-
Notifications
You must be signed in to change notification settings - Fork 68
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Move djl-serving project from djl repo
- Loading branch information
Showing
145 changed files
with
17,979 additions
and
16 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 +1,27 @@ | ||
| .gradle | ||
| .DS_Store | ||
| .idea | ||
| *.iml | ||
| build | ||
| libs | ||
| *.gz | ||
| *.params | ||
| *.zip | ||
| *.jar | ||
| *.so | ||
| *.dylib | ||
| *.dll | ||
| *.class | ||
|
|
||
| # Eclipse | ||
| .settings | ||
| .project | ||
| .classpath | ||
| bin/ | ||
|
|
||
| #node | ||
| node_modules/ | ||
| **/static/dist/ | ||
|
|
||
| #vscode | ||
| .vscode |
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 |
|---|---|---|
| @@ -1,17 +1,154 @@ | ||
| ## My Project | ||
| # DJL Serving | ||
|
|
||
| TODO: Fill this README out! | ||
| ## Overview | ||
|
|
||
| Be sure to: | ||
| DJL serving is a high performance universal model serving solution. You can use djl-serving serve the | ||
| following models out of the box: | ||
|
|
||
| * Change the title in this README | ||
| * Edit your repository description on GitHub | ||
| - PyTorch TorchScript model | ||
| - TensorFlow SavedModel bundle | ||
| - Apache MXNet model | ||
|
|
||
| ## Security | ||
| You can install extra extensions to enable the following models: | ||
|
|
||
| See [CONTRIBUTING](CONTRIBUTING.md#security-issue-notifications) for more information. | ||
| - ONNX model | ||
| - PaddlePaddle model | ||
| - TFLite model | ||
| - Neo DLR (TVM) model | ||
| - XGBoost model | ||
| - Sentencepiece model | ||
| - fastText/BlazingText model | ||
|
|
||
| ## License | ||
| ## Architecture | ||
|
|
||
| This project is licensed under the Apache-2.0 License. | ||
| DJL serving is built on top of [Deep Java Library](https://djl.ai). You can visit | ||
| [DJL github repository](https://github.com/deepjavalibrary/djl) to learn more about DJL. | ||
|
|
||
|  | ||
|
|
||
| ## Key features | ||
|
|
||
| - Performance, DJL serving running multithreading inference in a single JVM. Our benchmark shows | ||
| DJL serving has higher throughput than most of C++ based model server on the market. | ||
| - Ease of use, DJL serving can serve most of the models out of box. | ||
| - Easy to extend, DJL serving plugins make it easy for user to add their own extensions. | ||
| - Auto-scale, DJL serving automatically scales up/down worker threads based on the load. | ||
| - Dynamic batching, DJL serving supports dynamic batching to increase throughput. | ||
| - Model versioning, DJL allows user to load different version of a model on a single endpoint. | ||
| - Multi-engine support, DJL allows user to serve models from different engines at the same time. | ||
|
|
||
| ## Installation | ||
|
|
||
| For macOS | ||
|
|
||
| ``` | ||
| brew cask install djl-serving | ||
| # Start djl-serving as service: | ||
| brew services start djl-serving | ||
| # Stop djl-serving service | ||
| brew services stop djl-serving | ||
| ``` | ||
|
|
||
| For Ubuntu | ||
|
|
||
| ``` | ||
| curl -O https://publish.djl.ai/djl-serving/djl-serving_0.12.0-1_all.deb | ||
| sudo dpkg -i djl-serving_0.12.0-1_all.deb | ||
| ``` | ||
|
|
||
| For Windows | ||
|
|
||
| We are considering to create a `chocolatey` package for Windows. For the time being, you can | ||
