Comments (13)
参考的是
Get start of PaddleRec in ten minutes
10分钟开始不了啊
from paddlerec.
参考的是
Get start of PaddleRec in ten minutes
10分钟开始不了啊
看起来是paddlerec没有被正确安装~
你的运行环境是什么?前述的paddlerec安装步骤有顺利完成吗?
如果是用源代码安装,安装成功后,应该会有下面的标志
from paddlerec.
hi,大佬,安装成功后执行gnn,得到如下结果正常吗?感觉有点不对啊
python -m paddlerec.run -m paddlerec.models.recall.gnn
batch: 1, LOSS: [10.669986], train_acc: [0.]
batch: 2, LOSS: [10.683549], train_acc: [0.]
batch: 3, LOSS: [10.669658], train_acc: [0.]
batch: 4, LOSS: [10.671809], train_acc: [0.]
batch: 5, LOSS: [10.660932], train_acc: [0.]
batch: 6, LOSS: [10.66907], train_acc: [0.]
batch: 7, LOSS: [10.665563], train_acc: [0.]
batch: 8, LOSS: [10.674797], train_acc: [0.]
batch: 9, LOSS: [10.670386], train_acc: [0.]
batch: 10, LOSS: [10.662183], train_acc: [0.1]
batch: 11, LOSS: [10.667716], train_acc: [0.]
batch: 12, LOSS: [10.666458], train_acc: [0.]
batch: 13, LOSS: [10.66883], train_acc: [0.]
batch: 14, LOSS: [10.671003], train_acc: [0.]
batch: 15, LOSS: [10.6745], train_acc: [0.]
batch: 16, LOSS: [10.669777], train_acc: [0.]
batch: 17, LOSS: [10.675078], train_acc: [0.]
batch: 18, LOSS: [10.672194], train_acc: [0.]
batch: 19, LOSS: [10.670609], train_acc: [0.]
epoch 0 done, use time: 1.2693278789520264
batch: 1, LOSS: [10.416034], train_acc: [0.3]
batch: 2, LOSS: [10.377094], train_acc: [0.3]
batch: 3, LOSS: [10.380407], train_acc: [0.3]
batch: 4, LOSS: [10.326581], train_acc: [0.4]
batch: 5, LOSS: [10.391391], train_acc: [0.2]
batch: 6, LOSS: [10.374437], train_acc: [0.3]
batch: 7, LOSS: [10.369722], train_acc: [0.1]
batch: 8, LOSS: [10.341339], train_acc: [0.2]
batch: 9, LOSS: [10.202948], train_acc: [0.2]
batch: 10, LOSS: [10.351052], train_acc: [0.3]
batch: 11, LOSS: [10.318689], train_acc: [0.1]
batch: 12, LOSS: [10.254948], train_acc: [0.1]
batch: 13, LOSS: [10.40401], train_acc: [0.4]
batch: 14, LOSS: [10.613649], train_acc: [0.3]
batch: 15, LOSS: [10.598577], train_acc: [0.1]
batch: 16, LOSS: [10.588185], train_acc: [0.2]
batch: 17, LOSS: [10.5907], train_acc: [0.1]
batch: 18, LOSS: [10.577959], train_acc: [0.2]
batch: 19, LOSS: [10.562983], train_acc: [0.]
epoch 1 done, use time: 1.245844841003418
PaddleRec Finish
from paddlerec.
hi,大佬,安装成功后执行gnn,得到如下结果正常吗?感觉有点不对啊
python -m paddlerec.run -m paddlerec.models.recall.gnn batch: 1, LOSS: [10.669986], train_acc: [0.] batch: 2, LOSS: [10.683549], train_acc: [0.] batch: 3, LOSS: [10.669658], train_acc: [0.] batch: 4, LOSS: [10.671809], train_acc: [0.] batch: 5, LOSS: [10.660932], train_acc: [0.] batch: 6, LOSS: [10.66907], train_acc: [0.] batch: 7, LOSS: [10.665563], train_acc: [0.] batch: 8, LOSS: [10.674797], train_acc: [0.] batch: 9, LOSS: [10.670386], train_acc: [0.] batch: 10, LOSS: [10.662183], train_acc: [0.1] batch: 11, LOSS: [10.667716], train_acc: [0.] batch: 12, LOSS: [10.666458], train_acc: [0.] batch: 13, LOSS: [10.66883], train_acc: [0.] batch: 14, LOSS: [10.671003], train_acc: [0.] batch: 15, LOSS: [10.6745], train_acc: [0.] batch: 16, LOSS: [10.669777], train_acc: [0.] batch: 17, LOSS: [10.675078], train_acc: [0.] batch: 18, LOSS: [10.672194], train_acc: [0.] batch: 19, LOSS: [10.670609], train_acc: [0.] epoch 0 done, use time: 1.2693278789520264 batch: 1, LOSS: [10.416034], train_acc: [0.3] batch: 2, LOSS: [10.377094], train_acc: [0.3] batch: 3, LOSS: [10.380407], train_acc: [0.3] batch: 4, LOSS: [10.326581], train_acc: [0.4] batch: 5, LOSS: [10.391391], train_acc: [0.2] batch: 6, LOSS: [10.374437], train_acc: [0.3] batch: 7, LOSS: [10.369722], train_acc: [0.1] batch: 8, LOSS: [10.341339], train_acc: [0.2] batch: 9, LOSS: [10.202948], train_acc: [0.2] batch: 10, LOSS: [10.351052], train_acc: [0.3] batch: 11, LOSS: [10.318689], train_acc: [0.1] batch: 12, LOSS: [10.254948], train_acc: [0.1] batch: 13, LOSS: [10.40401], train_acc: [0.4] batch: 14, LOSS: [10.613649], train_acc: [0.3] batch: 15, LOSS: [10.598577], train_acc: [0.1] batch: 16, LOSS: [10.588185], train_acc: [0.2] batch: 17, LOSS: [10.5907], train_acc: [0.1] batch: 18, LOSS: [10.577959], train_acc: [0.2] batch: 19, LOSS: [10.562983], train_acc: [0.] epoch 1 done, use time: 1.245844841003418 PaddleRec Finish
这个是正常的,为了让用户简单跑起来,内置的是只有几十行的fake数据,所以loss和acc是不可参考的,替换成你的真实数据集跑下就可以了。
from paddlerec.
大佬,那么问题来了
咋改成gnn的train.txt数据集啊?
如下格式
1 1
2,3,4 4
2,3 4
2 3
5 6
7 8
9,10 11
9 10
12,13,14,14,15,16 17
12,13,14,14,15 16
12,13,14,14 15
12,13,14 14
12,13 14
12 13
18,18 19
18 18
20 21
5 22
23,24 25
23 24
26,11,10,27,28,29 30
26,11,10,27,28 29
26,11,10,27 28
26,11,10 27
26,11 10
26 11
31,32 33
31 32
34 35
36,37,38,39,40,37 41
36,37,38,39,40 37
36,37,38,39 40
36,37,38 39
36,37 38
36 37
42 43
44 45
46 47
48,49,48,50,51,22 51
48,49,48,50,51 22
48,49,48,50 51
48,49,48 50
48,49 48
48 49
52,52 53
52 52
54,55 56
54 55
57 57
58,59 58
58 59
60,61,61,62,63,62,64,65,62,66,67,68 68
60,61,61,62,63,62,64,65,62,66,67 68
60,61,61,62,63,62,64,65,62,66 67
60,61,61,62,63,62,64,65,62 66
60,61,61,62,63,62,64,65 62
from paddlerec.
大佬,那么问题来了
咋改成gnn的train.txt数据集啊?
如下格式1 1 2,3,4 4 2,3 4 2 3 5 6 7 8 9,10 11 9 10 12,13,14,14,15,16 17 12,13,14,14,15 16 12,13,14,14 15 12,13,14 14 12,13 14 12 13 18,18 19 18 18 20 21 5 22 23,24 25 23 24 26,11,10,27,28,29 30 26,11,10,27,28 29 26,11,10,27 28 26,11,10 27 26,11 10 26 11 31,32 33 31 32 34 35 36,37,38,39,40,37 41 36,37,38,39,40 37 36,37,38,39 40 36,37,38 39 36,37 38 36 37 42 43 44 45 46 47 48,49,48,50,51,22 51 48,49,48,50,51 22 48,49,48,50 51 48,49,48 50 48,49 48 48 49 52,52 53 52 52 54,55 56 54 55 57 57 58,59 58 58 59 60,61,61,62,63,62,64,65,62,66,67,68 68 60,61,61,62,63,62,64,65,62,66,67 68 60,61,61,62,63,62,64,65,62,66 67 60,61,61,62,63,62,64,65,62 66 60,61,61,62,63,62,64,65 62
- 克隆PaddleRec项目,如果已经克隆可跳过这一步
git clone https://github.com/PaddlePaddle/PaddleRec.git - 假设工程目录为/home/malin10/code/tmp/PaddleRec,进入gnn模型目录下:
- 执行sh data_prepare.sh准备真实数据
- 修改config.yaml中的workspace的路径,换成当前路径
pwd
,如下所示:
- 执行python -m paddlerec.run -m /home/malin10/code/tmp/PaddleRec/models/recall/gnn/config.yaml
from paddlerec.
