From 8c7d113ee3205844273bf434acf07dc95fa53185 Mon Sep 17 00:00:00 2001
From: zhang wenhui <frankwhzhang@126.com>
Date: Mon, 31 Aug 2020 10:53:23 +0800
Subject: [PATCH] fix esmm (#172)

* fix esmm

* fix esmm

Co-authored-by: wuzhihua <35824027+fuyinno4@users.noreply.github.com>
Co-authored-by: tangwei12 <tangwei12@baidu.com>
---
 models/multitask/esmm/README.md   | 23 ++++++++++++++++-------
 models/multitask/esmm/config.yaml | 14 +++++++-------
 2 files changed, 23 insertions(+), 14 deletions(-)

diff --git a/models/multitask/esmm/README.md b/models/multitask/esmm/README.md
index 91a1df76..aecd9eda 100644
--- a/models/multitask/esmm/README.md
+++ b/models/multitask/esmm/README.md
@@ -50,11 +50,6 @@ ESMM是发表在 SIGIR’2018 的论文[《Entire Space Multi-Task Model: An E
 
 数据地址：[Ali-CCP：Alibaba Click and Conversion Prediction](  https://tianchi.aliyun.com/datalab/dataSet.html?dataId=408  )
 
-```
-cd data 
-sh run.sh
-```
-
 数据格式参见demo数据：data/train
 
 
@@ -108,11 +103,25 @@ CPU环境
 
 ## 论文复现
 
-用原论文的完整数据复现论文效果需要在config.yaml中修改batch_size=1000, thread_num=8, epoch_num=4
+由于原论文的数据太大，我们选取了部分数据作为训练和测试数据, 建议使用gpu训练。
+
+我们的测试ctr auc为0.79+，ctcvr auc为0.82+。
 
+```
+wget https://paddlerec.bj.bcebos.com/esmm/traindata_10w.csv  
+wget https://paddlerec.bj.bcebos.com/esmm/testdata_10w.csv 
+mkdir data/train_data data/test_data
+mv traindata_10w.csv data/train_data
+mv testdata_10w.csv data/test_data
+```
 
-修改后运行方案：修改config.yaml中的'workspace'为config.yaml的目录位置，执行
+用原论文的完整数据复现论文效果需要在config.yaml中修改batch_size=1024, epoch=10, device=gpu, selected_gpus:"0"
 
+具体配置可以下载config_10w.yaml文件
+```
+wget https://paddlerec.bj.bcebos.com/esmm/config_10w.yaml
+```
+修改后运行
 ```
 python -m paddlerec.run -m /home/your/dir/config.yaml #调试模式 直接指定本地config的绝对路径
 ```
diff --git a/models/multitask/esmm/config.yaml b/models/multitask/esmm/config.yaml
index 2a4478ba..25cfbe91 100644
--- a/models/multitask/esmm/config.yaml
+++ b/models/multitask/esmm/config.yaml
@@ -17,19 +17,19 @@ workspace: "models/multitask/esmm"
 
 dataset:
   - name: dataset_train
-    batch_size: 1
+    batch_size: 5
     type: QueueDataset
     data_path: "{workspace}/data/train"
     data_converter: "{workspace}/esmm_reader.py"
   - name: dataset_infer
-    batch_size: 1
+    batch_size: 5
     type: QueueDataset
     data_path: "{workspace}/data/test"
     data_converter: "{workspace}/esmm_reader.py"
 
 hyper_parameters:
-  vocab_size: 10000
-  embed_size: 128
+  vocab_size: 737946
+  embed_size: 12
   optimizer:
     class: adam
     learning_rate: 0.001
@@ -43,15 +43,15 @@ runner:
     class: train
     device: cpu
     epochs: 3
-    save_checkpoint_interval: 2
+    save_checkpoint_interval: 1
     save_inference_interval: 4
-    save_checkpoint_path: "increment"
+    save_checkpoint_path: "increment_esmm"
     save_inference_path: "inference"
     print_interval: 10
     phases: [train]
   - name: infer_runner
     class: infer
-    init_model_path: "increment/1"
+    init_model_path: "increment_esmm/1"
     device: cpu
     print_interval: 1
     phases: [infer]
-- 
GitLab