berkeleyaligner - Project Hosting on Google Code

Earlier this year Sep 28, 2009 berkeleyaligner_unsupervised-2.1.tar.gz (Unsupervised aligner version 2.1) file uploaded by PapaJohn - Labels: Featured Type-Executable OpSys-All Labels: Featured Type-Executable OpSys-All Sep 28, 2009 berkeleyaligner.tar.gz (Unsupervised aligner (original) 2.1) file uploaded by PapaJohn - Labels: Featured Type-Executable OpSys-All Labels: Featured Type-Executable OpSys-All Sep 28, 2009 BerkeleyAligner Wiki page edited by PapaJohn - Revision r27 Edited wiki page through web user interface. Revision r27 Edited wiki page through web user interface. Aug 02, 2009 berkeleyaligner-2.0.tar.gz (Berkeley Aligner 2.0) file uploaded by blitzer - Labels: Featured Labels: Featured Aug 02, 2009 berkeleyaligner-2.0_source.tar.gz (Berkeley Aligner 2.0 source) file uploaded by blitzer Aug 02, 2009 berkeleyaligner-2.0_source.tar.gz (Berkeley Aligner 2.0 source) file uploaded by blitzer Aug 02, 2009 BerkeleyAligner Wiki page edited by blitzer - Revision r26 Edited wiki page through web user interface. Revision r26 Edited wiki page through web user interface. Aug 02, 2009 BerkeleyAligner Wiki page edited by blitzer - Revision r25 Edited wiki page through web user interface. Revision r25 Edited wiki page through web user interface. Aug 02, 2009 berkeleyAligner2.tar.gz (Berkeley Aligner 2.0 source) file uploaded by blitzer Aug 02, 2009 berkeleyaligner-2.0.tar.gz (Berkeley Aligner 2.0) file uploaded by PapaJohn - Labels: Featured Type-Executable OpSys-All Labels: Featured Type-Executable OpSys-All Jul 29, 2009 BerkeleyAligner Wiki page commented on by freeflying.yzhou - I want to use the aligner to compare with giza++ on speed and quality! I want to use the aligner to compare with giza++ on speed and quality! May 31, 2009 issue 2 (Different Alignment Result on the same input and setting) reported by paisarn.c - What steps will reproduce the problem? 1.) I rerun the aligner twice on the same data but I got slightly different results. I am not sure whether it is a bug of the program or not. Here is my command. java -server -Xmx1000m -ea -jar /home/paisarn/mt-util/berkeleyaligner-1.1/berkeleyaligner.jar -execDir /home/paisarn/aligner/out-enfr20-2-berkeley1.1 -englishSuffix en -foreignSuffix fr -exec.create true -Main.saveParams true -Main.alignTraining true -Main.testSources -Main.iters 5 5 -EMWordAligner.numThreads 4 -Main.trainSources /home/paisarn/aligner/enfr20k/ When I run this command as the second time, i just changed -execDir parameter. After that, I check both training.en-fr.align from both folders but they are slightly different. Could you please explain it to me whether it was normal or there was something wrong. 2. the result in training.en-fr.align generated by aligner version 1.1 seems to be swap between src and target word index. For example, in the training.en-fr.align Generated by Aligner 1.0: 6-8 3-2 4-3 2-1 7-10 5-5 Generated by Aligner 1.1: 7-5 1-2 2-3 0-1 9-6 4-4 Supposed for Aligner 1.1: 5-7 2-1 3-2 1-0 6-9 4-4 From what I understand, the result from Aligner 1.1 should be compatible with Giza++. So the word indexes should be just minus one from Aligner 1.0. What version of the product are you using? On what operating system? Currently I use Aligner 1.1 on Linux Fedora Core 9 (64bit) What steps will reproduce the problem? 1.) I rerun the aligner twice on the same data but I got slightly different results. I am not sure whether it is a bug of the program or not. Here is my command. java -server -Xmx1000m -ea -jar /home/paisarn/mt-util/berkeleyaligner-1.1/berkeleyaligner.jar -execDir /home/paisarn/aligner/out-enfr20-2-berkeley1.1 -englishSuffix en -foreignSuffix fr -exec.create true -Main.saveParams true -Main.alignTraining true -Main.testSources -Main.iters 5 5 -EMWordAligner.numThreads 4 -Main.trainSources /home/paisarn/aligner/enfr20k/ When I run this command as the second time, i just changed -execDir parameter. After that, I check both training.en-fr.align from both folders but they are slightly different. Could you please explain it to me whether it was normal or there was something wrong. 2. the result in training.en-fr.align generated by aligner version 1.1 seems to be swap between src and target word index. For example, in the training.en-fr.align Generated by Aligner 1.0: 6-8 3-2 4-3 2-1 7-10 5-5 Generated by Aligner 1.1: 7-5 1-2 2-3 0-1 9-6 4-4 Supposed for Aligner 1.1: 5-7 2-1 3-2 1-0 6-9 4-4 From what I understand, the result from Aligner 1.1 should be compatible with Giza++. So the word indexes should be just minus one from Aligner 1.0. What version of the product are you using? On what operating system? Currently I use Aligner 1.1 on Linux Fedora Core 9 (64bit) Mar 12, 2009 r24 (max sentence length option) committed by PapaJohn - max sentence length option max sentence length option Older Oct 13, 2008 issue 1 (No alignment score) commented on by pnkhanh - I have downloaded the new version of berkeley aligner you put on the website and re-try the experiments. I use the alignTraining option to align the training data. However, I still see the alignment score of all the sentence pairs in A3 file are 0. Do I have to set another parameter to make it correctly? Thank you so much for your great help and support. I have downloaded the new version of berkeley aligner you put on the website and re-try the experiments. I use the alignTraining option to align the training data. However, I still see the alignment score of all the sentence pairs in A3 file are 0. Do I have to set another parameter to make it correctly? Thank you so much for your great help and support. Oct 01, 2008 BerkeleyAligner Wiki page edited by PapaJohn Oct 01, 2008 issue 1 (No alignment score) changed by PapaJohn - Status: Accepted Owner: PapaJohn Status: Accepted Owner: PapaJohn Sep 17, 2008 issue 1 (No alignment score) reported by pnkhanh - Currently, the output alignment has no alignnment score associated with it. Would you please add this feature to Berkeley aligner Currently, the output alignment has no alignnment score associated with it. Would you please add this feature to Berkeley aligner Sep 17, 2008 r22 (Example data switched from 1-indexed to 0-indexed) committed by PapaJohn - Example data switched from 1-indexed to 0-indexed Example data switched from 1-indexed to 0-indexed Sep 17, 2008 r21 (Maximization checked for correctness & cleaned up.) committed by PapaJohn - Maximization checked for correctness & cleaned up. Maximization checked for correctness & cleaned up. Sep 17, 2008 r20 (Namespace conflicts removed between Berkeley internal softwa...) committed by PapaJohn - Namespace conflicts removed between Berkeley internal software and the Berkeley Aligner Namespace conflicts removed between Berkeley internal software and the Berkeley Aligner Sep 17, 2008 r19 (Namespace conflicts removed between Berkeley internal softwa...) committed by PapaJohn - Namespace conflicts removed between Berkeley internal software and the Berkeley Aligner Namespace conflicts removed between Berkeley internal software and the Berkeley Aligner

issue 2 (Different Alignment Result on the same input and setting) reported by paisarn.c - What steps will reproduce the problem? 1.) I rerun the aligner twice on the same data but I got slightly different results. I am not sure whether it is a bug of the program or not. Here is my command. java -server -Xmx1000m -ea -jar /home/paisarn/mt-util/berkeleyaligner-1.1/berkeleyaligner.jar -execDir /home/paisarn/aligner/out-enfr20-2-berkeley1.1 -englishSuffix en -foreignSuffix fr -exec.create true -Main.saveParams true -Main.alignTraining true -Main.testSources -Main.iters 5 5 -EMWordAligner.numThreads 4 -Main.trainSources /home/paisarn/aligner/enfr20k/ When I run this command as the second time, i just changed -execDir parameter. After that, I check both training.en-fr.align from both folders but they are slightly different. Could you please explain it to me whether it was normal or there was something wrong. 2. the result in training.en-fr.align generated by aligner version 1.1 seems to be swap between src and target word index. For example, in the training.en-fr.align Generated by Aligner 1.0: 6-8 3-2 4-3 2-1 7-10 5-5 Generated by Aligner 1.1: 7-5 1-2 2-3 0-1 9-6 4-4 Supposed for Aligner 1.1: 5-7 2-1 3-2 1-0 6-9 4-4 From what I understand, the result from Aligner 1.1 should be compatible with Giza++. So the word indexes should be just minus one from Aligner 1.0. What version of the product are you using? On what operating system? Currently I use Aligner 1.1 on Linux Fedora Core 9 (64bit)

What steps will reproduce the problem? 1.) I rerun the aligner twice on the same data but I got slightly different results. I am not sure whether it is a bug of the program or not. Here is my command. java -server -Xmx1000m -ea -jar /home/paisarn/mt-util/berkeleyaligner-1.1/berkeleyaligner.jar -execDir /home/paisarn/aligner/out-enfr20-2-berkeley1.1 -englishSuffix en -foreignSuffix fr -exec.create true -Main.saveParams true -Main.alignTraining true -Main.testSources -Main.iters 5 5 -EMWordAligner.numThreads 4 -Main.trainSources /home/paisarn/aligner/enfr20k/ When I run this command as the second time, i just changed -execDir parameter. After that, I check both training.en-fr.align from both folders but they are slightly different. Could you please explain it to me whether it was normal or there was something wrong. 2. the result in training.en-fr.align generated by aligner version 1.1 seems to be swap between src and target word index. For example, in the training.en-fr.align Generated by Aligner 1.0: 6-8 3-2 4-3 2-1 7-10 5-5 Generated by Aligner 1.1: 7-5 1-2 2-3 0-1 9-6 4-4 Supposed for Aligner 1.1: 5-7 2-1 3-2 1-0 6-9 4-4 From what I understand, the result from Aligner 1.1 should be compatible with Giza++. So the word indexes should be just minus one from Aligner 1.0. What version of the product are you using? On what operating system? Currently I use Aligner 1.1 on Linux Fedora Core 9 (64bit)

Earlier this year

Older