taku910 / crfpp Goto Github PK
View Code? Open in Web Editor NEWCRF++: Yet Another CRF toolkit
License: Other
CRF++: Yet Another CRF toolkit
License: Other
AM_CONFIG_HEADER was fatal in 1.13.1 but was reverted in 1.13.2. The bugs
should still be handled for future 1.14!
Plain replacing AM_CONFIG_HEADER with AC_CONFIG_HEADERS is enough.
Original issue reported on code.google.com by [email protected]
on 2 Jul 2013 at 2:01
Hi,
I'm using CRF++ version 0.57 on two different machines and I spotted an issue.
When I learn a model with the same training set in both the machines, one
running MACOSX 10.6.8 and one Ubuntu 8.04.4, both with 32bit architecture, I
get two different results when I perform the tests.
Is this normal?
Original issue reported on code.google.com by [email protected]
on 4 Apr 2012 at 2:13
What steps will reproduce the problem?
1. tar zxf CRF++-0.56.tar.gz && cd CRF++-0.56
2. ./configure
3. make
What is the expected output? What do you see instead?
crf_learn and crf_test should be created, but there are no binaries built.
What version of the product are you using? On what operating system?
product: CRF++-0.56.tar.gz
operating system: Mac OS X 10.7.3 + Xcode 4.2.1
Please provide any additional information below.
please see the attached log file.
Original issue reported on code.google.com by [email protected]
on 5 Mar 2012 at 2:58
Attachments:
This occurs when I'm using the resulting file in another application and I forget to close it before retraining. It would be helpful if this failed with an error saying "cannot access output file XYZ", etc.
i've used python setup.py install to install crf++,but error happend,error detail as blow:
C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\cl.exe /c /nologo /Ox
/W3 /GL /DNDEBUG /MD -ID:\DevInstall\Python\Python35-32\include -ID:\DevInstall
\Python\Python35-32\include "-IC:\Program Files (x86)\Microsoft Visual Studio 14
.0\VC\INCLUDE" "-IC:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\ATLMFC
INCLUDE" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.10240.0\ucrt" "-
IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.6\include\um" "-IC:\Program File
s (x86)\Windows Kits\8.1\include\shared" "-IC:\Program Files (x86)\Windows Kits
\8.1\include\um" "-IC:\Program Files (x86)\Windows Kits\8.1\include\winrt" /EH
sc /TpCRFPP_wrap.cxx /Fobuild\temp.win32-3.5\Release\CRFPP_wrap.obj
CRFPP_wrap.cxx
CRFPP_wrap.cxx(3046): fatal error C1083: Cannot open include file: 'crfpp.h': No
such file or directory
error: command 'C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\
cl.exe' failed with exit status 2
i've solved above error by copying crfpp.h file from crfpp-master dir to crfpp-master\python dir. then continue to install crf++. but i've got the blow error:
LINK : fatal error LNK1181: cannot open input file 'crfpp.lib'
error: command 'C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\
link.exe' failed with exit status 1181
S13 記号-空白 O
さらに M9 副詞-助詞類接続 O
、 S2 記号-読点 O
出先 M1 名詞-一般 O
から M9 助詞-格助詞-一般 O
I konw first column is token,third is pos tag,fourth si NE tag.
What the second column mean?
On http://taku910.github.io/crfpp/
http://www.cis.upenn.edu/~pereira/papers/crf.pdf is a dead link.
Fixes minor typo in Template example which makes it easier to understand.
Original issue reported on code.google.com by [email protected]
on 7 Jun 2012 at 11:09
Attachments:
I noticed that the version number return from JNI is incorrect.
Did you forget to run swig?
Original issue reported on code.google.com by [email protected]
on 17 Jul 2012 at 11:30
Attachments:
The documentation has the following example:
Input: Data
He PRP B-NP
reckons VBZ B-VP
the DT B-NP << CURRENT TOKEN
current JJ I-NP
account NN I-NP
template | expanded feature |
---|---|
%x[0,0] | the |
%x[0,1] | DT |
%x[-1,0] | reckons |
%x[-2,1] | PRP |
%x[0,0]/%x[0,1] | the/DT |
ABC%x[0,1]123 | ABCDT123 |
The documentation does not state what features are generated for out-of-bounds positions. For example, what features are generated in the following situations?
