = Page()
+
+                # Store the original text content
+                raw_text = pdf_page.extract_text()
+                tables = pdf_page.extract_tables()
+
+                # Remove each cell's content from the original text
+                for table_data in tables:
+                    for row in table_data:
+                        for cell in row:
+                            raw_text = raw_text.replace(cell, "", 1)
+
+                # Handling text
+                if raw_text:
+                    # Remove empty lines and leading/trailing whitespaces
+                    raw_text_lines = raw_text.splitlines()
+                    cleaned_raw_text_lines = [line.strip() for line in raw_text_lines if line.strip()]
+                    cleaned_raw_text = "\n".join(cleaned_raw_text_lines)
+
+                    text_content = Content(content_type=ContentType.TEXT, original=cleaned_raw_text)
+                    page.add_content(text_content)
+                    LOG.debug(f"[raw_text]\n {cleaned_raw_text}")
+
+
+
+                # Handling tables
+                if tables:
+                    table = TableContent(tables)
+                    page.add_content(table)
+                    LOG.debug(f"[table]\n{table}")
+
+                book.add_page(page)
+
+        return book

selected_homework/openai-translator_v4/ai_translator/translator/pdf_translator.py

@@ -0,0 +1,30 @@
+from typing import Optional
+from translator.pdf_parser import PDFParser
+from translator.writer import Writer
+from translator.translation_chain import TranslationChain
+
+class PDFTranslator:
+    def __init__(self, model_name: str):
+        self.translate_chain = TranslationChain(model_name)
+        self.pdf_parser = PDFParser()
+        self.writer = Writer()
+
+    def translate_pdf(self,
+                    input_file: str,
+                    output_file_format: str = 'markdown',
+                    source_language: str = "English",
+                    target_language: str = 'Chinese',
+                    translate_style: str = "Normal Style",
+                    pages: Optional[int] = None):
+
+        self.book = self.pdf_parser.parse_pdf(input_file, pages)
+
+        for page_idx, page in enumerate(self.book.pages):
+            for content_idx, content in enumerate(page.contents):
+                # Translate content.original
+                translation, status = self.translate_chain.run(
+                    content, source_language, target_language, translate_style)
+                # Update the content in self.book.pages directly
+                self.book.pages[page_idx].contents[content_idx].set_translation(translation, status)
+
+        return self.writer.save_translated_book(self.book, output_file_format)

selected_homework/openai-translator_v4/ai_translator/translator/translation_chain.py

@@ -0,0 +1,54 @@
+from langchain.llms import ChatGLM
+from langchain.chains import LLMChain
+
+from langchain.prompts.chat import (
+    ChatPromptTemplate,
+    SystemMessagePromptTemplate,
+    HumanMessagePromptTemplate,
+)
+
+from utils import LOG
+
+class TranslationChain:
+    def __init__(self, model_name: str = "chatglm2-6b", verbose: bool = True):
+
+        # 翻译任务指令始终由 System 角色承担
+        template = (
+            """You are a translation expert, proficient in various languages. \n
+            Translates {source_language} to {target_language}. \n
+            Speak like {translate_style}."""
+        )
+        system_message_prompt = SystemMessagePromptTemplate.from_template(template)
+
+        # 待翻译文本由 Human 角色输入
+        human_template = "{text}"
+        human_message_prompt = HumanMessagePromptTemplate.from_template(human_template)
+
+        # 使用 System 和 Human 角色的提示模板构造 ChatPromptTemplate
+        chat_prompt_template = ChatPromptTemplate.from_messages(
+            [system_message_prompt, human_message_prompt]
+        )
+
+        chat = ChatGLM(
+            endpoint_url="http://127.0.0.1:8000",
+            max_token=8000
+        )
+
+        self.chain = LLMChain(llm=chat, prompt=chat_prompt_template, verbose=verbose)
+
+    def run(self, text: str,
+            source_language: str,
+            target_language: str,
+            translate_style: str) -> (str, bool):
+        result = ""
+        try:
+            result = self.chain.run({
+                "text": text,
+                "source_language": source_language,
+                "target_language": target_language,
+                "translate_style": translate_style
+            })
+        except Exception as e:
+            LOG.error(f"An error occurred during translation: {e}")
+            return result, False
+        return result, True

selected_homework/openai-translator_v4/ai_translator/translator/translation_config.py

