From 6b29a07d9b9395344d09da601a4f1c9371d164df Mon Sep 17 00:00:00 2001 From: fangzehua Date: Mon, 10 May 2021 14:12:42 +0800 Subject: [PATCH] fix cache doc --- model_zoo/official/recommend/wide_and_deep/README.md | 2 +- model_zoo/official/recommend/wide_and_deep/README_CN.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/model_zoo/official/recommend/wide_and_deep/README.md b/model_zoo/official/recommend/wide_and_deep/README.md index 73d635f5194..1faa920dff4 100644 --- a/model_zoo/official/recommend/wide_and_deep/README.md +++ b/model_zoo/official/recommend/wide_and_deep/README.md @@ -34,7 +34,7 @@ Wide&Deep model is a classical model in Recommendation and Click Prediction area Wide&Deep model jointly trained wide linear models and deep neural network, which combined the benefits of memorization and generalization for recommender systems. -Currently we support host-device mode with column partition and parameter server mode, and we implement the cache mode that supported multi-dimensional slice for huge embedding table which cooperated with Noah's Ark Lab([ScaleFreeCTR](https://arxiv.org/abs/2104.08542)). +Currently we support host-device mode with multi-dimensional partition parallel for embedding table and parameter server mode, and we implement the cache mode for huge embedding table which cooperated with Noah's Ark Lab([ScaleFreeCTR](https://arxiv.org/abs/2104.08542)). # [Dataset](#contents) diff --git a/model_zoo/official/recommend/wide_and_deep/README_CN.md b/model_zoo/official/recommend/wide_and_deep/README_CN.md index 0c1a197e350..5b8179e738a 100644 --- a/model_zoo/official/recommend/wide_and_deep/README_CN.md +++ b/model_zoo/official/recommend/wide_and_deep/README_CN.md @@ -36,7 +36,7 @@ Wide&Deep模型是推荐和点击预测领域的经典模型。 [Wide&Deep推 Wide&Deep模型训练了宽线性模型和深度学习神经网络，结合了推荐系统的记忆和泛化的优点。 -目前我们支持列式分区的主机设备模式和参数服务器模式，且已和诺亚实验室合作实现了超大规模推荐网络多维度切分的缓存方案（[ScaleFreeCTR](https://arxiv.org/abs/2104.08542)）。 +目前我们支持embedding多维度切分并行的主机设备模式和参数服务器模式，且已和诺亚实验室合作实现了超大规模推荐网络的缓存方案（[ScaleFreeCTR](https://arxiv.org/abs/2104.08542)）。 # 数据集