初始化

.gitignore

+.vscode
+.idea
+.DS_Store
+__pycache__
+*.pyc
+*.zip
+*.out
+bin/
+debug/
+release/
\ No newline at end of file

README.md

 # skill_tree_cuda
 
-CUDA 技能树
+`CUAD入门技能树`是[技能森林](https://gitcode.net/csdn/skill_tree)的一部分。
\ No newline at end of file
+
+## 编辑环境初始化
+
+```
+pip install -r requirements.txt
+```
+
+## 目录结构说明
+技能树编辑仓库的 data 目录是主要的编辑目录，目录的结构是固定的
+
+* 技能树`骨架文件`：
+    * 位置：`data/tree.json`
+    * 说明：该文件是执行 `python main.py` 生成的，请勿人工编辑
+* 技能树`根节点`配置文件：
+    * 位置：`data/config.json`
+    * 说明：可编辑配置关键词等字段，其中 `node_id` 字段是生成的，请勿编辑
+* 技能树`难度节点`：
+    * 位置：`data/xxx`，例如: `data/1.CUAD入门初阶`
+    * 说明：
+        * 每个技能树有 3 个等级，目录前的序号是必要的，用来保持文件夹目录的顺序
+        * 每个目录下有一个 `config.json` 可配置关键词信息，其中 `node_id` 字段是生成的，请勿编辑
+* 技能树`章节点`：
+    * 位置：`data/xxx/xxx`，例如：`data/1.CUAD入门初阶/1.1.GPU架构及异构计算`
+    * 说明：
+        * 每个技能树的每个难度等级有 n 个章节，目录前的序号是必要的，用来保持文件夹目录的顺序
+        * 每个目录下有一个 `config.json` 可配置关键词信息，其中 `node_id` 字段是生成的，请勿编辑
+* 技能树`知识节点`：
+    * 位置：`data/xxx/xxx`，例如：`data/1.CUAD入门初阶/1.1.GPU架构及异构计算`
+    * 说明：
+        * 每个技能树的每章有 n 个知识节点，目录前的序号是必要的，用来保持文件夹目录的顺序
+        * 每个目录下有一个 `config.json`
+            * 其中 `node_id` 字段是生成的，请勿编辑
+            * 其中 `keywords` 可配置关键字字段
+            * 其中 `children` 可配置该`知识节点`下的子树结构信息，参考后面描述
+            * 其中 `export` 可配置该`知识节点`下的导出习题信息，参考后面描述
+
+## `知识节点` 子树信息结构
+
+例如 `data/1.CUAD入门初阶/1.1.GPU架构及异构计算/介绍GPU架构以及异构计算的基本原理/config.json` 里配置对该知识节点子树信息结构，这个配置是可选的：
+```json
+{
+    // ...
+
+    "children": [
+    {
+        "XX开发入门": {
+          "keywords": [
+            "XX开发",
+          ],
+          "children": [],
+          "keywords_must": [
+            "XX"
+          ],
+          "keywords_forbid": []
+        }
+    }
+  ],
+}
+```
+
+## `知识节点` 的导出习题编辑
+
+例如 `data/1.CUAD入门初阶/1.1.GPU架构及异构计算/介绍GPU架构以及异构计算的基本原理/config.json` 里配置对该知识节点导出的习题
+
+```json
+{
+    // ...
+    "export": [
+        "helloworld.json"
+    ]
+}
+```
+
+helloworld.json 的格式如下：
+```bash
+{
+  "type": "code_options",
+  "author": "xxx",
+  "source": "helloworld.md",
+  "notebook_enable": false,
+  "exercise_id": "xxx"
+}
+```
+
+其中 
+* "type": "code_options" 表示是一个选择题
+* "author" 可以放作者的 CSDN id，
+* "source" 指向了习题 MarkDown文件
+* "notebook_enable" 目前都是false
+* "exercise_id" 是工具生成的，不填
+
+
+习题格式模版如下：
+
+````mardown
+# {标题}
+
+{习题描述}
+
+以下关于上述游戏代码说法[正确/错误]的是？
+
+## 答案
+
+{目标选项}
+
+## 选项
+
+### A
+
+{混淆选项1}
+
+### B
+
+{混淆选项2}
+
+### C
+
+{混淆选项3}
+
+````
+
+## 技能树合成
+
+在根目录下执行 `python main.py` 会合成技能树文件，合成的技能树文件: `data/tree.json`
+* 合成过程中，会自动检查每个目录下 `config.json` 里的 `node_id` 是否存在，不存在则生成
+* 合成过程中，会自动检查每个知识点目录下 `config.json` 里的 `export` 里导出的习题配置，检查是否存在`exercise_id` 字段，如果不存在则生成
+* 在 节 目录下根据需要，可以添加一些子目录用来测试代码。
+* 开始游戏入门技能树构建之旅，GoodLuck! 
+
+
+## FAQ
+
+**难度目录是固定的么？**
+
+1. data/xxx 目录下的子目录是固定的初/中/高三个难度等级目录
+
+**如何增加章目录？**
+
+1. 在VSCode里打开项目仓库
+2. 在对应的难度等级目录新建章目录，例如在 data/1.xxx初阶/ 下新建章文件夹，data/1.xxx初阶/1.yyy
+3. 在项目根目录下执行 python main.py 脚本，会自动生成章的配置文件 data/1.xxx初阶/1.yyy/config.json
+
+**如何增加节目录?**:
+1. 直接在VSCode里创建文件夹，例如 "data/1.xxx初阶/1.yyy/2.zzz"
+2. 项目根目录下执行 python main.py 会自动为新增节创建配置文件 data/1.xxx初阶/1.yyy/2.zzz/config.json
+
+**如何在节下新增一个习题**:
+3. 在"data/1.xxx初阶/1.yyy/2.zzz" 目录下添加一个 markdown 文件编辑，例如 yyy.md，按照习题markdown格式编辑习题。
+4. md编辑完后，可以再次执行  python main.py 会自动生成同名的 yyy.json，并将 yyy.json 添加到config.json 的export数组里。
+5. yyy.json里的author信息放作者 CSDN ID。
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/1.介绍GPU架构以及异构计算的基本原理/config.json

+{
+  "node_id": "cuda-bdd8df6c59d0460bbf30d3a4a6203b06",
+  "keywords": [],
+  "children": [],
+  "export": [
+    "helloworld.json"
+  ],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/1.介绍GPU架构以及异构计算的基本原理/helloworld.json

+{
+  "type": "code_options",
+  "author": null,
+  "source": "helloworld.md",
+  "notebook_enable": false,
+  "exercise_id": "02176393bfff4c7ea60d9f85fd095224"
+}
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/1.介绍GPU架构以及异构计算的基本原理/helloworld.md

+# {标题}
+
+{习题描述}
+
+以下关于上述游戏代码说法[正确/错误]的是？
+
+## 答案
+
+{目标选项}
+
+## 选项
+
+### A
+
+{混淆选项1}
+
+### B
+
+{混淆选项2}
+
+### C
+
+{混淆选项3}
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/2.介绍GPU硬件平台/config.json

+{
+  "node_id": "cuda-8acef8aa3f7b479d90b7eaf77ff752eb",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/3.环境安装配置/config.json

+{
+  "node_id": "cuda-f15df586747c4e648b6c6824c6b9b3e1",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/1.GPU架构及异构计算/config.json

+{
+  "node_id": "cuda-8cc14ba50dbb4b00ba62a8070d2b599c",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/1.CUDA程序的编译/config.json

+{
+  "node_id": "cuda-25c9eacb75e64d00bff3d14ffdec7ea7",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/2.GPU线程的调用/config.json

+{
+  "node_id": "cuda-4756f243773643fd8064aa5b4ffdb789",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/3.GPU和CPU的通讯/config.json

