-
Notifications
You must be signed in to change notification settings - Fork 1.9k
/
igl_3states_1.stderr
31 lines (30 loc) · 1.19 KB
/
igl_3states_1.stderr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
creating quadratic features for pairs: UA
final_regressor = igl_1.vw
predictions = igl_1.pred
Enabling FTRL based optimization
Algorithm used: Coin Betting
ftrl_alpha = 4
ftrl_beta = 1
using no cache
Reading datafile = train-sets/igl_3states_1.dsjson
num sources = 1
Num weight bits = 18
learning rate = 0.5
initial_t = 0
power_t = 0.5
cb_type = mtr
Enabled learners: ftrl-Coin Betting, scorer-identity, csoaa_ldf-rank, cb_adf, cb_explore_adf_greedy, experimental_igl, shared_feature_merger
Input label = CB_WITH_OBSERVATIONS
Output pred = ACTION_PROBS
average since example example current current current
loss last counter weight label predict features
0.000000 0.000000 1 1.0 3:0:0.25 0:0.25 21
0.000000 0.000000 2 2.0 1:0:0.25 0:0.25 21
0.000000 0.000000 4 4.0 2:0:0.25 0:0.25 21
-0.03676 -0.07352 8 8.0 2:0:0.85 0:0.25 21
finished run
number of examples = 10
weighted example sum = 10.000000
weighted label sum = 0.000000
average loss = -0.029412
total feature number = 280