ai-se / TuneData

Transfer Learning for Parameter Tuning
0 stars 0 forks source link

new observation? tuning range #3

Closed WeiFoo closed 8 years ago

WeiFoo commented 8 years ago

CART:T: I have two different tuning results when I set the following range for threshold!!

threshold in [0.01,1]

best configurations are [0.43, 2, 5, 9, 0.55]

++++++++++++++++++++
 DataSet: antV0
++++++++++++++++++++

********** pd **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      40  ,     0 (*              |              ), 0.40,  0.40,  0.40,  0.40,  0.40
   1 ,    Tuned_CART: Y-Def ,      56  ,     0 (               |             *), 0.56,  0.56,  0.56,  0.56,  0.56

********** pf **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Tuned_CART: Y-Def ,       8  ,     0 (*              |              ), 0.08,  0.08,  0.08,  0.08,  0.08
   1 ,    Naive_CART: Y-Def ,      27  ,     0 (               |             *), 0.27,  0.27,  0.27,  0.27,  0.27

********** prec **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      15  ,     0 (*              |              ), 0.15,  0.15,  0.15,  0.15,  0.15
   1 ,    Tuned_CART: Y-Def ,      45  ,     0 (               |             *), 0.45,  0.45,  0.45,  0.45,  0.45

********** f **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      22  ,     0 (*              |              ), 0.22,  0.22,  0.22,  0.22,  0.22
   1 ,    Tuned_CART: Y-Def ,      50  ,     0 (               |             *), 0.50,  0.50,  0.50,  0.50,  0.50

********** g **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      51  ,     0 (*              |              ), 0.51,  0.51,  0.51,  0.51,  0.51
   1 ,    Tuned_CART: Y-Def ,      69  ,     0 (               |             *), 0.69,  0.69,  0.69,  0.69,  0.69

 In terms of 2 : the times of better tuners are{}

threshold in [0.1,1]

best configurations are [0.84, 22, 9, 1, 0.58]


++++++++++++++++++++
 DataSet: antV0
++++++++++++++++++++

********** pd **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Tuned_CART: Y-Def ,      37  ,     0 (*              |              ), 0.37,  0.37,  0.37,  0.37,  0.37
   1 ,    Naive_CART: Y-Def ,      40  ,     0 (               |             *), 0.40,  0.40,  0.40,  0.40,  0.40

********** pf **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Tuned_CART: Y-Def ,      13  ,     0 (*              |              ), 0.13,  0.13,  0.13,  0.13,  0.13
   1 ,    Naive_CART: Y-Def ,      27  ,     0 (               |             *), 0.27,  0.27,  0.27,  0.27,  0.27

********** prec **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      15  ,     0 (*              |              ), 0.15,  0.15,  0.15,  0.15,  0.15
   1 ,    Tuned_CART: Y-Def ,      25  ,     0 (               |             *), 0.25,  0.25,  0.25,  0.25,  0.25

********** f **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      22  ,     0 (*              |              ), 0.22,  0.22,  0.22,  0.22,  0.22
   1 ,    Tuned_CART: Y-Def ,      30  ,     0 (               |             *), 0.30,  0.30,  0.30,  0.30,  0.30

********** g **********

rank ,                 name ,    med   ,  iqr 
----------------------------------------------------
   1 ,    Naive_CART: Y-Def ,      51  ,     0 (*              |              ), 0.51,  0.51,  0.51,  0.51,  0.51
   1 ,    Tuned_CART: Y-Def ,      52  ,     0 (               |             *), 0.52,  0.52,  0.52,  0.52,  0.52