| download djl-serving zip file from [here](https://publish.djl.ai/djl-serving/serving-0.12.0.zip). | ||
|
|
||
| ### Docker | ||
|
|
||
| You can also use docker to run DJL Serving: | ||
|
|
||
| ``` | ||
| docker run -itd -p 8080:8080 deepjavalibrary/djl-serving | ||
| ``` | ||
|
|
||
| ## Run DJL Serving | ||
|
|
||
| Use the following command to start model server locally: | ||
|
|
||
| ```sh | ||
| djl-serving | ||
| ``` | ||
|
|
||
| The model server will be listening on port 8080. You can also load a model for serving on start up: | ||
|
|
||
| ```sh | ||
| djl-serving -m "https://resources.djl.ai/demo/mxnet/resnet18_v1.zip" | ||
| ``` | ||
|
|
||
| Open another terminal, and type the following command to test the inference REST API: | ||
|
|
||
| ```sh | ||
| curl -O https://resources.djl.ai/images/kitten.jpg | ||
| curl -X POST http://localhost:8080/predictions/resnet18_v1 -T kitten.jpg | ||
|
|
||
| or: | ||
|
|
||
| curl -X POST http://localhost:8080/predictions/resnet18_v1 -F "[email protected]" | ||
|
|
||
| [ | ||
| { | ||
| "className": "n02123045 tabby, tabby cat", | ||
| "probability": 0.4838452935218811 | ||
| }, | ||
| { | ||
| "className": "n02123159 tiger cat", | ||
| "probability": 0.20599420368671417 | ||
| }, | ||
| { | ||
| "className": "n02124075 Egyptian cat", | ||
| "probability": 0.18810515105724335 | ||
| }, | ||
| { | ||
| "className": "n02123394 Persian cat", | ||
| "probability": 0.06411745399236679 | ||
| }, | ||
| { | ||
| "className": "n02127052 lynx, catamount", | ||
| "probability": 0.010215568356215954 | ||
| } | ||
| ] | ||
| ``` | ||
|
|
||
| For more command line options: | ||
|
|
||
| ```sh | ||
| djl-serving --help | ||
| usage: djl-serving [OPTIONS] | ||
| -f,--config-file <CONFIG-FILE> Path to the configuration properties file. | ||
| -h,--help Print this help. | ||
| -m,--models <MODELS> Models to be loaded at startup. | ||
| -s,--model-store <MODELS-STORE> Model store location where models can be loaded. | ||
| ``` | ||
|
|
||
| ## REST API | ||
|
|
||
| DJL Serving use RESTful API for both inference and management calls. | ||
|
|
||
| When DJL Serving startup, it starts two web services: | ||
| * [Inference API](docs/inference_api.md) | ||
| * [Management API](docs/management_api.md) | ||
|
|
||
| By default, DJL Serving listening on 8080 port and only accessible from localhost. | ||
| Please see [DJL Serving Configuration](docs/configuration.md) for how to enable access from remote host. | ||
|
|
||
| # Plugin management | ||
|
|
||
| DJL Serving supports plugins, user can implement their own plugins to enrich DJL Serving features. | ||
| See [DJL Plugin Management](docs/plugin_management.md) for how to install plugins to DJL Serving. | ||
|
|
||
| ## Logging | ||
| you can set the logging level on the command-line adding a parameter for the JVM | ||
|
|
||
| ```sh | ||
| -Dai.djl.logging.level={FATAL|ERROR|WARN|INFO|DEBUG|TRACE} | ||
| ``` |
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 +1,116 @@ | ||
| plugins { | ||
| id "com.github.spotbugs" version "4.7.1" apply false | ||
| } | ||
|
|
||
| defaultTasks 'build' | ||
|
|
||
| def djl_version = System.getenv("DJL_VERSION") | ||
| djl_version = (djl_version == null) ? "0.13.0-SNAPSHOT" : djl_version | ||
|
|
||
| allprojects { | ||
| group 'ai.djl' | ||
| version "${djl_version}" | ||
|
|
||