@@ -0,0 +1,29 @@
+import yaml
+
+class TranslationConfig:
+    _instance = None
+
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super(TranslationConfig, cls).__new__(cls)
+            cls._instance._config = None
+        return cls._instance
+
+    def initialize(self, args):
+        with open(args.config_file, "r") as f:
+            config = yaml.safe_load(f)
+
+        # Use the argparse Namespace to update the configuration
+        overridden_values = {
+            key: value for key, value in vars(args).items() if key in config and value is not None
+        }
+        config.update(overridden_values)    
+
+        # Store the original config dictionary
+        self._instance._config = config
+
+    def __getattr__(self, name):
+        # Try to get attribute from _config
+        if self._instance._config and name in self._instance._config:
+            return self._instance._config[name]
+        raise AttributeError(f"'TranslationConfig' object has no attribute '{name}'")

selected_homework/openai-translator_v4/ai_translator/translator/writer.py

@@ -0,0 +1,114 @@
+import os
+from reportlab.lib import colors, pagesizes, units
+from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
+from reportlab.pdfbase import pdfmetrics
+from reportlab.pdfbase.ttfonts import TTFont
+from reportlab.platypus import (
+    SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle, PageBreak
+)
+
+from book import Book, ContentType
+from utils import LOG
+
+class Writer:
+    def __init__(self):
+        pass
+
+    def save_translated_book(self, book: Book, ouput_file_format: str):
+        LOG.debug(ouput_file_format)
+
+        if ouput_file_format.lower() == "pdf":
+            output_file_path = self._save_translated_book_pdf(book)
+        elif ouput_file_format.lower() == "markdown":
+            output_file_path = self._save_translated_book_markdown(book)
+        else:
+            LOG.error(f"不支持文件类型: {ouput_file_format}")
+            return ""
+
+        LOG.info(f"翻译完成，文件保存至: {output_file_path}")
+
+        return output_file_path
+
+
+    def _save_translated_book_pdf(self, book: Book, output_file_path: str = None):
+
+        output_file_path = book.pdf_file_path.replace('.pdf', f'_translated.pdf')
+
+        LOG.info(f"开始导出: {output_file_path}")
+
+        # Register Chinese font
+        font_path = "../fonts/simsun.ttc"  # 请将此路径替换为您的字体文件路径
+        pdfmetrics.registerFont(TTFont("SimSun", font_path))
+
+        # Create a new ParagraphStyle with the SimSun font
+        simsun_style = ParagraphStyle('SimSun', fontName='SimSun', fontSize=12, leading=14)
+
+        # Create a PDF document
+        doc = SimpleDocTemplate(output_file_path, pagesize=pagesizes.letter)
+        styles = getSampleStyleSheet()
+        story = []
+
+        # Iterate over the pages and contents
+        for page in book.pages:
+            for content in page.contents:
+                if content.status:
+                    if content.content_type == ContentType.TEXT:
+                        # Add translated text to the PDF
+                        text = content.translation
+                        para = Paragraph(text, simsun_style)
+                        story.append(para)
+
+                    elif content.content_type == ContentType.TABLE:
+                        # Add table to the PDF
+                        table = content.translation
+                        table_style = TableStyle([
+                            ('BACKGROUND', (0, 0), (-1, 0), colors.grey),
+                            ('TEXTCOLOR', (0, 0), (-1, 0), colors.whitesmoke),
+                            ('ALIGN', (0, 0), (-1, -1), 'CENTER'),
+                            ('FONTNAME', (0, 0), (-1, 0), 'SimSun'),  # 更改表头字体为 "SimSun"
+                            ('FONTSIZE', (0, 0), (-1, 0), 14),
+                            ('BOTTOMPADDING', (0, 0), (-1, 0), 12),
+                            ('BACKGROUND', (0, 1), (-1, -1), colors.beige),
+                            ('FONTNAME', (0, 1), (-1, -1), 'SimSun'),  # 更改表格中的字体为 "SimSun"
+                            ('GRID', (0, 0), (-1, -1), 1, colors.black)
+                        ])
+                        pdf_table = Table(table.values.tolist())
+                        pdf_table.setStyle(table_style)
+                        story.append(pdf_table)
+            # Add a page break after each page except the last one
+            if page != book.pages[-1]:
+                story.append(PageBreak())
+
+        # Save the translated book as a new PDF file
+        doc.build(story)
+        return output_file_path
+
+
+    def _save_translated_book_markdown(self, book: Book, output_file_path: str = None):
+        output_file_path = book.pdf_file_path.replace('.pdf', f'_translated.md')
+
+        LOG.info(f"开始导出: {output_file_path}")
+        with open(output_file_path, 'w', encoding='utf-8') as output_file:
+            # Iterate over the pages and contents
+            for page in book.pages:
+                for content in page.contents:
+                    if content.status:
+                        if content.content_type == ContentType.TEXT:
+                            # Add translated text to the Markdown file
+                            text = content.translation
+                            output_file.write(text + '\n\n')
+
+                        elif content.content_type == ContentType.TABLE:
+                            # Add table to the Markdown file
+                            table = content.translation
+                            header = '| ' + ' | '.join(str(column) for column in table.columns) + ' |' + '\n'
+                            separator = '| ' + ' | '.join(['---'] * len(table.columns)) + ' |' + '\n'
+                            # body = '\n'.join(['| ' + ' | '.join(row) + ' |' for row in table.values.tolist()]) + '\n\n'
+                            body = '\n'.join(['| ' + ' | '.join(str(cell) for cell in row) + ' |' for row in table.values.tolist()]) + '\n\n'
+                            output_file.write(header + separator + body)
+
+                # Add a page break (horizontal rule) after each page except the last one
+                if page != book.pages[-1]:
+                    output_file.write('---\n\n')
+
+        return output_file_path

