Skip to content

Commit e6f590c

Browse files
committed
add recall/readme
1 parent bce83d0 commit e6f590c

File tree

1 file changed

+61
-0
lines changed

1 file changed

+61
-0
lines changed

models/recall/readme.md

Lines changed: 61 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
# 召回模型库
2+
3+
## 简介
4+
我们提供了常见的召回任务中使用的模型算法的PaddleRec实现, 单机训练&预测效果指标以及分布式训练&预测性能指标等。实现的排序模型包括 [SR-GNN](http://gitlab.baidu.com/tangwei12/paddlerec/tree/develop/models/recall/gnn)[GRU4REC](http://gitlab.baidu.com/tangwei12/paddlerec/tree/develop/models/recall/gru4rec)[Sequence Semantic Retrieval Model](http://gitlab.baidu.com/tangwei12/paddlerec/tree/develop/models/recall/ssr)[Tree-based Deep Model](http://gitlab.baidu.com/tangwei12/paddlerec/tree/develop/models/recall/tdm)[Word2Vector](http://gitlab.baidu.com/tangwei12/paddlerec/tree/develop/models/recall/word2vec)
5+
6+
模型算法库在持续添加中,欢迎关注。
7+
8+
## 目录
9+
* [整体介绍](#整体介绍)
10+
* [召回模型列表](#召回模型列表)
11+
* [使用教程](#使用教程)
12+
* [数据处理](#数据处理)
13+
* [训练](#训练)
14+
* [预测](#预测)
15+
* [效果对比](#效果对比)
16+
* [模型效果列表](#模型效果列表)
17+
* [分布式](#分布式)
18+
* [模型性能列表](#模型性能列表)
19+
20+
## 整体介绍
21+
### 召回模型列表
22+
23+
| 模型 | 简介 | 论文 |
24+
| :------------------: | :--------------------: | :---------: |
25+
| GNN | SR-GNN | [Session-based Recommendation with Graph Neural Networks](https://arxiv.org/abs/1811.00855)(2018) |
26+
| GRU4REC | SR-GRU | [Session-based Recommendations with Recurrent Neural Networks](https://arxiv.org/abs/1511.06939)(2015) |
27+
| SSR | Sequence Semantic Retrieval Model | [Multi-Rate Deep Learning for Temporal Recommendation](http://sonyis.me/paperpdf/spr209-song_sigir16.pdf)(2016) |
28+
| TDM | Tree-based Deep Model | [Learning Tree-based Deep Model for Recommender Systems](https://arxiv.org/pdf/1801.02294.pdf)(2018) |
29+
| Word2Vec | word2vector | [Distributed Representations of Words and Phrases and their Compositionality](https://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf)(2013) |
30+
31+
## 使用教程
32+
### 数据处理
33+
```shell
34+
sh data_process.sh
35+
```
36+
### 训练
37+
```shell
38+
python -m fleetrec.run -m config.yaml -d cpu -e single
39+
```
40+
### 预测
41+
42+
## 效果对比
43+
### 模型效果列表
44+
45+
| 数据集 | 模型 | loss | 测试Recall@20 |
46+
| :------------------: | :--------------------: | :---------: |:---------: |
47+
| DIGINETICA | GNN | -- | 0.507 |
48+
| RSC15 | GRU4REC | -- | 0.67 |
49+
| RSC15 | SSR | -- | -- |
50+
| - | TDM | -- | -- |
51+
| 1 Billion Word Language Model Benchmark | Word2Vec | -- | 0.54 |
52+
53+
## 分布式
54+
### 模型性能列表
55+
| 数据集 | 模型 | 单机 | 多机(同步) | 多机(异步) | GPU |
56+
| :------------------: | :--------------------: | :---------: |:---------: |:---------: |:---------: |
57+
| DIGINETICA | GNN | -- | -- | -- | -- |
58+
| RSC15 | GRU4REC | -- | -- | -- | -- |
59+
| RSC15 | SSR | -- | -- | -- | -- |
60+
| - | TDM | -- | -- | -- | -- |
61+
| 1 Billion Word Language Model Benchmark | Word2Vec | -- | -- | -- | -- |

0 commit comments

Comments
 (0)