@123malin
大佬data_prepare.sh下载的数据如下示例:
session_id;user_id;item_id;timeframe;eventdate
1;NA;81766;526309;2016-05-09
1;NA;31331;1031018;2016-05-09
1;NA;32118;243569;2016-05-09
1;NA;9654;75848;2016-05-09
1;NA;32627;1112408;2016-05-09
1;NA;33043;173912;2016-05-09
1;NA;12352;329870;2016-05-09
1;NA;35077;390072;2016-05-09
1;NA;36118;487369;2016-05-09
1;NA;129055;991416;2016-05-09
2;NA;3147;883039;2016-05-09
2;NA;32754;359839;2016-05-09
2;NA;100747;38317;2016-05-09
2;NA;32971;182759;2016-05-09
2;NA;10657;387340;2016-05-09
2;NA;35606;756335;2016-05-09
2;NA;35606;53684;2016-05-09
2;NA;36246;332000;2016-05-09
2;NA;36246;722492;2016-05-09
2;NA;196110;1045018;2016-05-09
4;NA;13931;21812;2016-05-09
5;NA;5140;22430;2016-05-09
5;NA;35472;121036;2016-05-09
6;NA;73191;53123;2016-05-09
6;NA;382683;88337;2016-05-09
7;NA;59248;38806;2016-05-09
7;NA;59248;328086;2016-05-09
7;NA;9416;271202;2016-05-09
7;NA;40639;14808;2016-05-09
7;NA;40639;432980;2016-05-09
7;NA;40639;161072;2016-05-09
7;NA;73816;69100;2016-05-09
7;NA;32742;186920;2016-05-09
7;NA;79626;358743;2016-05-09
7;NA;388131;583388;2016-05-09
8;NA;11542;10067;2016-04-10
10;NA;69597;193275;2016-04-05
10;NA;5103;5639;2016-04-05
.....
295645;NA;32950;263151;2016-03-22
295645;NA;68708;381295;2016-03-22
295645;NA;7017;22863;2016-03-22
295646;117396;20717;371823;2016-03-17
295646;117396;15330;822956;2016-03-17
295646;117396;30605;342923;2016-03-17
295646;117396;93790;909621;2016-03-17
295646;117396;15714;898750;2016-03-17
295646;117396;73107;29211;2016-03-17
295646;117396;129720;1032305;2016-03-17
295646;117396;129720;1011780;2016-03-17
295646;117396;65789;1081925;2016-03-17
295646;117396;69199;969417;2016-03-17
295646;117396;6961;1019633;2016-03-17
....
请教下session-id是啥啊?
里面为啥好多NA啊,user-id是NA表示啥意思呢?
多谢大佬
from paddlerec.
ai, rensheng jiannan a
from paddlerec.
这里面的等于1为啥要去掉啊
# Filter out length 1 sessions
for s in list(sess_clicks):
if len(sess_clicks[s]) == 1:
del sess_clicks[s]
del sess_date[s]
from paddlerec.
请教下timeframe是啥意思啊?
在一个session_id内按照timeframe对item_id排序是什么意思呢?咋不考虑eventdate呢?话说eventdate是什么呢?都是时间,一个具体一个不具体??是这样吗?
多谢大佬
from paddlerec.
@123malin
大佬data_prepare.sh下载的数据如下示例:session_id;user_id;item_id;timeframe;eventdate 1;NA;81766;526309;2016-05-09 1;NA;31331;1031018;2016-05-09 1;NA;32118;243569;2016-05-09 1;NA;9654;75848;2016-05-09 1;NA;32627;1112408;2016-05-09 1;NA;33043;173912;2016-05-09 1;NA;12352;329870;2016-05-09 1;NA;35077;390072;2016-05-09 1;NA;36118;487369;2016-05-09 1;NA;129055;991416;2016-05-09 2;NA;3147;883039;2016-05-09 2;NA;32754;359839;2016-05-09 2;NA;100747;38317;2016-05-09 2;NA;32971;182759;2016-05-09 2;NA;10657;387340;2016-05-09 2;NA;35606;756335;2016-05-09 2;NA;35606;53684;2016-05-09 2;NA;36246;332000;2016-05-09 2;NA;36246;722492;2016-05-09 2;NA;196110;1045018;2016-05-09 4;NA;13931;21812;2016-05-09 5;NA;5140;22430;2016-05-09 5;NA;35472;121036;2016-05-09 6;NA;73191;53123;2016-05-09 6;NA;382683;88337;2016-05-09 7;NA;59248;38806;2016-05-09 7;NA;59248;328086;2016-05-09 7;NA;9416;271202;2016-05-09 7;NA;40639;14808;2016-05-09 7;NA;40639;432980;2016-05-09 7;NA;40639;161072;2016-05-09 7;NA;73816;69100;2016-05-09 7;NA;32742;186920;2016-05-09 7;NA;79626;358743;2016-05-09 7;NA;388131;583388;2016-05-09 8;NA;11542;10067;2016-04-10 10;NA;69597;193275;2016-04-05 10;NA;5103;5639;2016-04-05 ..... 295645;NA;32950;263151;2016-03-22 295645;NA;68708;381295;2016-03-22 295645;NA;7017;22863;2016-03-22 295646;117396;20717;371823;2016-03-17 295646;117396;15330;822956;2016-03-17 295646;117396;30605;342923;2016-03-17 295646;117396;93790;909621;2016-03-17 295646;117396;15714;898750;2016-03-17 295646;117396;73107;29211;2016-03-17 295646;117396;129720;1032305;2016-03-17 295646;117396;129720;1011780;2016-03-17 295646;117396;65789;1081925;2016-03-17 295646;117396;69199;969417;2016-03-17 295646;117396;6961;1019633;2016-03-17 ....
请教下session-id是啥啊?
里面为啥好多NA啊,user-id是NA表示啥意思呢?
多谢大佬
session可以理解为一个浏览器session概念,创建一个网页会分配一个session。比如用户打开一个网页,点击了一系列物品,这整个过程都对应同一个session_id
user_id 为NA 应该是这个这个模型/数据集不需要用户用户信息
from paddlerec.
这里面的等于1为啥要去掉啊
# Filter out length 1 sessions for s in list(sess_clicks): if len(sess_clicks[s]) == 1: del sess_clicks[s] del sess_date[s]
这是个序列预测问题,长度为1的序列没有训练意义,就过滤掉了
from paddlerec.
此问题暂时没有收到有效的回复, 会close, 如有需要可reopen或重开issue.
from paddlerec.
Related Issues (20)
- ValueError: The ``path`` (output_model_xdeepfm/0/rec.pdparams) to load model not exists.
- 关于MHCN代码的Loss function问题请教。
- 同样是Session Based Recommendation算法,为什么GRU4Rec是用于召回,而BERT4Rec用于排序? HOT 1
- Ple模型获取数据时出错 HOT 1
- 关于escm2中的操作 HOT 1
- 运行paddlerec报错OSError: (External) adam XDNN Error, XDNN_RUNTIME_ERROR (at /data/paddle/Paddle/paddle/phi/kernels/selected_rows/xpu/adam_kernel.cc:318)
- 运行paddlerec报错OSError: (External) adam XDNN Error, XDNN_RUNTIME_ERROR (at /data/paddle/Paddle/paddle/phi/kernels/selected_rows/xpu/adam_kernel.cc:318) HOT 1
- Implement Fastformer in the rank models for the experiments on MIND dataset HOT 1
- config.yaml文件中save_inference_fetch_varnames参数的问题 HOT 3
- Warning:: 0D Tensor cannot be used as 'Tensor.numpy()[0]' . HOT 1
- 排序模型
- ModuleNotFoundError: No module named 'paddle.fluid' HOT 5
- 运行paddle_infer.py报错
- PaddleRec的deepfm模型不明原因出core #61629
- mmoe模型配置缺少auc_num导致AUC计算错误
- Word2Vec 负样本 id 没有做到单词的映射
- Requirement.parse Error
- MMOE\PLE模型sparse特征为什么没有转化为embedding啊?
- ESCM2模型的IPW代码
- config.yaml开启了use_inference: True,为什么执行静态图训练(static_trainer.py)保存的模型缺少rec_inference.pdiparams
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from paddlerec.