selected_homework/openai-translator_v4/ai_translator/utils/__init__.py

@@ -0,0 +1,2 @@
+from .argument_parser import ArgumentParser
+from .logger import LOG

selected_homework/openai-translator_v4/ai_translator/utils/argument_parser.py

@@ -0,0 +1,15 @@
+import argparse
+
+class ArgumentParser:
+    def __init__(self):
+        self.parser = argparse.ArgumentParser(description='A translation tool that supports translations in any language pair.')
+        self.parser.add_argument('--config_file', type=str, default='config.yaml', help='Configuration file with model and API settings.')
+        self.parser.add_argument('--model_name', type=str, help='Name of the Large Language Model.')
+        self.parser.add_argument('--input_file', type=str, help='PDF file to translate.')
+        self.parser.add_argument('--output_file_format', type=str, help='The file format of translated book. Now supporting PDF and Markdown')
+        self.parser.add_argument('--source_language', type=str, help='The language of the original book to be translated.')
+        self.parser.add_argument('--target_language', type=str, help='The target language for translating the original book.')
+
+    def parse_arguments(self):
+        args = self.parser.parse_args()
+        return args

selected_homework/openai-translator_v4/ai_translator/utils/logger.py

@@ -0,0 +1,32 @@
+from loguru import logger
+import os
+import sys
+
+LOG_FILE = "translation.log"
+ROTATION_TIME = "02:00"
+
+class Logger:
+    def __init__(self, name="translation", log_dir="logs", debug=False):
+        if not os.path.exists(log_dir):
+            os.makedirs(log_dir)
+        log_file_path = os.path.join(log_dir, LOG_FILE)
+
+        # Remove default loguru handler
+        logger.remove()
+
+        # Add console handler with a specific log level
+        level = "DEBUG" if debug else "INFO"
+        logger.add(sys.stdout, level=level)
+        # Add file handler with a specific log level and timed rotation
+        logger.add(log_file_path, rotation=ROTATION_TIME, level="DEBUG")
+        self.logger = logger
+
+LOG = Logger(debug=True).logger
+
+if __name__ == "__main__":
+    log = Logger().logger
+
+    log.debug("This is a debug message.")
+    log.info("This is an info message.")
+    log.warning("This is a warning message.")
+    log.error("This is an error message.")

selected_homework/openai-translator_v4/openai_api_demo/docker-compose.yml

@@ -0,0 +1,44 @@
+version: "3.6"
+
+services:
+        glm3_api:
+                image: python:3.10.13-slim
+                restart: unless-stopped
+                working_dir: /glm3
+                container_name: glm3_api
+                env_file: ./.env
+                networks:
+                        - v_glm3
+                deploy:
+                        resources:
+                                reservations:
+                                        devices:
+                                                - driver: nvidia
+                                                  count: 1
+                                                  capabilities: [gpu]
+                environment:
+                        - MODEL_PATH=/models/chatglm3-6b
+                        - EMBEDDING_PATH=/models/bge-large-zh-v1.5
+                        - TZ=Asia/Shanghai
+                        - PYTHONDONTWRITEBYTECODE=1
+                        - PYTHONUNBUFFERED=1
+                        - DOCKER=True
+                ports:
+                        - 8100:8000
+                volumes:
+                        - ./:/glm3
+                        - ${LOCAL_MODEL_PATH}:/models/chatglm3-6b
+                        - ${LOCAL_EMBEDDING_MODEL_PATH}:/models/bge-large-zh-v1.5
+                command:
+                        - sh
+                        - -c
+                        - |
+                          sed -i s/deb.debian.org/mirrors.tencentyun.com/g /etc/apt/sources.list
+                          sed -i s/security.debian.org/mirrors.tencentyun.com/g /etc/apt/sources.list
+                          apt-get update
+                          python -m pip install -i  https://mirror.sjtu.edu.cn/pypi/web/simple --upgrade pip
+                          pip install -r requirements.txt -i  https://pypi.tuna.tsinghua.edu.cn/simple
+                          python api_server.py
+networks:
+        v_glm3:
+                driver: bridge

selected_homework/openai-translator_v4/openai_api_demo/langchain_openai_api.py

@@ -0,0 +1,55 @@
+"""
+This script is designed for interacting with a local GLM3 AI model using the `ChatGLM3` class
+from the `langchain_community` library. It facilitates continuous dialogue with the GLM3 model.
+
+1. Start the Local Model Service: Before running this script, you need to execute the `api_server.py` script
+to start the GLM3 model's service.
+2. Run the Script: The script includes functionality for initializing the LLMChain object and obtaining AI responses,
+allowing the user to input questions and receive AI answers.
+3. This demo is not support for streaming.
+
+"""
+from langchain.chains import LLMChain
+from langchain.prompts import PromptTemplate
+from langchain.schema.messages import HumanMessage, SystemMessage, AIMessage
+from langchain_community.llms.chatglm3 import ChatGLM3
+
+
+def initialize_llm_chain(messages: list):
+    template = "{input}"
+    prompt = PromptTemplate.from_template(template)
+
+    endpoint_url = "http://127.0.0.1:8000/v1/chat/completions"
+    llm = ChatGLM3(
+        endpoint_url=endpoint_url,
+        max_tokens=4096,
+        prefix_messages=messages,
+        top_p=0.9
+    )
+    return LLMChain(prompt=prompt, llm=llm)
+
+
+def get_ai_response(llm_chain, user_message):
+    ai_response = llm_chain.invoke({"input": user_message})
+    return ai_response
+
+
+def continuous_conversation():
+    messages = [
+        SystemMessage(content="You are an intelligent AI assistant, named ChatGLM3."),
+    ]
+    while True:
+        user_input = input("Human (or 'exit' to quit): ")
+        if user_input.lower() == 'exit':
+            break
+        llm_chain = initialize_llm_chain(messages=messages)
+        ai_response = get_ai_response(llm_chain, user_input)
+        print("ChatGLM3: ", ai_response["text"])
+        messages += [
+            HumanMessage(content=user_input),
+            AIMessage(content=ai_response["text"]),
+        ]
+
+
+if __name__ == "__main__":
+    continuous_conversation()

selected_homework/openai-translator_v4/openai_api_demo/openai_api_request.py

@@ -0,0 +1,99 @@
+"""
+This script is an example of using the OpenAI API to create various interactions with a ChatGLM3 model.
+It includes functions to:
+
+1. Conduct a basic chat session, asking about weather conditions in multiple cities.
+2. Initiate a simple chat in Chinese, asking the model to tell a short story.
+3. Retrieve and print embeddings for a given text input.
+
+Each function demonstrates a different aspect of the API's capabilities, showcasing how to make requests
+and handle responses.
+"""
+
+from openai import OpenAI
+
+base_url = "http://127.0.0.1:8000/v1/"
+client = OpenAI(api_key="EMPTY", base_url=base_url)
+
+
+def function_chat():
+    messages = [{"role": "user", "content": "What's the weather like in San Francisco, Tokyo, and Paris?"}]
+    tools = [
+        {
+            "type": "function",
+            "function": {
+                "name": "get_current_weather",
+                "description": "Get the current weather in a given location",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "location": {
+                            "type": "string",
+                            "description": "The city and state, e.g. San Francisco, CA",
+                        },
+                        "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
+                    },
+                    "required": ["location"],
+                },
+            },
+        }
+    ]
+
+    response = client.chat.completions.create(
+        model="chatglm3-6b",
+        messages=messages,
+        tools=tools,
+        tool_choice="auto",
+    )
+    if response:
+        content = response.choices[0].message.content
+        print(content)
+    else:
+        print("Error:", response.status_code)
+
+
+def simple_chat(use_stream=True):
+    messages = [
+        {
+            "role": "system",
+            "content": "You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's "
+                       "instructions carefully. Respond using markdown.",
+        },
+        {
+            "role": "user",
+            "content": "你好，请你用生动的话语给我讲一个小故事吧"
+        }
+    ]
+    response = client.chat.completions.create(
+        model="chatglm3-6b",
+        messages=messages,
+        stream=use_stream,
+        max_tokens=256,
+        temperature=0.8,
+        presence_penalty=1.1,
+        top_p=0.8)
+    if response:
+        if use_stream:
+            for chunk in response:
+                print(chunk.choices[0].delta.content)
+        else:
+            content = response.choices[0].message.content
+            print(content)
+    else:
+        print("Error:", response.status_code)
+
+
+def embedding():
+    response = client.embeddings.create(
+        model="bge-large-zh-1.5",
+        input=["你好，给我讲一个故事，大概100字"],
+    )
+    embeddings = response.data[0].embedding
+    print("嵌入完成，维度：", len(embeddings))
+
+
+if __name__ == "__main__":
+    simple_chat(use_stream=False)
+    simple_chat(use_stream=True)
+    embedding()
+    function_chat()

selected_homework/openai-translator_v4/openai_api_demo/tools/schema.py

@@ -0,0 +1,34 @@
+
+"""
+Description: You can customize the developed langchain tool overview information here,
+just like the sample code already given in this script.
+"""
+
+
+tool_param_start_with = "```python\ntool_call"
+
+
+""" Fill this dictionary with the mapping from tool class names to tool classes that you defined.  
+Like: 
+         from tools.Calculator import Calculator
+         tool_class = {"Calculator": Calculator, ...}
+
+It is required that your customized tool class must define the format for the langchain tool 
+and implement the parameter verification function in the class:
+       parameter_validation(self, para: str) -> bool
+       
+Tool class definition reference: ChatGLM3/langchain_demo/tools.
+"""
+tool_class = {}
+
+
+""" Describe your tool names and parameters in this dictionary.
+Like:
+  tool_def = [
+    {"name": "Calculator", 
+     "description": "数学计算器，计算数学问题", 
+      "parameters": {"type": "object", "properties": {"symbol": {"description": "要计算的数学公式"}}, "required": []}
+     },...
+  ]
+"""
+tool_def = []