He PRP B-NP << CURRENT TOKEN
reckons VBZ B-VP
the DT B-NP
current JJ I-NP
account NN I-NP
template | expanded feature |
---|---|
%x[-1,0] | ??? |
He PRP B-NP
reckons VBZ B-VP
the DT B-NP
current JJ I-NP
account NN I-NP << CURRENT TOKEN
template | expanded feature |
---|---|
%x[1,0] | ??? |
Or are these features perhaps not generated at all? I need to know if I should be adding my own BOS/EOS tokens.
where is the readme file ?
Rgs,
Ben
hi all,
I am git clone the src into my centos 7 os , and configure and make crfpp,
but the compiler told me that :
crf_learn.cpp:9:21: fatal error: winmain.h: No such file or directory
#include "winmain.h"
^
compilation terminated.
make[1]: *** [crf_learn.o] Error 1
How can I solve this issue ?
thanks
I simply follow the readme to install the python-api
when I run
python setup.py build
I got following error
running build
running build_py
running build_ext
building '_CRFPP' extension
gcc -Wno-unused-result -Wsign-compare -Wunreachable-code -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -I/Users/xxx/anaconda3/include -arch x86_64 -I/Users/xxx/anaconda3/include -arch x86_64 -I/Users/xxx/anaconda3/include/python3.7m -c CRFPP_wrap.cxx -o build/temp.macosx-10.7-x86_64-3.7/CRFPP_wrap.o
warning: include path for stdlibc++ headers not found; pass '-stdlib=libc++' on the command line to
use the libc++ standard library instead [-Wstdlibcxx-not-found]
CRFPP_wrap.cxx:2375:23: warning: explicitly assigning value of variable of type 'int' to itself
[-Wself-assign]
res = SWIG_AddCast(res);
~~~ ^ ~~~
CRFPP_wrap.cxx:2378:23: warning: explicitly assigning value of variable of type 'int' to itself
[-Wself-assign]
res = SWIG_AddCast(res);
~~~ ^ ~~~
CRFPP_wrap.cxx:2900:9: warning: variable 'res' is used uninitialized whenever 'if' condition is true
[-Wsometimes-uninitialized]
if (PyType_Ready(tp) < 0)
^~~~~~~~~~~~~~~~~~~~
CRFPP_wrap.cxx:2924:10: note: uninitialized use occurs here
return res;
^~~
CRFPP_wrap.cxx:2900:5: note: remove the 'if' if its condition is always false
if (PyType_Ready(tp) < 0)
^~~~~~~~~~~~~~~~~~~~~~~~~
CRFPP_wrap.cxx:2881:10: note: initialize the variable 'res' to silence this warning
int res;
^
= 0
CRFPP_wrap.cxx:2981:10: fatal error: 'stdexcept' file not found
#include <stdexcept>
^~~~~~~~~~~
4 warnings and 1 error generated.
error: command 'gcc' failed with exit status 1
I tried to add '-stdlib=libc++' args in the setup.py in following way,
from distutils.core import setup,Extension,os
import string
import platform
if platform.system() == "Windows":
extra_args = ["/std:c++latest", "/EHsc"]
elif platform.system() == "Darwin":
extra_args = ['-std=c++11', "-Os", "-stdlib=libc++", "-mmacosx-version-min=10.7"]
else:
extra_args = []
setup(name = "mecab-python",
py_modules=["CRFPP"],
ext_modules = [Extension("_CRFPP",
["CRFPP_wrap.cxx",],
libraries=["crfpp", "pthread"],
extra_compile_args=extra_args)
])
it did not work, I got more warnings
ld: library not found for -lstdc++
I used this command to train my data:
CRF++-0.58/crf_learn -a MIRA -f 30 -H 10 template train_data model
and it shows:
Number of sentences: 299769
Number of features: 3260280
Number of thread(s): 32
Freq: 30
eta: 0.00010
C: 1.00000
shrinking size: 10
but in training console :
iter=148 terr=0.00038 serr=0.00215 act=669 uact=135 obj=9552.22259 kkt=0.98669
iter=149 terr=0.00038 serr=0.00212 act=669 uact=126 obj=9552.24988 kkt=0.89123
iter=150 terr=0.00038 serr=0.00212 act=667 uact=127 obj=9552.24988 kkt=0.00000
iter=151 terr=0.00520 serr=0.02583 act=299769 uact=549 obj=9932.48463 kkt=17.21649
iter=152 terr=0.00358 serr=0.01923 act=299769 uact=656 obj=10176.21849 kkt=12.91036
I don't know why it continues training when kkt=0.