| repositories { | ||
| mavenCentral() | ||
| maven { | ||
| url 'https://oss.sonatype.org/content/repositories/snapshots/' | ||
| } | ||
| } | ||
|
|
||
| apply plugin: 'idea' | ||
| idea { | ||
| module { | ||
| outputDir = file('build/classes/java/main') | ||
| testOutputDir = file('build/classes/java/test') | ||
| // inheritOutputDirs = true | ||
| } | ||
| } | ||
| } | ||
|
|
||
| def javaProjects() { | ||
| return subprojects.findAll { new File(it.projectDir, "src/main").exists() } | ||
| } | ||
|
|
||
| configure(javaProjects()) { | ||
| apply plugin: 'java-library' | ||
| sourceCompatibility = 1.8 | ||
| targetCompatibility = 1.8 | ||
| compileJava.options.encoding = "UTF-8" | ||
| compileTestJava.options.encoding = "UTF-8" | ||
| if (JavaVersion.current() != JavaVersion.VERSION_1_8) { | ||
| compileJava.options.compilerArgs.addAll(["--release", "8"]) | ||
| } | ||
|
|
||
| apply plugin: 'eclipse' | ||
|
|
||
| eclipse { | ||
| jdt.file.withProperties { props -> | ||
| props.setProperty "org.eclipse.jdt.core.circularClasspath", "warning" | ||
| } | ||
| classpath { | ||
| sourceSets.test.java { | ||
| srcDirs = ["src/test/java"] | ||
| exclude "**/package-info.java" | ||
| } | ||
| } | ||
| } | ||
|
|
||
| apply from: file("${rootProject.projectDir}/tools/gradle/java-formatter.gradle") | ||
| apply from: file("${rootProject.projectDir}/tools/gradle/check.gradle") | ||
|
|
||
| test { | ||
| // tensorflow mobilenet and resnet require more cpu memory | ||
| maxHeapSize = "4096m" | ||
| doFirst { | ||
| if (JavaVersion.current() != JavaVersion.VERSION_1_8) { | ||
| jvmArgs = [ | ||
| '--add-opens', "java.base/jdk.internal.loader=ALL-UNNAMED" | ||
| ] | ||
| } | ||
| } | ||
|
|
||
| useTestNG() { | ||
| // suiteXmlFiles << new File(rootDir, "testng.xml") //This is how to add custom testng.xml | ||
| } | ||
|
|
||
| testLogging { | ||
| showStandardStreams = true | ||
| events "passed", "skipped", "failed", "standardOut", "standardError" | ||
| } | ||
|
|
||
| doFirst { | ||
| systemProperties System.getProperties() | ||
| systemProperties.remove("user.dir") | ||
| // systemProperty "ai.djl.logging.level", "debug" | ||
| systemProperty "org.slf4j.simpleLogger.defaultLogLevel", "debug" | ||
| systemProperty "org.slf4j.simpleLogger.log.org.mortbay.log", "warn" | ||
| systemProperty "disableProgressBar", "true" | ||
| systemProperty "nightly", System.getProperty("nightly", "false") | ||
| if (gradle.startParameter.offline) { | ||
| systemProperty "offline", "true" | ||
| } | ||
| // This is used to avoid overriding on default engine for modules: | ||
| // mxnet-engine, mxnet-model-zoo, api (MockEngine), basicdataset, fasttext, etc | ||
| if (project.name != "integration" && project.name != "examples") { | ||
| systemProperties.remove("ai.djl.default_engine") | ||
| } | ||
| } | ||
| } | ||
|
|
||
| compileJava { | ||
| options.compilerArgs << "-proc:none" << "-Xlint:all,-options,-static" << "-Werror" | ||
| } | ||
|
|
||
| compileTestJava { | ||
| options.compilerArgs << "-proc:none" << "-Xlint:all,-options,-static" << "-Werror" | ||
| } | ||
|
|
||
| jar { | ||
| manifest { | ||
| attributes("Automatic-Module-Name": "ai.djl.${project.name.replace('-', '_')}") | ||
| } | ||
| } | ||
| } | ||
|
|
||
| apply from: file("${rootProject.projectDir}/tools/gradle/jacoco.gradle") |
Oops, something went wrong.