+{
+  "node_id": "cuda-bf38379916ce44978c6bfa3ef3487c71",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/4.使用多个线程的核函数/config.json

+{
+  "node_id": "cuda-8c79c6a5e3ca441a9cee430f312407bf",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/5.使用线程索引/config.json

+{
+  "node_id": "cuda-b7fb5b2b91234dd89968918460ae506f",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/6.多维网络/config.json

+{
+  "node_id": "cuda-ec664e92f8e2410b88226408d9bb9a9f",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/7.网格与线程块/config.json

+{
+  "node_id": "cuda-70b301e021ef435f92c0f07b22adaa09",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/2.初识CUDA/config.json

+{
+  "node_id": "cuda-8b007eb550f842058b4bc0b2bc457c7e",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/1.设备初始化/config.json

+{
+  "node_id": "cuda-5f99fc5469cf4907ba3ebe615287a6e9",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/2.GPU的存储单元/config.json

+{
+  "node_id": "cuda-f336472c4c244a68b4a2ee80dac8fdd0",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/3.GPU存储单元的分配与释放/config.json

+{
+  "node_id": "cuda-3443edc2ea5140b0a39f690382bf91e2",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/4.数据的传输/config.json

+{
+  "node_id": "cuda-2244dd4b61cc478094ba7013770f1f29",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/5.数据与线程之间的对应关系/config.json

+{
+  "node_id": "cuda-3fa4ca2a53d74b9d92a11ba6a7f23306",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/3.CUDA存储单元的使用/config.json

+{
+  "node_id": "cuda-d7adf776b61b4829aea19181b8bd188b",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/4.多种CUDA存储单元详解/1.CUDA中的存储单元种类/config.json

+{
+  "node_id": "cuda-73f87372510e4546a282aea26e371e53",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/4.多种CUDA存储单元详解/2.CUDA中的各种存储单元的使用方法/config.json

+{
+  "node_id": "cuda-3347d61cb34745a4adbe216f6eca305c",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/4.多种CUDA存储单元详解/3.CUDA中的各种存储单元的适用条件/config.json

+{
+  "node_id": "cuda-73179a76ecd34f5dbf8e53b3a1e84228",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/4.多种CUDA存储单元详解/config.json

+{
+  "node_id": "cuda-95a89e2e13dc4c5c94c41eeb648107a5",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/5.利用共享存储单元优化应用/1.共享存储单元详解/config.json

+{
+  "node_id": "cuda-8965afc1396c4aad8fa3eaa203b6e3ac",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/5.利用共享存储单元优化应用/2.共享内存的Bank conflict/config.json

+{
+  "node_id": "cuda-3c183e532dcd44d88311a778f6958916",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/5.利用共享存储单元优化应用/3.利用共享存储单元进行矩阵转置和矩阵乘积/config.json

+{
+  "node_id": "cuda-6560fcc5a5c9465b84aa0f15b9576b79",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/5.利用共享存储单元优化应用/config.json

+{
+  "node_id": "cuda-288bccf96d9645e294723329219375eb",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/6.统一内存/1.统一内存的基本概念和使用/config.json

+{
+  "node_id": "cuda-e2ec9573577d42fdbed66392ffc835d3",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/6.统一内存/config.json

+{
+  "node_id": "cuda-545f2bfb1f8b4102ad4b50dd376ebc23",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/1.CUDA入门初阶/config.json

+{
+  "node_id": "cuda-0f5c808b1251449fb296063ff1d324d1",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/1.CUA错误检测与事件/1.CUDA应用程序运行时的错误检测/config.json

+{
+  "node_id": "cuda-634b3d9492044bafb7a089431c878879",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/1.CUA错误检测与事件/2.CUDA中的事件/config.json

+{
+  "node_id": "cuda-5a542af3254b49a8ab5266364d421e47",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/1.CUA错误检测与事件/config.json

+{
+  "node_id": "cuda-346e5f64d0904f63baccd001ef8b676b",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/2.原子操作/1.CUDA中的原子操作/config.json

+{
+  "node_id": "cuda-10a7cead25f14eeaaf0730ff2468cb90",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/2.原子操作/2.原子操作的适用场景/config.json

+{
+  "node_id": "cuda-9e74b1fb163e46bb8cc2e72b3c9990a9",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/2.原子操作/config.json

+{
+  "node_id": "cuda-6f61fc7ce2ef413886354ac2dfab2b27",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/3.CUDA stream/1.CUDA流的基本概念/config.json

+{
+  "node_id": "cuda-a95b22ce131e42979daabc1f2ec82886",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/3.CUDA stream/2.默认流与非默认流/config.json

+{
+  "node_id": "cuda-93c84af94c154430b93094befb70f5c8",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/3.CUDA stream/3.利用CUDA流重叠计算和数据传输/config.json

+{
+  "node_id": "cuda-24b5313252644192b40894d9d677a40f",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/3.CUDA stream/config.json

+{
+  "node_id": "cuda-5d11b746f9f34965a77bb8e6777c3ba9",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/2.CUDA入门中阶/config.json

+{
+  "node_id": "cuda-c77a77ca93f546dda09ec621380b04f1",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/1.CUDA 调试分析/1.利用Nsight等分析工具对程序性能进行分析/config.json

+{
+  "node_id": "cuda-1698eb54bf894cf48d27b11baaf03916",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/1.CUDA 调试分析/2.根据实际硬件调整程序/config.json

+{
+  "node_id": "cuda-e1ca07a17c9443b5af7c0d22b4bff705",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/1.CUDA 调试分析/config.json

+{
+  "node_id": "cuda-60427e3bc2ea4308a250e2716efe4ed8",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/2.CUDA 优化/1.存储优化/config.json

+{
+  "node_id": "cuda-5f1d7d26ebff499abb0ee0aeb66da328",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/2.CUDA 优化/2.执行设置优化/config.json

+{
+  "node_id": "cuda-8e9c726f99b84fa685d6e5cce061fd9d",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/2.CUDA 优化/3.指令级优化/config.json

+{
+  "node_id": "cuda-d1a8d39fbc4247f394c9979f1cf2cac7",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/2.CUDA 优化/4.控制流优化/config.json

+{
+  "node_id": "cuda-18f6a05677194f78bf61c456dddf0905",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/2.CUDA 优化/config.json

+{
+  "node_id": "cuda-fb7d53daa17e4db09234d87558508e81",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/1.cuBLAS/config.json

+{
+  "node_id": "cuda-19170998ef3c4b9a9ab808a989ed29f3",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/2.cuFFT/config.json

+{
+  "node_id": "cuda-9c80239f362c4b28aa40bf55a5f2de81",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/3.cuRAND/config.json

+{
+  "node_id": "cuda-e1f4431b5a10485985345315c55762bc",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/4.cuSPARSE/config.json

+{
+  "node_id": "cuda-88eb3bc83d4c425583fea7bf547483a6",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/5.cuDNN/config.json

+{
+  "node_id": "cuda-083766b88b3647679d2067ec60ccfcdd",
+  "keywords": [],
+  "children": [],
+  "export": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/3.CUDA 加速库/config.json

+{
+  "node_id": "cuda-923b4ee7f3134962933344a5d7b8ff4c",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/3.CUDA入门高阶/config.json

+{
+  "node_id": "cuda-ade73748ee304a6997e6a24cbc0d69ac",
+  "keywords": [],
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/config.json

+{
+  "tree_name": "cuda",
+  "keywords": [],
+  "node_id": "cuda-a39f9a57edfe4dd48fce8d68e950db24",
+  "keywords_must": [],
+  "keywords_forbid": []
+}
\ No newline at end of file

data/tree.json

+{
+  "cuda": {
+    "node_id": "cuda-a39f9a57edfe4dd48fce8d68e950db24",
+    "keywords": [],
+    "children": [
+      {
+        "CUDA入门初阶": {
+          "node_id": "cuda-0f5c808b1251449fb296063ff1d324d1",
+          "keywords": [],
+          "children": [
+            {
+              "GPU架构及异构计算": {
+                "node_id": "cuda-8cc14ba50dbb4b00ba62a8070d2b599c",
+                "keywords": [],
+                "children": [
+                  {
+                    "介绍GPU架构以及异构计算的基本原理": {
+                      "node_id": "cuda-bdd8df6c59d0460bbf30d3a4a6203b06",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "介绍GPU硬件平台": {
+                      "node_id": "cuda-8acef8aa3f7b479d90b7eaf77ff752eb",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "环境安装配置": {
+                      "node_id": "cuda-f15df586747c4e648b6c6824c6b9b3e1",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "初识CUDA": {
+                "node_id": "cuda-8b007eb550f842058b4bc0b2bc457c7e",
+                "keywords": [],
+                "children": [
+                  {
+                    "CUDA程序的编译": {
+                      "node_id": "cuda-25c9eacb75e64d00bff3d14ffdec7ea7",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "GPU线程的调用": {
+                      "node_id": "cuda-4756f243773643fd8064aa5b4ffdb789",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "GPU和CPU的通讯": {
+                      "node_id": "cuda-bf38379916ce44978c6bfa3ef3487c71",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "使用多个线程的核函数": {
+                      "node_id": "cuda-8c79c6a5e3ca441a9cee430f312407bf",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "使用线程索引": {
+                      "node_id": "cuda-b7fb5b2b91234dd89968918460ae506f",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "多维网络": {
+                      "node_id": "cuda-ec664e92f8e2410b88226408d9bb9a9f",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "网格与线程块": {
+                      "node_id": "cuda-70b301e021ef435f92c0f07b22adaa09",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "CUDA存储单元的使用": {
+                "node_id": "cuda-d7adf776b61b4829aea19181b8bd188b",
+                "keywords": [],
+                "children": [
+                  {
+                    "设备初始化": {
+                      "node_id": "cuda-5f99fc5469cf4907ba3ebe615287a6e9",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "GPU的存储单元": {
+                      "node_id": "cuda-f336472c4c244a68b4a2ee80dac8fdd0",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "GPU存储单元的分配与释放": {
+                      "node_id": "cuda-3443edc2ea5140b0a39f690382bf91e2",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "数据的传输": {
+                      "node_id": "cuda-2244dd4b61cc478094ba7013770f1f29",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "数据与线程之间的对应关系": {
+                      "node_id": "cuda-3fa4ca2a53d74b9d92a11ba6a7f23306",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "多种CUDA存储单元详解": {
+                "node_id": "cuda-95a89e2e13dc4c5c94c41eeb648107a5",
+                "keywords": [],
+                "children": [
+                  {
+                    "CUDA中的存储单元种类": {
+                      "node_id": "cuda-73f87372510e4546a282aea26e371e53",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "CUDA中的各种存储单元的使用方法": {
+                      "node_id": "cuda-3347d61cb34745a4adbe216f6eca305c",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "CUDA中的各种存储单元的适用条件": {
+                      "node_id": "cuda-73179a76ecd34f5dbf8e53b3a1e84228",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "利用共享存储单元优化应用": {
+                "node_id": "cuda-288bccf96d9645e294723329219375eb",
+                "keywords": [],
+                "children": [
+                  {
+                    "共享存储单元详解": {
+                      "node_id": "cuda-8965afc1396c4aad8fa3eaa203b6e3ac",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "共享内存的Bank conflict": {
+                      "node_id": "cuda-3c183e532dcd44d88311a778f6958916",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "利用共享存储单元进行矩阵转置和矩阵乘积": {
+                      "node_id": "cuda-6560fcc5a5c9465b84aa0f15b9576b79",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "统一内存": {
+                "node_id": "cuda-545f2bfb1f8b4102ad4b50dd376ebc23",
+                "keywords": [],
+                "children": [
+                  {
+                    "统一内存的基本概念和使用": {
+                      "node_id": "cuda-e2ec9573577d42fdbed66392ffc835d3",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            }
+          ],
+          "keywords_must": [],
+          "keywords_forbid": []
+        }
+      },
+      {
+        "CUDA入门中阶": {
+          "node_id": "cuda-c77a77ca93f546dda09ec621380b04f1",
+          "keywords": [],
+          "children": [
+            {
+              "CUA错误检测与事件": {
+                "node_id": "cuda-346e5f64d0904f63baccd001ef8b676b",
+                "keywords": [],
+                "children": [
+                  {
+                    "CUDA应用程序运行时的错误检测": {
+                      "node_id": "cuda-634b3d9492044bafb7a089431c878879",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "CUDA中的事件": {
+                      "node_id": "cuda-5a542af3254b49a8ab5266364d421e47",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "原子操作": {
+                "node_id": "cuda-6f61fc7ce2ef413886354ac2dfab2b27",
+                "keywords": [],
+                "children": [
+                  {
+                    "CUDA中的原子操作": {
+                      "node_id": "cuda-10a7cead25f14eeaaf0730ff2468cb90",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "原子操作的适用场景": {
+                      "node_id": "cuda-9e74b1fb163e46bb8cc2e72b3c9990a9",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "CUDA stream": {
+                "node_id": "cuda-5d11b746f9f34965a77bb8e6777c3ba9",
+                "keywords": [],
+                "children": [
+                  {
+                    "CUDA流的基本概念": {
+                      "node_id": "cuda-a95b22ce131e42979daabc1f2ec82886",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "默认流与非默认流": {
+                      "node_id": "cuda-93c84af94c154430b93094befb70f5c8",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "利用CUDA流重叠计算和数据传输": {
+                      "node_id": "cuda-24b5313252644192b40894d9d677a40f",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            }
+          ],
+          "keywords_must": [],
+          "keywords_forbid": []
+        }
+      },
+      {
+        "CUDA入门高阶": {
+          "node_id": "cuda-ade73748ee304a6997e6a24cbc0d69ac",
+          "keywords": [],
+          "children": [
+            {
+              "CUDA 调试分析": {
+                "node_id": "cuda-60427e3bc2ea4308a250e2716efe4ed8",
+                "keywords": [],
+                "children": [
+                  {
+                    "利用Nsight等分析工具对程序性能进行分析": {
+                      "node_id": "cuda-1698eb54bf894cf48d27b11baaf03916",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "根据实际硬件调整程序": {
+                      "node_id": "cuda-e1ca07a17c9443b5af7c0d22b4bff705",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "CUDA 优化": {
+                "node_id": "cuda-fb7d53daa17e4db09234d87558508e81",
+                "keywords": [],
+                "children": [
+                  {
+                    "存储优化": {
+                      "node_id": "cuda-5f1d7d26ebff499abb0ee0aeb66da328",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "执行设置优化": {
+                      "node_id": "cuda-8e9c726f99b84fa685d6e5cce061fd9d",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "指令级优化": {
+                      "node_id": "cuda-d1a8d39fbc4247f394c9979f1cf2cac7",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "控制流优化": {
+                      "node_id": "cuda-18f6a05677194f78bf61c456dddf0905",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            },
+            {
+              "CUDA 加速库": {
+                "node_id": "cuda-923b4ee7f3134962933344a5d7b8ff4c",
+                "keywords": [],
+                "children": [
+                  {
+                    "cuBLAS": {
+                      "node_id": "cuda-19170998ef3c4b9a9ab808a989ed29f3",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "cuFFT": {
+                      "node_id": "cuda-9c80239f362c4b28aa40bf55a5f2de81",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "cuRAND": {
+                      "node_id": "cuda-e1f4431b5a10485985345315c55762bc",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "cuSPARSE": {
+                      "node_id": "cuda-88eb3bc83d4c425583fea7bf547483a6",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  },
+                  {
+                    "cuDNN": {
+                      "node_id": "cuda-083766b88b3647679d2067ec60ccfcdd",
+                      "keywords": [],
+                      "children": [],
+                      "keywords_must": [],
+                      "keywords_forbid": []
+                    }
+                  }
+                ],
+                "keywords_must": [],
+                "keywords_forbid": []
+              }
+            }
+          ],
+          "keywords_must": [],
+          "keywords_forbid": []
+        }
+      }
+    ],
+    "keywords_must": [],
+    "keywords_forbid": []
+  }
+}
\ No newline at end of file

data/tree_simple.json

+{
+    "level":{
+        "level_0": [
+            "GPU架构及异构计算",
+            "初识CUDA",
+            "CUDA存储单元的使用",
+            "多种CUDA存储单元详解",
+            "利用共享存储单元优化应用",
+            "统一内存"
+        ],
+        "level_1":[
+            "CUA错误检测与事件",
+            "原子操作",
+            "CUDA stream"
+        ],
+        "level_2": [
+            "CUDA 调试分析",
+            "CUDA 优化",
+            "CUDA 加速库"
+        ]
+    },
+    "tree": {
+        "GPU架构及异构计算":[
+            "介绍GPU架构以及异构计算的基本原理",
+            "介绍GPU硬件平台",
+            "环境安装配置"
+        ],
+        "初识CUDA": [
+            "CUDA程序的编译",
+            "GPU线程的调用",
+            "GPU和CPU的通讯",
+            "使用多个线程的核函数",
+            "使用线程索引",
+            "多维网络",
+            "网格与线程块"
+        ],
+        "CUDA存储单元的使用": [
+            "设备初始化",
+            "GPU的存储单元",
+            "GPU存储单元的分配与释放",
+            "数据的传输",
+            "数据与线程之间的对应关系"
+        ],
+        "多种CUDA存储单元详解": [
+            "CUDA中的存储单元种类",
+            "CUDA中的各种存储单元的使用方法",
+            "CUDA中的各种存储单元的适用条件"
+        ],
+        "利用共享存储单元优化应用":  [
+            "共享存储单元详解",
+            "共享内存的Bank conflict",
+            "利用共享存储单元进行矩阵转置和矩阵乘积"
+        ],
+        "统一内存": [
+            "统一内存的基本概念和使用"
+        ],
+        "CUA错误检测与事件": [
+            "CUDA应用程序运行时的错误检测",
+            "CUDA中的事件"
+        ],
+        "原子操作": [
+            "CUDA中的原子操作",
+            "原子操作的适用场景"
+        ],
+        "CUDA stream": [
+            "CUDA流的基本概念",
+            "默认流与非默认流",
+            "利用CUDA流重叠计算和数据传输"
+        ],
+        "CUDA 调试分析": [
+            "利用Nsight等分析工具对程序性能进行分析",
+            "根据实际硬件调整程序"
+        ],
+        "CUDA 优化": [
+            "存储优化",
+            "执行设置优化",
+            "指令级优化",
+            "控制流优化"
+        ],
+        "CUDA 加速库": [
+            "cuBLAS",
+            "cuFFT",
+            "cuRAND",
+            "cuSPARSE",
+            "cuDNN"
+        ]
+    }
+}
\ No newline at end of file

main.py

+from skill_tree.tree import TreeWalker, load_json, dump_json
+
+if __name__ == '__main__':
+    walker = TreeWalker("data", "cuda", "CUDA入门", ignore_keywords=True)
+    walker.walk()

requirements.txt

+.pre_commit
+skill-tree-parser
\ No newline at end of file