Navi is a high-performance, versatile machine learning serving server implemented in Rust and tailored for production usage. It's designed to efficiently serve within the Twitter tech stack, offering top-notch performance while focusing on core features.
While Navi's features may not be as comprehensive as its open-source counterparts, its performance-first mindset makes it highly efficient.
navi
: The main code repository for Navidr_transform
: Twitter-specific converter that converts BatchPredictionRequest Thrift to ndarraysegdense
: Twitter-specific config to specify how to retrieve feature values from BatchPredictionRequestthrift_bpr_adapter
: generated thrift code for BatchPredictionRequestWe have included all *.rs source code files that make up the main Navi binaries for you to examine. However, we have not included the test and benchmark code, or various configuration files, due to data security concerns.
In navi/navi, you can run the following commands:
scripts/run_tf2.sh
for TensorFlow
scripts/run_onnx.sh
for Onnx
Do note that you need to create a models directory and create some versions, preferably using epoch time, e.g., 1679693908377
.
so the models structure looks like:
models/
-web_click
- 1809000
- 1809010
You can adapt the above scripts to build using Cargo.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。