diff options
Diffstat (limited to 'doc/libpinyin.1')
| -rw-r--r-- | doc/libpinyin.1 | 38 |
1 files changed, 38 insertions, 0 deletions
diff --git a/doc/libpinyin.1 b/doc/libpinyin.1 new file mode 100644 index 0000000..b8a0261 --- /dev/null +++ b/doc/libpinyin.1 @@ -0,0 +1,38 @@ +.TH LIBPINYIN "1" "Fed 2012" "libpinyin" "User Commands" + +.SH NAME +libpinyin \- Library to deal with pinyin + +.SH DESCRIPTION +The libpinyin project aims to provide the algorithms core for intelligent sentence-based Chinese pinyin input methods. + +.SH TOOLS +gen_binary_files \- generate initially binary pinyin libraries +import_interpolation \- import libpinyin textual format model data +gen_unigram \- increase the unigram frequency for all phrases + +.SH USAGE +.HP +gen_binary_files --table-dir <DIRNAME> +.RS +.HP +.B --table-dir +Read textual format files from the <DIRNAME> directory. +.RE +.HP +import_interpolation \< <MODELFILE> +.HP +gen_unigram + +.SH EXAMPLE +Download the model.text.tar.gz, and extracts all files into a folder, then run the commands below to generate the binary model data. + +.RS +rm gb_char.bin gbk_char.bin phrase_index.bin pinyin_index.bin bigram.db + +gen_binary_files --table-dir ../data + +import_interpolation < ../data/interpolation.text + +utils/training/gen_unigram +.RE |