It seems no ends. T T
Anything wrong?
I encounter a question when reading the source code in "param.cpp" at line 205: "if (size == sizeof(ptr)) break;". I think this condition is used to check if the index size
is out of range of array ptr
, and it should be "if (size == sizeof(ptr)/sizeof(char *)) break;".
I'm not sure if my idea is right. ..
What steps will reproduce the problem?
crf_learn -t -a crf nerd.i.tok1.pattern nerd.i.tok1 nerd.i.tok1.crfpp.model
on provided files
What is the expected output? What do you see instead?
expect starting training but seg fault
What version of the product are you using? On what operating system?
0.58 (last at writing time)
Please provide any additional information below.
extract 6004 sequences and 306984496 features then seg fault.
Original issue reported on code.google.com by [email protected]
on 10 Dec 2013 at 12:28
Attachments:
i want to debug the crf++. how to create a debuging crf_learn object file. should i alter the makefile, how to do it ?thank you
The Swig wrappers can't be compiled right now because the link to CRFPP::Model::get_template
is missing. Here's what happens when you run make
in the java directory:
c++ -O3 -c -fpic CRFPP_wrap.cxx -I/usr/lib/jvm/oracle-jdk-bin-1.8/include/ -I/usr/lib/jvm/oracle-jdk-bin-1.8/include/linux
CRFPP_wrap.cxx: In function ‘_jstring* Java_org_chasen_crfpp_CRFPPJNI_Model_1getTemplate(JNIEnv*, jclass, jlong, jobject)’:
CRFPP_wrap.cxx:311:54: error: ‘const class CRFPP::Model’ has no member named ‘getTemplate’
Makefile:14: recipe for target 'all' failed
make: *** [all] Error 1
What steps will reproduce the problem?
1. load libCRFPP.so and libcrfpp.so.0
2. run java test example
What version of the product are you using? On what operating system?
0.58 Linux
Please provide any additional information below.
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007f208a4a8a49, pid=3986, tid=139778720790272
#
# JRE version: 7.0_05-b05
# Java VM: Java HotSpot(TM) 64-Bit Server VM (23.1-b03 mixed mode linux-amd64
compressed oops)
# Problematic frame:
# C [libCRFPP.so+0x3a49] Java_org_chasen_crfpp_CRFPPJNI_Tagger_1x+0x29
#
# Failed to write core dump. Core dumps have been disabled.
Original issue reported on code.google.com by [email protected]
on 16 Aug 2014 at 10:52
This is a python script, not ruby.
Original issue reported on code.google.com by pierre.francois
on 9 Jan 2013 at 7:27
Attachments:
In python, I run python test.py, but there has an error:
import _CRFPP
ImportError: libcrfpp.so.0: cannot open shared object file: No such file or directory
How can I represent category of sentence predicted from a text classifier as a feature in CRF++?
For instance, if the sentence, Tumblr merges with Yahoo
., is classified as Business
, then while composing the training file for crf, where can I indicate the label Business
as a feature? And how should then the template be modeled?
Should the train file be like this
Tumblr business ORG
merges business O
with business O
Yahoo business ORG
Or only include the category with the ORG
label? How so? And the template file?
If you could help out.
From the documents I know
unigram: |output tag| x |all possible strings expanded with a macro|
bigram: |output tag| x |output tag| x |all possible strings expanded with a macro|
Now I have configured a template like this
Unigram
U01:%x[-1,0]/%x[0,0]
in this case is it bigram or unigram ? And how many features it will generate, L * N or L * L * N?
I have learned the model via CRF++ 0.59 on Ubuntu 18.04.2,
and then test the model on my mac pro, CRF++ is 0.58
the bug is
feature_index.cpp(249) [ptr == end] model file is broken.
How to fix it, whether upgrade the version toward 0.59 is ok?
how can u explain the meaning of the following template:
U15:%x[-2,1]/%x[-1,1]
What steps will reproduce the problem?
1.
2.
3.
What is the expected output? What do you see instead?
What version of the product are you using? On what operating system?
Please provide any additional information below.
Original issue reported on code.google.com by [email protected]
on 27 Sep 2013 at 1:24
In the migration from google code to Github, the tag for version 0.58 appears to have been lost. Please could you tag the appropriate commit with the version?
Thanks
The crf model is ready, as I have trained successfully using crf++, now I want to make prediction on some short sentence using the model in our python code.
Unfortunately, I didn't find any help, any advice?
I followed the README to configure, make and install this program on my Ubuntu 14.04, and in the process of making I encountered an error reads that:
#include "winmain.h"
^
compilation terminated.
make[1]: *** [crf_learn.o] Error 1
make[1]: Leaving directory `/opt/crfpp'
make: *** [all] Error 2
and then I added this from #15 to the project root directory and can pass the make and install processes.
After installation, I tended to run it in eclipse and also tried it in my terminal but both trigger an error again:
Cannot load the example native code.
Make sure your LD_LIBRARY_PATH contains '.'
java.lang.UnsatisfiedLinkError: no CRFPP in java.library.path
I got to know that First of all we must verify that the parameter passed in the System.loadLibrary method is correct and that the library actually exists.
, so to use the native C++ code we need to load the library as shown bellow in the source code:
static {
try {
System.loadLibrary("CRFPP");
} catch (UnsatisfiedLinkError e) {
System.err.println("Cannot load the example native code.\nMake sure your LD_LIBRARY_PATH contains \'.\'\n" + e);
System.exit(1);
}
}
So according to the the above snippet there should be a CRFPP.so file? But I've not found it. Or where can I find the CRFPP library which I must add to the java.library.path
? I guess it may be caused by the later added and non-authoritive winmain.h file. Am I right?
What steps will reproduce the problem?
1. give crf_learn a bad path to the new model
2.
3.
What is the expected output? What do you see instead?
I expect an error message that says something about the bad path.
Instead I get a SIGABRT when the error message code runs through and deletes
taggers that have already been deleted.
What version of the product are you using? On what operating system?
head of the trunk as of a week ago.
Please provide any additional information below.
I fixed it locally by setting the tagger pointers to NULL after deleting them.
Then, in the error loop I check for NULL before deleting possibly a second time.
Original issue reported on code.google.com by croeder6000
on 10 May 2013 at 2:51
What steps will reproduce the problem?
download 5.8 unzip follow readme
./configure
make
What is the expected output? What do you see instead?
make
What version of the product are you using? On what operating system?
Please provide any additional information below.
mkdir m4 then make
Original issue reported on code.google.com by [email protected]
on 11 Aug 2014 at 6:52
I need to change the window size as 3, 5 and 7 to train my data. Also help me how to include suffix and prefix in template?
This is either a problem with the project setup or with documentation. After compiling the Java binding, running java test
gives this error:
Exception in thread "main" java.lang.RuntimeException: feature_index.cpp(193) [mmap_.open(model_filename)] c:\workspace\crfpp\mmap.h(111) [hFile != INVALID_HANDLE_VALUE] CreateFile() failed: ../model
at org.chasen.crfpp.CRFPPJNI.new_Tagger(Native Method)
at org.chasen.crfpp.Tagger.<init>(Tagger.java:183)
at test.main(test.java:6)
The file test.java
uses a hard-coded path, ../model
, which does not exist. The documentation in the README file in the java directory also says to use -d ../dic
, which not only does not exist, but is also ignored inside of the test file.
I've been using CRF++ quite for a while and never faced such a problem before.
Training fails on some training files. They are successfully read and the training process interrupts suddenly after "iter = 0..." appears. For some training files it helps just to launch training once again and then everything works as expected. But for some files it doesn't help. From what I can see, it happens when the amount of features is about 7 millions. Is such a behaviour somewhat OK for CRF++ or is there something wrong?
