iamyourboss / deeprec Goto Github PK

DeepRec is a recommendation engine based on TensorFlow 1.15, Intel-TensorFlow and NVIDIA-TensorFlow.

Sparse model is a type of deep learning model that accounts for a relatively high proportion of discrete feature calculation logic in the model structure. Discrete features are usually expressed as non-numeric features that cannot be directly processed by algorithms such as id, tag, text, and phrases. They are widely used in high-value businesses such as search, advertising, and recommendation.

DeepRec has been deeply cultivated since 2016, which supports core businesses such as Taobao Search, recommendation and advertising. It precipitates a large number of optimized operators on basic frameworks and has excellent performance in sparse models training.Facing a wide variety of external needs and the environment of deep learning framework embracing open source, deeperec open source is conducive to establishing standardized interfaces, cultivating user habits, greatly reducing the cost of external customers working on cloud and establishing the brand value.

DeepRec has super large-scale distributed training capability, supporting model training of trillion samples and 100 billion Embedding Processing. For sparse model scenarios, in-depth performance optimization has ben conducted across CPU and GPU platform. It contains 3 kinds of key features to improve usability and performance for super-scale scenarios.

• Dynamic and elastic embedding processing
• Embedding elimination and access
• Dynamic elastic dimension based on embedding frequency
• Adaptive dynamic elastic embedding

• Distributed Training Framework Optimization, such as grpc+seastar,FuseRecv etc
• Runtime Optimization
• Operator level optimization, such as BF16 mixed precision optimizaiton, sparse operator optimizaiton and EmbeddingVariable on PMEM and GPU, new hardware feature enabling, etc.
• Graph level optimization, such as embedding fusion, Attention and MLP, etc.

• Incremental model loading and exporting
• Super-scale sparse model distributed serving
• Multilevel hybrid storage and multi backend supportcd ..
• Online deep learning with low latency

CPU Platform

registry.cn-shanghai.aliyuncs.com/pai-dlc/tensorflow-developer:1.15deeprec2106-cpu-py36-ubuntu18.04

GPU Platform

registry.cn-shanghai.aliyuncs.com/pai-dlc/tensorflow-developer:1.15deeprec2106-gpu-py36-cu110-ubuntu18.04

configure

$ ./configure

Compile for CPU and GPU defaultly

$ bazel build -c opt --config=opt //tensorflow/tools/pip_package:build_pip_package

Compile for CPU and GPU: ABI=0

$ bazel build --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --host_cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" -c opt --config=opt //tensorflow/tools/pip_package:build_pip_package

Compile for CPU optimization: oneDNN + Unified Eigen Thread pool

$ bazel build  -c opt --config=opt  --config=mkl_threadpool --define build_with_mkl_dnn_v1_only=true //tensorflow/tools/pip_package:build_pip_package

Compile for CPU optimization and ABI=0

$ bazel build --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --host_cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" -c opt --config=opt --config=mkl_threadpool --define build_with_mkl_dnn_v1_only=true //tensorflow/tools/pip_package:build_pip_package

$ ./bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg

$ pip3 install /tmp/tensorflow_pkg/tensorflow-1.15.5+deeprec2106-cp36-cp36m-linux_x86_64.whl

registry.cn-shanghai.aliyuncs.com/pai-dlc/tensorflow-training:deeprec-nightly-gpu-py36-cu110-ubuntu18.04

registry.cn-shanghai.aliyuncs.com/pai-dlc/tensorflow-training:deeprec-nightly-cpu-py36-ubuntu18.04

$ ./configure
$ bazel build --config opt //tensorflow/java:tensorflow   //tensorflow/java:libtensorflow_jni
$ javac -cp bazel-bin/tensorflow/java/libtensorflow.jar ...
$ java -cp bazel-bin/tensorflow/java/libtensorflow.jar  -Djava.library.path=bazel-bin/tensorflow/java  ...

Apache License 2.0