selected_homework/openai-translator_v4/openai_api_demo/utils.py

@@ -0,0 +1,192 @@
+import gc
+import json
+import torch
+from transformers import PreTrainedModel, PreTrainedTokenizer
+from transformers.generation.logits_process import LogitsProcessor
+from typing import Union, Tuple
+
+
+class InvalidScoreLogitsProcessor(LogitsProcessor):
+    def __call__(
+            self, input_ids: torch.LongTensor, scores: torch.FloatTensor
+    ) -> torch.FloatTensor:
+        if torch.isnan(scores).any() or torch.isinf(scores).any():
+            scores.zero_()
+            scores[..., 5] = 5e4
+        return scores
+
+
+def process_response(output: str, use_tool: bool = False) -> Union[str, dict]:
+    content = ""
+    for response in output.split("<|assistant|>"):
+        metadata, content = response.split("\n", maxsplit=1)
+        if not metadata.strip():
+            content = content.strip()
+            content = content.replace("[[训练时间]]", "2023年")
+        else:
+            if use_tool:
+                content = "\n".join(content.split("\n")[1:-1])
+
+                def tool_call(**kwargs):
+                    return kwargs
+
+                parameters = eval(content)
+                content = {
+                    "name": metadata.strip(),
+                    "arguments": json.dumps(parameters, ensure_ascii=False)
+                }
+            else:
+                content = {
+                    "name": metadata.strip(),
+                    "content": content
+                }
+    return content
+
+
+@torch.inference_mode()
+def generate_stream_chatglm3(model: PreTrainedModel, tokenizer: PreTrainedTokenizer, params: dict):
+    messages = params["messages"]
+    tools = params["tools"]
+    temperature = float(params.get("temperature", 1.0))
+    repetition_penalty = float(params.get("repetition_penalty", 1.0))
+    top_p = float(params.get("top_p", 1.0))
+    max_new_tokens = int(params.get("max_tokens", 256))
+    echo = params.get("echo", True)
+    messages = process_chatglm_messages(messages, tools=tools)
+    query, role = messages[-1]["content"], messages[-1]["role"]
+
+    inputs = tokenizer.build_chat_input(query, history=messages[:-1], role=role)
+    inputs = inputs.to(model.device)
+    input_echo_len = len(inputs["input_ids"][0])
+
+    if input_echo_len >= model.config.seq_length:
+        print(f"Input length larger than {model.config.seq_length}")
+
+    eos_token_id = [
+        tokenizer.eos_token_id,
+        tokenizer.get_command("<|user|>"),
+        tokenizer.get_command("<|observation|>")
+    ]
+
+    gen_kwargs = {
+        "max_new_tokens": max_new_tokens,
+        "do_sample": True if temperature > 1e-5 else False,
+        "top_p": top_p,
+        "repetition_penalty": repetition_penalty,
+        "logits_processor": [InvalidScoreLogitsProcessor()],
+    }
+    if temperature > 1e-5:
+        gen_kwargs["temperature"] = temperature
+
+    total_len = 0
+    for total_ids in model.stream_generate(**inputs, eos_token_id=eos_token_id, **gen_kwargs):
+        total_ids = total_ids.tolist()[0]
+        total_len = len(total_ids)
+        if echo:
+            output_ids = total_ids[:-1]
+        else:
+            output_ids = total_ids[input_echo_len:-1]
+
+        response = tokenizer.decode(output_ids)
+        if response and response[-1] != "�":
+            response, stop_found = apply_stopping_strings(response, ["<|observation|>"])
+
+            yield {
+                "text": response,
+                "usage": {
+                    "prompt_tokens": input_echo_len,
+                    "completion_tokens": total_len - input_echo_len,
+                    "total_tokens": total_len,
+                },
+                "finish_reason": "function_call" if stop_found else None,
+            }
+
+            if stop_found:
+                break
+
+    # Only last stream result contains finish_reason, we set finish_reason as stop
+    ret = {
+        "text": response,
+        "usage": {
+            "prompt_tokens": input_echo_len,
+            "completion_tokens": total_len - input_echo_len,
+            "total_tokens": total_len,
+        },
+        "finish_reason": "stop",
+    }
+    yield ret
+
+    gc.collect()
+    torch.cuda.empty_cache()
+
+
+def process_chatglm_messages(messages, tools=None):
+    _messages = messages
+    messages = []
+    msg_has_sys = False
+    if tools:
+        messages.append(
+            {
+                "role": "system",
+                "content": "Answer the following questions as best as you can. You have access to the following tools:",
+                "tools": tools
+            }
+        )
+        msg_has_sys = True
+
+    for m in _messages:
+        role, content, func_call = m.role, m.content, m.function_call
+        if role == "function":
+            messages.append(
+                {
+                    "role": "observation",
+                    "content": content
+                }
+            )
+
+        elif role == "assistant" and func_call is not None:
+            for response in content.split("<|assistant|>"):
+                metadata, sub_content = response.split("\n", maxsplit=1)
+                messages.append(
+                    {
+                        "role": role,
+                        "metadata": metadata,
+                        "content": sub_content.strip()
+                    }
+                )
+        else:
+            if role == "system" and msg_has_sys:
+                msg_has_sys = False
+                continue
+            messages.append({"role": role, "content": content})
+    return messages
+
+
+def generate_chatglm3(model: PreTrainedModel, tokenizer: PreTrainedTokenizer, params: dict):
+    for response in generate_stream_chatglm3(model, tokenizer, params):
+        pass
+    return response
+
+
+def apply_stopping_strings(reply, stop_strings) -> Tuple[str, bool]:
+    stop_found = False
+    for string in stop_strings:
+        idx = reply.find(string)
+        if idx != -1:
+            reply = reply[:idx]
+            stop_found = True
+            break
+
+    if not stop_found:
+        # If something like "\nYo" is generated just before "\nYou: is completed, trim it
+        for string in stop_strings:
+            for j in range(len(string) - 1, 0, -1):
+                if reply[-j:] == string[:j]:
+                    reply = reply[:-j]
+                    break
+            else:
+                continue
+
+            break
+
+    return reply, stop_found

selected_homework/openai-translator_v4/openai_api_demo/zhipu_api_request.py

@@ -0,0 +1,100 @@
+"""
+This script is an example of using the Zhipu API to create various interactions with a ChatGLM3 model. It includes
+functions to:
+
+1. Conduct a basic chat session, asking about weather conditions in multiple cities.
+2. Initiate a simple chat in Chinese, asking the model to tell a short story.
+3. Retrieve and print embeddings for a given text input.
+Each function demonstrates a different aspect of the API's capabilities,
+showcasing how to make requests and handle responses.
+
+Note: Make sure your Zhipu API key is set as an environment
+variable formate as xxx.xxx (just for check, not need a real key).
+"""
+
+from zhipuai import ZhipuAI
+
+base_url = "http://127.0.0.1:8000/v1/"
+client = ZhipuAI(api_key="EMP.TY", base_url=base_url)
+
+
+def function_chat():
+    messages = [{"role": "user", "content": "What's the weather like in San Francisco, Tokyo, and Paris?"}]
+    tools = [
+        {
+            "type": "function",
+            "function": {
+                "name": "get_current_weather",
+                "description": "Get the current weather in a given location",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "location": {
+                            "type": "string",
+                            "description": "The city and state, e.g. San Francisco, CA",
+                        },
+                        "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
+                    },
+                    "required": ["location"],
+                },
+            },
+        }
+    ]
+
+    response = client.chat.completions.create(
+        model="chatglm3_6b",
+        messages=messages,
+        tools=tools,
+        tool_choice="auto",
+    )
+    if response:
+        content = response.choices[0].message.content
+        print(content)
+    else:
+        print("Error:", response.status_code)
+
+
+def simple_chat(use_stream=True):
+    messages = [
+        {
+            "role": "system",
+            "content": "You are ChatGLM3, a large language model trained by Zhipu.AI. Follow "
+                       "the user's instructions carefully. Respond using markdown.",
+        },
+        {
+            "role": "user",
+            "content": "你好，请你介绍一下chatglm3-6b这个模型"
+        }
+    ]
+    response = client.chat.completions.create(
+        model="chatglm3_",
+        messages=messages,
+        stream=use_stream,
+        max_tokens=256,
+        temperature=0.8,
+        top_p=0.8)
+    if response:
+        if use_stream:
+            for chunk in response:
+                print(chunk.choices[0].delta.content)
+        else:
+            content = response.choices[0].message.content
+            print(content)
+    else:
+        print("Error:", response.status_code)
+
+
+def embedding():
+    response = client.embeddings.create(
+        model="bge-large-zh-1.5",
+        input=["ChatGLM3-6B 是一个大型的中英双语模型。"],
+    )
+    embeddings = response.data[0].embedding
+    print("嵌入完成，维度：", len(embeddings))
+
+
+if __name__ == "__main__":
+    simple_chat(use_stream=False)
+    simple_chat(use_stream=True)
+    embedding()
+    function_chat()