kenlm

kenlm copied to clipboard

lyjzsyzlt

Interpolation of different language models in kenlm

1

Is there any easy-to-use binary like 'compute-best-mix' from srilm? I've glanced through the [code](https://github.com/kpu/kenlm/blob/master/lm/builder/interpolate.cc) in kenlm but I'm not sure how can I use it.

Some-random

interrupted by signal 11: SIGSEGV after running "model = new lm::ngram::Model()" in ubuntu 18.04.5

2

model = new lm::ngram::Model(); when i run it in deepin, there no error; but i run it in ubuntu18.04.5, it report "interrupted by signal 11: SIGSEGV". who can help me?

lucky-dersan

cmake failed on CentOS Linux 7

4

When i used follow command mkdir build cd build cmake .. occurred follow error. > CMake Error at /usr/local/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:218 (message): > Could NOT find Boost (missing: Boost_INCLUDE_DIR program_options system >...

cfangpp

Segmentation fault after running lmplz

5

Hi. I've just tried to compile the lmplz and faced with the Segmentation fault. Moreover, I was facing with the errors while installation KenLM with Boost version 1.65 which actually...

OOps717

Python binding for the training

2

Add the training in the Python wrapper. Can be used like this : ``` from kenlm import lmplz lmplz('a.txt', 'b.txt', memory="500M", order=3) ```

Garvys

Hello, I'm wondering..., how to solve segmentation error in inserting `spelling_idxs, word_idx, score` to trie? the script is in [here](https://github.com/pytorch/fairseq/blob/master/examples/speech_recognition/w2l_decoder.py), line 156. what i do is described in [here](https://github.com/pytorch/fairseq/issues/2493#issuecomment-719915281). the...

wahyubram82

@kpu Hi,when i run the command `./bin/lmplz -o 2 -S 5% --verbose_header --skip_symbols --limit_vocab_file ../../fst/mandarin/wlist --intermediate ../../tmp/Oral.intermediate --text ../../tmp/Oral.dat ` it goes wrong and tell me `lm/common/joint_order.hh:62 n void lm::JointOrder(const...

xiaosimao

Best way to add kenlm to your own C++ project using Cmake?

2

digital10111

Hello, I'd like to use arpa format LM as input into the `interpolate` binary. It's probably not possible to do directly. But isn't there any workaround, e.g. reformatting the arpa...

tomassykora