MLcomp version alpha Statistics: 1403 users, 461 datasets, 360 programs, 16577 runs (0 queued, 0 running), 0 workers

Run 14165

Creator	favre
Program	btagger
Dataset	srl-media1-fra
Task type	SequenceTagging
Created	28d17h ago

Status: Done!

Total Time	10h53m
Max Memory Usage	1181M
Domain	SequenceTagging
Learn time	9h17m
Train acc.	0.947
Train F1	0.933
Predict train time	1h13m
Test acc.	0.881
Test F1	0.849
Predict test time

Log file

===== MAIN: learn based on training data =====
=== START program1: ./run learn ../dataset2/train

Starting time: Fri Jul 15 03:39:16 UTC 2011
-----
TRAIN
-----
Opening Feature Script File: featureTemplates
PARAMETERS:
	KSOCKET = 3
	KHYPO = 3
	TAU = 0.03
	RADIUS = 40.0
	MARGIN_RATE = 48.0
	MAXROUND = 3
	SENTENCE_LIMIT_CHAR = false
	BASEFEAT = true
	EXTENDFEAT = true
	ADDITIONALFEAT = true
	DOCUMENTFEAT = false
	DOCUMENTCOLUMN = []
	BARRIERFEAT = false
	GAZETTEERS = false
	GAZETTEERS_CASESENSITIVE_MAINWORD = true
	GAZETTEERS_CASESENSITIVE_CONTEXTWORDS = false
	LEXICON = null
FORMAT:
	W,C1,C2!
ADDITIONAL TAGS:
	[C1]
ADDITIONAL FEATURES:
	W:0
	W:0,W:1
	W:0,W:1,W:2
	W:0,W:-1
	W:0,W:-1,W:-2
	W:0,C2:-1,C2:1
	W:0,C2:-1
	W:0,C2:1
	C1:0
	C1:0,C1:1
	C1:0,C1:1,C1:2
	C1:0,C1:-1
	C1:0,C1:-1,C1:-2
	C1:0,C2:-1,C2:1
	C1:0,C2:-1
	C1:0,C2:1
	C2:1
	C2:1,C2:2
	C2:-1
	C2:-1,C2:-2
		20 additional features loaded
TRAIN FILE:
	../dataset2/train
		12990 sentences loaded in this file
LOOP: 86
LOOP: 1666
LOOP: 4958
LOOP: 4990
LOOP: 5668
LOOP: 9678

OUT: Feature weights saved in: model.0.fea
LOOP: 1666
LOOP: 4990

OUT: Feature weights saved in: model.1.fea
LOOP: 1666

OUT: Feature weights saved in: model.2.fea

OUT: Tag set saved in: model.tag

Starting time: Fri Jul 15 03:39:16 UTC 2011
Ending time: Fri Jul 15 12:56:51 UTC 2011
Elapsed time: 9 hours, 17 min, 34 s
=== END program1: ./run learn ../dataset2/train --- OK [33455s]

===== MAIN: predict/evaluate on train data =====
=== START program3: ./run stripLabels ../dataset2/train ../program0/evalTrain.in
=== END program3: ./run stripLabels ../dataset2/train ../program0/evalTrain.in --- OK [2s]
=== START program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out

Starting time: Fri Jul 15 12:56:53 UTC 2011
-------
PREDICT
-------
Opening Feature Script File: featureTemplates
PARAMETERS:
	KSOCKET = 3
	KHYPO = 3
	TAU = 0.03
	RADIUS = 40.0
	MARGIN_RATE = 48.0
	MAXROUND = 3
	SENTENCE_LIMIT_CHAR = false
	BASEFEAT = true
	EXTENDFEAT = true
	ADDITIONALFEAT = true
	DOCUMENTFEAT = false
	DOCUMENTCOLUMN = []
	BARRIERFEAT = false
	GAZETTEERS = false
	GAZETTEERS_CASESENSITIVE_MAINWORD = true
	GAZETTEERS_CASESENSITIVE_CONTEXTWORDS = false
	LEXICON = null
FORMAT:
	W,C1,C2!
ADDITIONAL TAGS:
	[C1]
ADDITIONAL FEATURES:
	W:0
	W:0,W:1
	W:0,W:1,W:2
	W:0,W:-1
	W:0,W:-1,W:-2
	W:0,C2:-1,C2:1
	W:0,C2:-1
	W:0,C2:1
	C1:0
	C1:0,C1:1
	C1:0,C1:1,C1:2
	C1:0,C1:-1
	C1:0,C1:-1,C1:-2
	C1:0,C2:-1,C2:1
	C1:0,C2:-1
	C1:0,C2:1
	C2:1
	C2:1,C2:2
	C2:-1
	C2:-1,C2:-2
		20 additional features loaded
UNTAGGED FILE:
	../program0/evalTrain.in
		12990 sentences loaded in this file
FEATURE TABLE:
	model.2.fea
		610074 features loaded
		145 labels loaded

OUT: Tagged results saved in: test.outTagged.txt

Starting time: Fri Jul 15 12:56:53 UTC 2011
Ending time: Fri Jul 15 14:10:47 UTC 2011
Elapsed time: 1 hours, 13 min, 53 s
=== END program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out --- OK [4434s]
=== START program4: ./run evaluate ../dataset2/train ../program0/evalTrain.out
=== END program4: ./run evaluate ../dataset2/train ../program0/evalTrain.out --- OK [9s]

===== MAIN: predict/evaluate on test data =====
=== START program3: ./run stripLabels ../dataset2/test ../program0/evalTest.in
=== END program3: ./run stripLabels ../dataset2/test ../program0/evalTest.in --- OK [0s]
=== START program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out

Starting time: Fri Jul 15 14:10:56 UTC 2011
-------
PREDICT
-------
Opening Feature Script File: featureTemplates
PARAMETERS:
	KSOCKET = 3
	KHYPO = 3
	TAU = 0.03
	RADIUS = 40.0
	MARGIN_RATE = 48.0
	MAXROUND = 3
	SENTENCE_LIMIT_CHAR = false
	BASEFEAT = true
	EXTENDFEAT = true
	ADDITIONALFEAT = true
	DOCUMENTFEAT = false
	DOCUMENTCOLUMN = []
	BARRIERFEAT = false
	GAZETTEERS = false
	GAZETTEERS_CASESENSITIVE_MAINWORD = true
	GAZETTEERS_CASESENSITIVE_CONTEXTWORDS = false
	LEXICON = null
FORMAT:
	W,C1,C2!
ADDITIONAL TAGS:
	[C1]
ADDITIONAL FEATURES:
	W:0
	W:0,W:1
	W:0,W:1,W:2
	W:0,W:-1
	W:0,W:-1,W:-2
	W:0,C2:-1,C2:1
	W:0,C2:-1
	W:0,C2:1
	C1:0
	C1:0,C1:1
	C1:0,C1:1,C1:2
	C1:0,C1:-1
	C1:0,C1:-1,C1:-2
	C1:0,C2:-1,C2:1
	C1:0,C2:-1
	C1:0,C2:1
	C2:1
	C2:1,C2:2
	C2:-1
	C2:-1,C2:-2
		20 additional features loaded
UNTAGGED FILE:
	../program0/evalTest.in
		3524 sentences loaded in this file
FEATURE TABLE:
	model.2.fea
		610074 features loaded
		145 labels loaded

OUT: Tagged results saved in: test.outTagged.txt

Starting time: Fri Jul 15 14:10:56 UTC 2011
Ending time: Fri Jul 15 14:32:28 UTC 2011
Elapsed time: 0 hours, 21 min, 31 s
=== END program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out --- OK [1292s]
=== START program4: ./run evaluate ../dataset2/test ../program0/evalTest.out
=== END program4: ./run evaluate ../dataset2/test ../program0/evalTest.out --- OK [2s]


real	653m14.607s
user	642m45.626s
sys	1m22.161s

Run specification

• supervised-learning: Main entry for supervised learning for training and testing a program on a dataset.
  • (learner:Program) btagger: Guided learning bidirectionnal tagger (https://github.com/agesmundo/BTagger)
  • (dataset:Dataset) srl-media1-fra: Semantic role tagging on telephone dialogue transcripts. For more information: "Semantic annotation of the french media dialog corpus", Bonneau-Maynard et al., 2005.
  • (stripper:Program[Strip]) sequence-conll-utils: Validates and inspects SequenceTagging (CoNLL Shared Task) datasets.
  • (evaluator:Program[Evaluate]) sequence-conll-evaluator: Evaluator for SequenceTagging.

Results

• doTest:
  • evaluate:
    • accuracy: 0.8811
    • errorRate: 0.1509
    • f1: 0.8491
    • precision: 0.8509
    • recall: 0.8474
    • success: true
    • time: 2
  • predict:
    • success: true
    • time: 1292
  • strip:
    • success: true
    • time: 0
• doTrain:
  • evaluate:
    • accuracy: 0.9472
    • errorRate: 0.0670999999999999
    • f1: 0.9329
    • precision: 0.9278
    • recall: 0.9379
    • success: true
    • time: 9
  • predict:
    • success: true
    • time: 4434
  • strip:
    • success: true
    • time: 2
• exitCode: 0
• learn:
  • success: true
  • time: 33455
• success: true

Comments: