Skip to content

estimate-lm segfaults when specifying a vocab file #31

Description

@GoogleCodeExporter
What steps will reproduce the problem?

Run estimate-lm -v vocab.file -t training.txt -wl lm.arpa. Where vocab.file 
contains a subset of the vocab in the LM training text.


What is the expected output? What do you see instead?

Program output:
0.000   Loading vocab vocab...
0.010   Loading corpus training.txt...
Segmentation fault


What version of the product are you using? On what operating system?

Latest SVN, ubuntu 10.04 64bit


Original issue reported on code.google.com by edohashira@gmail.com on 14 Jul 2011 at 4:34

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions