diff options
Diffstat (limited to 'dtrain/test/example')
| -rw-r--r-- | dtrain/test/example/dtrain.ini | 3 | ||||
| -rw-r--r-- | dtrain/test/example/expected-output | 143 |
2 files changed, 74 insertions, 72 deletions
diff --git a/dtrain/test/example/dtrain.ini b/dtrain/test/example/dtrain.ini index e43d6b34..c8ac7c3f 100644 --- a/dtrain/test/example/dtrain.ini +++ b/dtrain/test/example/dtrain.ini @@ -5,7 +5,7 @@ decoder_config=test/example/cdec.ini # config for cdec # weights for these features will be printed on each iteration print_weights=Glue WordPenalty LanguageModel LanguageModel_OOV PhraseModel_0 PhraseModel_1 PhraseModel_2 PhraseModel_3 PhraseModel_4 PhraseModel_5 PhraseModel_6 PassThrough tmp=/tmp -stop_after=20 # stop epoch after 20 inputs +stop_after=10 # stop epoch after 10 inputs # interesting stuff epochs=3 # run over input 3 times @@ -19,3 +19,4 @@ filter=uniq # only unique entries in kbest (surface form) pair_sampling=XYX hi_lo=0.1 # 10 vs 80 vs 10 and 80 vs 10 here pair_threshold=0 # minimum distance in BLEU (this will still only use pairs with diff > 0) +loss_margin=0 diff --git a/dtrain/test/example/expected-output b/dtrain/test/example/expected-output index 08733dd4..25d2c069 100644 --- a/dtrain/test/example/expected-output +++ b/dtrain/test/example/expected-output @@ -15,7 +15,7 @@ State is 0 bytes for feature RuleNgramFeatures feature: RuleShape (no config parameters) Example feature: Shape_S00000_T00000 State is 0 bytes for feature RuleShape -Seeding random number sequence to 380245307 +Seeding random number sequence to 1072059181 dtrain Parameters: @@ -27,6 +27,7 @@ Parameters: filter 'uniq' learning rate 0.0001 gamma 0 + loss margin 0 pairs 'XYX' hi lo 0.1 pair threshold 0 @@ -35,90 +36,90 @@ Parameters: cdec cfg 'test/example/cdec.ini' input 'test/example/nc-wmt11.1k.gz' output '-' - stop_after 20 + stop_after 10 (a dot represents 10 inputs) Iteration #1 of 3. - .. 20 -Stopping after 20 input sentences. + . 10 +Stopping after 10 input sentences. WEIGHTS - Glue = -0.1015 - WordPenalty = -0.0152 - LanguageModel = +0.21493 - LanguageModel_OOV = -0.3257 - PhraseModel_0 = -0.050844 - PhraseModel_1 = +0.25074 - PhraseModel_2 = +0.27944 - PhraseModel_3 = -0.038384 - PhraseModel_4 = -0.12041 - PhraseModel_5 = +0.1047 - PhraseModel_6 = -0.1289 - PassThrough = -0.3094 + Glue = -0.0293 + WordPenalty = +0.049075 + LanguageModel = +0.24345 + LanguageModel_OOV = -0.2029 + PhraseModel_0 = +0.0084102 + PhraseModel_1 = +0.021729 + PhraseModel_2 = +0.014922 + PhraseModel_3 = +0.104 + PhraseModel_4 = -0.14308 + PhraseModel_5 = +0.0247 + PhraseModel_6 = -0.012 + PassThrough = -0.2161 --- - 1best avg score: 0.17508 (+0.17508) - 1best avg model score: -1.2392 (-1.2392) - avg # pairs: 1329.8 - avg # rank err: 649.1 - avg # margin viol: 677.5 - non0 feature count: 874 - avg list sz: 88.6 - avg f count: 85.643 -(time 0.25 min, 0.75 s/S) + 1best avg score: 0.16872 (+0.16872) + 1best avg model score: -1.8276 (-1.8276) + avg # pairs: 1121.1 + avg # rank err: 555.6 + avg # margin viol: 0 + non0 feature count: 277 + avg list sz: 77.2 + avg f count: 90.96 +(time 0.1 min, 0.6 s/S) Iteration #2 of 3. - .. 20 + . 10 WEIGHTS - Glue = -0.0792 - WordPenalty = -0.056198 - LanguageModel = +0.31038 - LanguageModel_OOV = -0.4011 - PhraseModel_0 = +0.072188 - PhraseModel_1 = +0.11473 - PhraseModel_2 = +0.049774 - PhraseModel_3 = -0.18448 - PhraseModel_4 = -0.12092 - PhraseModel_5 = +0.1599 - PhraseModel_6 = -0.0606 - PassThrough = -0.3848 + Glue = -0.3526 + WordPenalty = +0.067576 + LanguageModel = +1.155 + LanguageModel_OOV = -0.2728 + PhraseModel_0 = -0.025529 + PhraseModel_1 = +0.095869 + PhraseModel_2 = +0.094567 + PhraseModel_3 = +0.12482 + PhraseModel_4 = -0.36533 + PhraseModel_5 = +0.1068 + PhraseModel_6 = -0.1517 + PassThrough = -0.286 --- - 1best avg score: 0.24015 (+0.065075) - 1best avg model score: -10.131 (-8.8914) - avg # pairs: 1324.7 - avg # rank err: 558.65 - avg # margin viol: 752.85 - non0 feature count: 1236 - avg list sz: 84.9 - avg f count: 88.306 -(time 0.22 min, 0.65 s/S) + 1best avg score: 0.18394 (+0.015221) + 1best avg model score: 3.205 (+5.0326) + avg # pairs: 1168.3 + avg # rank err: 594.8 + avg # margin viol: 0 + non0 feature count: 543 + avg list sz: 77.5 + avg f count: 85.916 +(time 0.083 min, 0.5 s/S) Iteration #3 of 3. - .. 20 + . 10 WEIGHTS - Glue = -0.051 - WordPenalty = -0.077956 - LanguageModel = +0.33699 - LanguageModel_OOV = -0.4726 - PhraseModel_0 = +0.040228 - PhraseModel_1 = +0.18 - PhraseModel_2 = +0.15618 - PhraseModel_3 = -0.098908 - PhraseModel_4 = -0.036555 - PhraseModel_5 = +0.1619 - PhraseModel_6 = +0.0078 - PassThrough = -0.4563 + Glue = -0.392 + WordPenalty = +0.071963 + LanguageModel = +0.81266 + LanguageModel_OOV = -0.4177 + PhraseModel_0 = -0.2649 + PhraseModel_1 = -0.17931 + PhraseModel_2 = +0.038261 + PhraseModel_3 = +0.20261 + PhraseModel_4 = -0.42621 + PhraseModel_5 = +0.3198 + PhraseModel_6 = -0.1437 + PassThrough = -0.4309 --- - 1best avg score: 0.25527 (+0.015113) - 1best avg model score: -13.906 (-3.7756) - avg # pairs: 1356.3 - avg # rank err: 562.1 - avg # margin viol: 757.35 - non0 feature count: 1482 - avg list sz: 86.65 - avg f count: 87.475 -(time 0.23 min, 0.7 s/S) + 1best avg score: 0.2962 (+0.11225) + 1best avg model score: -36.274 (-39.479) + avg # pairs: 1109.6 + avg # rank err: 515.9 + avg # margin viol: 0 + non0 feature count: 741 + avg list sz: 77 + avg f count: 88.982 +(time 0.083 min, 0.5 s/S) Writing weights file to '-' ... done --- -Best iteration: 3 [SCORE 'stupid_bleu'=0.25527]. -This took 0.7 min. +Best iteration: 3 [SCORE 'stupid_bleu'=0.2962]. +This took 0.26667 min. |