update test

test/pred-sets/ref/cats.predict

@@ -1,57 +1,57 @@
-16793.773,0.00016043647
-17118.012,0.00016043647
-17713.61,0.00016043647
-15724.643,0.00016043647
-13239.782,0.00016043647
-15331.797,0.00016043647
-12142.141,0.00016043647
-15138.831,0.00016043647
-14645.7,0.00016043647
-13393.029,0.00016043647
-17545.05,0.00016043647
-12710.977,0.00016043647
-12617.485,0.00016043647
-14503.707,0.00016043647
-15984.484,0.00016043647
-15936.399,0.00016043647
-14643.835,0.00016043647
-17257.074,0.00016043647
-17299.61,0.00016043647
-17208.654,0.00016043647
-12631.003,0.00016043647
-14921.187,0.00016043647
-12721.814,0.00016043647
-16660.594,0.00016043647
-15443.514,0.00016043647
-13486.455,0.00016043647
-12181.869,0.00016043647
-6641.853,2.1031378e-06
-15241.56,0.00016043647
-14801.344,0.00016043647
-910.76746,2.1031378e-06
-17399.625,0.00016043647
-12541.766,0.00016043647
-16820.953,0.00016043647
-12102.344,0.00016043647
-13160.788,0.00016043647
-16463.285,0.00016043647
-15094.871,0.00016043647
-17983.957,0.00016043647
-13487.721,0.00016043647
-16005.059,0.00016043647
-17514.398,0.00016043647
-14627.654,0.00016043647
-12287.391,0.00016043647
-17782.832,0.00016043647
-17394.316,0.00016043647
-16736.158,0.00016043647
-12249.621,0.00016043647
-15287.943,0.00016043647
-12830.798,0.00016043647
-17067.023,0.00016043647
-14759.012,0.00016043647
-16284.014,0.00016043647
-17850.668,0.00016043647
-17372.887,0.00016043647
-15202.805,0.00016043647
-17974.736,0.00016043647
+17540.906,0.00016043648
+15536.194,0.00016043648
+17381.867,0.00016043648
+12421.002,0.00016043648
+13341.685,0.00016043648
+15988.09,0.00016043648
+13866.5,0.00016043648
+17106.008,0.00016043648
+15123.611,0.00016043648
+15783.712,0.00016043648
+22992.766,2.1031378e-06
+7194.4995,2.1031378e-06
+15160.824,0.00016043648
+17345.219,0.00016043648
+14788.842,0.00016043648
+14527.182,0.00016043648
+16322.045,0.00016043648
+12985.183,0.00016043648
+13590.578,0.00016043648
+14753.838,0.00016043648
+17058.365,0.00016043648
+13427.974,0.00016043648
+16906.275,0.00016043648
+14033.336,0.00016043648
+12143.846,0.00016043648
+14244.833,0.00016043648
+16280.45,0.00016043648
+16732.473,0.00016043648
+14155.6875,0.00016043648
+15759.547,0.00016043648
+14163.108,0.00016043648
+14782.351,0.00016043648
+12062.178,0.00016043648
+22556.402,2.1031378e-06
+16126.391,0.00016043648
+17084.584,0.00016043648
+17965.973,0.00016043648
+12816.858,0.00016043648
+14803.017,0.00016043648
+16711.629,0.00016043648
+14877.602,0.00016043648
+15326.174,0.00016043648
+13539.921,0.00016043648
+15957.607,0.00016043648
+15291.404,0.00016043648
+13974.891,0.00016043648
+17413.562,0.00016043648
+13018.842,0.00016043648
+17922.018,0.00016043648
+17293.945,0.00016043648
+15109.656,0.00016043648
+16252.369,0.00016043648
+14785.5625,0.00016043648
+17082.445,0.00016043648
+15951.468,0.00016043648
+12971.434,0.00016043648
+12969.782,0.00016043648

test/train-sets/ref/cats-predict.stderr

@@ -15,17 +15,17 @@ Input label = CONTINUOUS
 Output pred = ACTION_PDF_VALUE
 average  since         example        example        current        current  current
 loss     last          counter         weight          label        predict features
-0.000000 0.000000            1            1.0 {185.12,0.6...     16793.77,0       10
-0.000000 0.000000            2            2.0 {772.59,0.4...     17118.01,0       10
-0.226921 0.453841            4            4.0 {14122,0.02,0}     15724.64,0       10
-0.321934 0.416948            8            8.0 {12715.1,0....     15138.83,0       10
-0.200445 0.078955           16           16.0 {669.12,0.4...      15936.4,0       10
-0.191026 0.181608           32           32.0 {10786.7,0....     17399.62,0       10
+0.000000 0.000000            1            1.0 {185.12,0.6...     17540.91,0       10
+0.000000 0.000000            2            2.0 {772.59,0.4...     15536.19,0       10
+0.226921 0.453841            4            4.0 {14122,0.02,0}        12421,0       10
+0.321934 0.416948            8            8.0 {12715.1,0....     17106.01,0       10
+0.184223 0.046511           16           16.0 {669.12,0.4...     14527.18,0       10
+0.192736 0.201250           32           32.0 {10786.7,0....     14782.35,0       10
 
 finished run
 number of examples = 57
 weighted example sum = 57.000000
 weighted label sum = 57.000000
-average loss = 0.184755
+average loss = 0.174316
 total feature number = 570
 Learn() count per node: id=0, #l=17; id=1, #l=0; id=2, #l=0;