Add VecMVM - Githubissues

cloudml / zen

Zen aims to provide the largest scale and the most efficient machine learning platform on top of Spark, including but not limited to logistic regression, latent dirichilet allocation, factorization machines and DNN.

Apache License 2.0

170 stars 75 forks source link

Training loss : 3.7500901095175676 Training loss : 3.6481977670037447 Training loss : 3.0154451631968673 Training loss : 2.301574750976185 Training loss : 1.64470397565321 (Iteration 5/1000) Test RMSE: 1.233504 Training loss : 1.2284820306510036 Training loss : 1.4732805597950736 Training loss : 1.1205931927870145 Training loss : 0.9359770573525823 Training loss : 0.9216058314346698 (Iteration 10/1000) Test RMSE: 0.925369 Training loss : 0.913410135170783 Training loss : 1.1816797199323703 Training loss : 1.651559537221694 Training loss : 1.1079543313491855 Training loss : 0.9330512755211859 (Iteration 15/1000) Test RMSE: 0.933812 Training loss : 0.91655996522135 Training loss : 1.3159227421256716 Training loss : 1.4240307849058782 Training loss : 1.105259133711867 Training loss : 0.9165354277926252 (Iteration 20/1000) Test RMSE: 0.919473 Training loss : 0.9028139663645766 Training loss : 1.3253980953053717 Training loss : 1.350138999697169 Training loss : 0.9381364146316845 Training loss : 0.902270442685908 (Iteration 25/1000) Test RMSE: 0.910946 Training loss : 0.8880398655525505 Training loss : 1.2689394561424192 Training loss : 1.3283529096666535 Training loss : 1.0666820486691853 Training loss : 0.8905154300146964 (Iteration 30/1000) Test RMSE: 0.901018 Training loss : 0.8713604279066419 Training loss : 1.3965301563797694 Training loss : 1.2681387865382572 Training loss : 0.8955762717430986 Training loss : 0.8638836408180803 (Iteration 35/1000) Test RMSE: 0.892930 Training loss : 0.8515089350824213 Training loss : 1.236073300499966 Training loss : 1.3715636757285579 Training loss : 1.0443572333235276 Training loss : 0.860208250298429 (Iteration 40/1000) Test RMSE: 0.893269 Training loss : 0.8388994489050412 Training loss : 1.3173294754531584 Training loss : 1.3666662175047999 Training loss : 0.8870536556003747 Training loss : 0.8364603386191529 (Iteration 45/1000) Test RMSE: 0.889765 Training loss : 0.8201107027514734 Training loss : 1.2157862845113319 Training loss : 1.3496795985855927 Training loss : 1.002764761280667 Training loss : 0.831910395941351 (Iteration 50/1000) Test RMSE: 0.894383

Haipeng, would you please take a look?

Get Outlook for iOS

On Sat, Aug 13, 2016 at 10:07 PM -0700, "Guoqiang Li" notifications@github.com wrote:

似乎有一些bug ,我跑的结果不太对:


Training loss : 3.7500901095175676
Training loss : 3.6481977670037447
Training loss : 3.0154451631968673
Training loss : 2.301574750976185
Training loss : 1.64470397565321
(Iteration 5/1000) Test RMSE:                     1.233504
Training loss : 1.2284820306510036
Training loss : 1.4732805597950736
Training loss : 1.1205931927870145
Training loss : 0.9359770573525823
Training loss : 0.9216058314346698
(Iteration 10/1000) Test RMSE:                     0.925369
Training loss : 0.913410135170783
Training loss : 1.1816797199323703
Training loss : 1.651559537221694
Training loss : 1.1079543313491855
Training loss : 0.9330512755211859
(Iteration 15/1000) Test RMSE:                     0.933812
Training loss : 0.91655996522135
Training loss : 1.3159227421256716
Training loss : 1.4240307849058782
Training loss : 1.105259133711867
Training loss : 0.9165354277926252
(Iteration 20/1000) Test RMSE:                     0.919473
Training loss : 0.9028139663645766
Training loss : 1.3253980953053717
Training loss : 1.350138999697169
Training loss : 0.9381364146316845
Training loss : 0.902270442685908
(Iteration 25/1000) Test RMSE:                     0.910946
Training loss : 0.8880398655525505
Training loss : 1.2689394561424192
Training loss : 1.3283529096666535
Training loss : 1.0666820486691853
Training loss : 0.8905154300146964
(Iteration 30/1000) Test RMSE:                     0.901018
Training loss : 0.8713604279066419
Training loss : 1.3965301563797694
Training loss : 1.2681387865382572
Training loss : 0.8955762717430986
Training loss : 0.8638836408180803
(Iteration 35/1000) Test RMSE:                     0.892930
Training loss : 0.8515089350824213
Training loss : 1.236073300499966
Training loss : 1.3715636757285579
Training loss : 1.0443572333235276
Training loss : 0.860208250298429
(Iteration 40/1000) Test RMSE:                     0.893269
Training loss : 0.8388994489050412
Training loss : 1.3173294754531584
Training loss : 1.3666662175047999
Training loss : 0.8870536556003747
Training loss : 0.8364603386191529
(Iteration 45/1000) Test RMSE:                     0.889765
Training loss : 0.8201107027514734
Training loss : 1.2157862845113319
Training loss : 1.3496795985855927
Training loss : 1.002764761280667
Training loss : 0.831910395941351
(Iteration 50/1000) Test RMSE:                     0.894383

You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/cloudml/zen/pull/61#issuecomment-239656456

cloudml / zen

Add VecMVM #61