DeepRec is a training and inference engine for search, recommendation, and advertising scenarios based on TensorFlow1.15, Intel-TensorFlow and NVIDIA-TensorFlow. Background sparse model refers to a general term for a class of deep learning models with a high proportion of discrete feature calculation logic in the model structure. Discrete features are usually non-numerical features that cannot be directly processed by algorithms such as id, tag, text, and phrase. They are widely used in high-value services such as search, advertising, and recommendation. DeepRec(PAI-TF), as the training/prediction engine of Alibaba Group’s unified sparse model business scenario, supports Taobao…

#DeepRec #trainingprediction #engine #sparse #model #business #scenarios

Leave a Comment

Your email address will not be published. Required fields are marked *