feat(project): release 0.1.0beta2

.gitignore

@@ -19,3 +19,4 @@ pyo3
 docs/_build
 *.dot
 .clippy.toml
+run_with_plugins_inner

Cargo.toml

@@ -2,7 +2,9 @@
 members = [
     "flow-rs",
     "flow-derive",
-    "flow-plugins"
+    "flow-plugins",
+    "flow-debugger",
+    "flow-message"
 ]
 
 [profile.dev]

README.md

@@ -23,27 +23,24 @@ MegFlow 提供快速视觉应用落地流程，最快 15 分钟搭建起视频
 * how to use
   * [tutorial01: image classification service](docs/how-to-add-my-service/01-single-classification-model.zh.md)
   * [tutorial02: detect and classify on video stream](docs/how-to-add-my-service/02-single-det-classify.zh.md)
+  * [tutorial03: batching and pipeline test](docs/how-to-add-my-service/03-batching-and-pipeline-test.zh.md)
 * [how to debug](docs/how-to-debug.zh.md)
 * [how to contribute](docs/how-to-contribute.zh.md)
 * [FAQ](docs/FAQ.zh.md)
 
-## Current Support List
+## Current Support Matrix
 
-| 系统环境               | 支持情况 |
-| ---------------------- | -------- |
-| Windows 10 wsl2        | ✔️        |
-| Windows 10 docker      | ✔️        |
-| x86 Ubuntu16.04 有 GPU | ✔️        |
-| x86 Ubuntu18.04 无 GPU | ✔️        |
-| x86 macos              | ✔️        |
-| ARM                    | ✔️        |
+| 系统环境 | win10 docker/wsl2 | ubuntu | centOS | macos |
+| -----------  | ------------------------- | ---------- | ---------  | ----- |
+| x86 | ✔️ | ✔️ | ✔️ | ✔️ |  
+| ARMv8 | - | ✔️ | ✔️ | - |
 
 | Python 版本 | 支持情况 |
 | ----------- | -------- |
 | 3.6         | ✔️        |
 | 3.7         | ✔️        |
 | 3.8         | ✔️        |
-| 3.9         | /        |
+| 3.9         | ✔️        |
 
 ## Built-in Applications
 * 猫猫围栏
@@ -59,12 +56,12 @@ MegFlow 提供快速视觉应用落地流程，最快 15 分钟搭建起视频
 - 支持 demux/reorder/transform 等通用函数式组件
 - Python 插件内置有栈协程，不依赖 asyncio
 - 基础测试工具，支持插件沙盒，用于单测插件
+- 基础调试工具，支持建图实时预览/qps profile
 
 ## Coming Soon
 - 进程级别的节点、子图支持
 - 插件自动化测试部署
-- 调试工具，建图实时预览 profile 工具
-- 性能监控，inspect 工具
+- 性能监控，inspect 等工具
 - 更多内置应用和组件
 
 ## Contact Us

ci/build_doc.sh

@@ -0,0 +1,11 @@
+#!/bin/bash -ex
+
+rm -rf public
+
+: ${TARGET_DIR:=/target/${CI_COMMIT_SHA}}
+
+cargo doc --no-deps
+
+echo '<meta http-equiv=refresh content=0;url=flow_rs/index.html>' > ${TARGET_DIR}/doc/index.html 
+
+mv ${TARGET_DIR}/doc public

ci/run_test.sh

@@ -0,0 +1,9 @@
+#!/bin/bash -ex
+
+cd flow-python
+
+python3 setup.py install
+
+cd examples
+
+cargo run --example run_with_plugins -- -p logical_test

docs/FAQ.zh.md

@@ -1,12 +1,5 @@
 # FAQ
 
-Q：`error while loading shared libraries: libpython3.8.xxx`。
-
-A：如果使用 conda 只需要 
-```bash
-$ export LD_LIBRARY_PATH=`conda info --base`/pkgs/python-3.8.11-xxx/lib:${LD_LIBRARY_PATH}
-```
-___
 Q：`run_with_plugins -p logical_test ` 无法运行怎么办？
 
 A1：如果报错 `message: "No such file or directory" }'`，确认是否`cd flow-python/examples`

docs/how-to-add-my-service/01-single-classification-model.zh.md

@@ -99,7 +99,7 @@ connections = [
 * `classify` 结点的参数。最重要的是 `ty="Classify"`指明了类名，MegFlow 将在当前目录搜索`Classify`类。path/device/device_id 分别是模型路径/用 CPU 推理/用哪个核，属于用户自定义配置
 * 服务类型。这里想运行图片服务 `ty = "ImageServer"`，如果想运行视频解析服务改 `ty = "VideoServer"`；图片服务默认返回图片，想返回 string 需要配置 `response = "json"`
 
-[完整的计算图 config 定义](appendix-A-graph-definition.md)
+[完整的计算图 config 定义](appendix-A-graph-definition.zh.md)
 
 ## 实现配置中的 node
 

docs/how-to-add-my-service/02-single-det-classify.zh.md

@@ -25,7 +25,7 @@ $ cat ${MegFlow}/flow-python/examples/simple_det_classify/dump.py
         # div 
         _div = mge.Tensor(np.array([57.375, 57.120, 58.395], dtype=np.float32))
         out = F.div(out, _div)
-        # dimshuffile 
+        # dimshuffle 
         out = F.transpose(out, (0,3,1,2))
 
         outputs = model(out)