-
Notifications
You must be signed in to change notification settings - Fork 4
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
11 changed files
with
169 additions
and
17 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,3 @@ | ||
| { | ||
| "makefile.configureOnOpen": false | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,91 @@ | ||
| package ai | ||
|
|
||
| import ( | ||
| "context" | ||
|
|
||
| "github.com/redish101/reblog/internal/core" | ||
| "github.com/redish101/reblog/internal/log" | ||
| "github.com/yankeguo/zhipu" | ||
| ) | ||
|
|
||
| const prompt = ` | ||
| 【角色定义】 | ||
| 你是一位资深技术文档工程师，专注于为复杂的技术博客生成精准摘要。请严格基于原文内容生成专业摘要，保持技术严谨性。 | ||
| 【核心要求】 | ||
| 内容忠实度： | ||
| 完全基于原文事实，禁止添加任何外部知识或个人见解 | ||
| 保留关键专业术语（如Transformer、CUDA核心、RAG架构等） | ||
| 准确呈现技术方案的核心创新点 | ||
| 结构规范： | ||
| 首段：总览句（20字内）点明文章核心价值 | ||
| 主体内容包含： | ||
| • 待解决的技术问题 | ||
| • 采用的解决方案架构 | ||
| • 关键算法/工具链（如使用PyTorch 2.1+TorchDynamo） | ||
| • 实验结果（量化指标需精确，如"QPS提升37.2%"） | ||
| • 实际应用场景 | ||
| 语言风格： | ||
| 专业而不晦涩，避免过度学术化表达 | ||
| 使用技术领域标准表达（如"微调"而非"调整"，"消融实验"而非"对比测试"） | ||
| 保持客观中立，区分作者主张和客观事实 | ||
| 格式规范： | ||
| 纯文本输出，禁用任何Markdown格式、禁止换行 | ||
| 中文字数严格控制在280-320字 | ||
| 段落间自然过渡，保持技术逻辑连贯性 | ||
| 不要出现任何文章内没有出现的内容，包括类似“实验结果：无具体量化指标“ | ||
| 【典型示例】 | ||
| [合格摘要] | ||
| "本文探讨PyTorch 2.1的编译优化技术。针对动态图执行效率问题，提出基于TorchDynamo的图捕获方案，配合TensorRT的后端优化，实现训练速度提升。核心创新包括：1）通过字节码分析实现Python动态特性兼容 2）选择性子图JIT编译策略 3）与CUDA 11.7的深度适配。在ResNet-152训练任务中，相较原生PyTorch获得37.2%的加速比，内存占用降低19%。该方案适用于需要兼顾开发灵活性和执行效率的DL训练场景。" | ||
| [不合格摘要] | ||
| "这篇文章讲PyTorch的优化方法，作者搞了个新的编译方案，速度提升明显，这对AI训练很有帮助。"（缺乏技术细节和量化指标） | ||
| **GN(Generate Ninja) 快速入门指南**\n\n本文介绍了 Google 开发的元构建系统 GN 的使用方法。GN 用于生成 build.ninja 文件，类似于 cmake。文章通过构建一个基于 glfw + glad 的 C++ 程序，详细讲解了 GN 的安装、配置编译工具链、声明构建配置以及构建过程。\n\n**待解决的技术问题**：GN 的使用方法和配置流程。\n\n**解决方案架构**：通过构建一个简单的 C++ 程序，展示 GN 的使用步骤。\n\n**关键算法/工具链**：GN、ninja、glfw、glad。\n\n**实验结果**：无具体量化指标。\n\n**实际应用场景**：GN 适用于规模较大的项目中，更为清晰的描述构建过程。（使用了markdown语法以及换行） | ||
| 你需要牢记： 禁止使用任何markdown语法、禁止换行！ | ||
| 你需要牢记： 禁止使用任何markdown语法、禁止换行！ | ||
| 你需要牢记： 禁止使用任何markdown语法、禁止换行！ | ||
| 请严格遵循以上规范生成专业技术摘要，现在请处理以下技术博客内容： | ||
| ` | ||
| func Summary(app *core.App, text string) string { | ||
| log.Debug("[AI] 正生成摘要") | ||
| client, err := zhipu.NewClient(zhipu.WithAPIKey(app.Config().Ai.ApiKey)) | ||
|
|
||
| service := client.ChatCompletion("glm-4-flash"). | ||
| AddMessage(zhipu.ChatCompletionMessage{ | ||
| Role: "user", | ||
| Content: prompt + text, | ||
|
|
||
| }) | ||
|
|
||
| res, err := service.Do(context.Background()) | ||
|
|
||
| if err != nil { | ||
| zhipu.GetAPIErrorCode(err) | ||
| log.Warnf("[AI] 摘要生成失败： %s", err) | ||
| } else { | ||
| log.Debugf("[AI] 成功生成摘要： %s", res.Choices[0].Message.Content) | ||
| return res.Choices[0].Message.Content | ||
| } | ||
|
|
||
| return "生成摘要失败" | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -12,6 +12,9 @@ rss: | |
| limit: 10 | ||
| content_link_format: "%s/article/%s" | ||
|
|
||
| ai: | ||
| apikey: your_api_key | ||
|
|
||
| plugins: | ||
| # - example/plugin | ||
|
|
||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters