Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/moses-smt/vowpal_wabbit.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJohn Langford <jl@hunch.net>2011-06-26 08:20:44 +0400
committerJohn Langford <jl@hunch.net>2011-06-26 08:20:44 +0400
commitafaedc47326688b011e3147b1e86aa7a8deca382 (patch)
treeb9de98fef61b3d351f986ee0263b5c94713c468a /test/test-sets/ref/0002c.stderr
parente07d7376e9333905c7346462223a351d2336b887 (diff)
a better normalized adaptive update
Diffstat (limited to 'test/test-sets/ref/0002c.stderr')
-rw-r--r--test/test-sets/ref/0002c.stderr32
1 files changed, 16 insertions, 16 deletions
diff --git a/test/test-sets/ref/0002c.stderr b/test/test-sets/ref/0002c.stderr
index 0770dbe7..110dbfae 100644
--- a/test/test-sets/ref/0002c.stderr
+++ b/test/test-sets/ref/0002c.stderr
@@ -9,27 +9,27 @@ predictions = 0002c.predict
only testing
average since example example current current current
loss last counter weight label predict features
-0.001747 0.001747 3 3.0 0.5498 0.5289 184
-0.027184 0.052621 6 6.0 0.2681 0.6451 184
-0.024323 0.020890 11 11.0 0.4315 0.5887 184
-0.034173 0.044024 22 22.0 0.5519 0.5049 184
-0.021027 0.007881 44 44.0 0.5514 0.5307 184
-0.020252 0.019460 87 87.0 0.5140 0.4891 184
-0.015007 0.009762 174 174.0 0.5596 0.5018 184
-0.012415 0.009822 348 348.0 0.5475 0.4353 184
-0.011138 0.009862 696 696.0 0.3421 0.7196 184
-0.011188 0.011238 1392 1392.0 0.4996 0.4918 184
-0.009234 0.007279 2784 2784.0 0.5090 0.4131 184
-0.008574 0.007915 5568 5568.0 0.6413 0.6159 184
-0.007983 0.007391 11135 11135.0 0.3869 0.4991 184
-0.008372 0.008762 22269 22269.0 0.5063 0.4469 184
-0.011464 0.014555 44537 44537.0 0.4905 0.4894 184
+0.001778 0.001778 3 3.0 0.5498 0.5288 184
+0.026911 0.052045 6 6.0 0.2681 0.6416 184
+0.024174 0.020889 11 11.0 0.4315 0.5869 184
+0.033938 0.043701 22 22.0 0.5519 0.4996 184
+0.021159 0.008381 44 44.0 0.5514 0.5242 184
+0.020368 0.019557 87 87.0 0.5140 0.4852 184
+0.015119 0.009871 174 174.0 0.5596 0.4972 184
+0.012454 0.009789 348 348.0 0.5475 0.4336 184
+0.011079 0.009703 696 696.0 0.3421 0.7151 184
+0.011197 0.011316 1392 1392.0 0.4996 0.4925 184
+0.009159 0.007120 2784 2784.0 0.5090 0.4137 184
+0.008413 0.007667 5568 5568.0 0.6413 0.5986 184
+0.007835 0.007257 11135 11135.0 0.3869 0.5028 184
+0.008269 0.008702 22269 22269.0 0.5063 0.4464 184
+0.011104 0.013939 44537 44537.0 0.4905 0.4881 184
finished run
number of examples = 74746
weighted example sum = 6.952e+04
weighted label sum = 3.511e+04
-average loss = 0.00979
+average loss = 0.009452
best constant = 0.5051
best constant's loss = 0.25
total feature number = 13753264