UPD. Just checked: the problem does not occur on Windows 7. I face it only with Windows 10.
I am trying to install CRF++-0.51 for an old shallow parser. It requires CRF++-0.51.
When I run the installation instructions (written in INSTALL
file), I observed that the command failed at the make step of CRF++
I tried to install CRF++-0.51 separately from the official website. I followed these commands: cd Downloads/CRF++-0.51/
--> ./configure
--> make
but I observed the same error which is as follows.
Makefile:375: recipe for target 'feature_index.lo' failed
make[1]: *** [feature_index.lo] Error 1
make[1]: Leaving directory '/home/ritwik/Downloads/CRF++-0.51'
Makefile:240: recipe for target 'all' failed
make: *** [all] Error 2
The file mentioned above is present in the directory
$ ls feature*
feature_cache.cpp feature_cache.h feature_cache.lo feature_cache.o feature.cpp feature_index.cpp feature_index.h feature_index.loT feature.lo feature.o
However CRF++-0.58 installs successfully. Unfortunately, the shallow parser is not able to run on it.
$ shallow_parser_hin < INPFILE > OUTFILE
/home/ritwik/sampark/shallow_parser_hin/bin/sl/morph/hin/morph_hin.exe: error while loading shared libraries: libglib-2.0.so.0: cannot open shared object file: No such file or directory
Can't open /tmp/sl21031.out: No such file or directory at /home/ritwik/sampark/shallow_parser_hin/bin/sl/morph/hin/addsentencetag.pl line 4.
File "/home/ritwik/sampark/shallow_parser_hin/bin/sl/guess-morph/hin/guess_morph.py", line 5
print 'Arguments missing\nPlease provide input and output files'
^
SyntaxError: Missing parentheses in call to 'print'. Did you mean print('Arguments missing\nPlease provide input and output files')?
The shallow parser issue is not related to this repository. I just want to know why there is a make error for installing CRF++-0.51.
I use a machine with 64GB memory and it has only 12GB memory. I run a test with tagger->parse() and it costs long time about 150ms. But when I run a test in machine with enough free memory, it only costs 2-3ms.
I find that most of the time spend on feature_index_->buildFeatures(this) and buildLattice() operations.
Hi, I want to aks if anybody has used crf++ in 2-dimension problems.
My main problem is understand how must be the training file format.
Thanks in advance for any help.
Original issue reported on code.google.com by [email protected]
on 1 Jul 2012 at 1:50
I’m a beginner of CRF++. A question have plagued me for many days. I’m exhausted~
In void Node::calcBeta() {
Why the code is so like calcAlpha()
You know they two are very different in original forward-backward algorithm of HMM.
What steps will reproduce the problem?
Create a tagger and delete it:
// 1st way to create a tagger:
Tagger tagger = new Tagger("-m someModel -v 3 -n2");
tagger.delete();
// 2nd way to create a tagger:
Model model = new Model("-m someModel -v 3 -n2");
Tagger tagger = model.createTagger();
tagger.delete();
What is the expected output? What do you see instead?
In both cases, the delete method should call the c++ code to destroy the tagger
from memory.
It is the case for the first exemple but not the second one.
A simple way to check it is to fill an ArrayList with taggers and then remove
all of them by calling tagger.delete:
// 1st way to create a tagger:
// List that will contain all the taggers
List<Tagger> taggers = new ArrayList<Tagger>();
for (int i = 0; i < 10000; i++) {
// create a tagger by giving the path to the model
Tagger tagger = new Tagger("-m someModel -v 3 -n2");
taggers.add(tagger);
}
// remove all objects from TaggerVector
for (Tagger currTagger : taggers) {
currTagger.delete();
}
// 2nd way to create a tagger:
// List that will contain all the taggers
List<Tagger> taggers = new ArrayList<Tagger>();
// The model used to create taggers
Model model = new Model("-m someModel -v 3 -n2");
for (int i = 0; i < 10000; i++) {
// create a tagger from the model created above
Tagger tagger = model.createTagger();
taggers.add(tagger);
}
// remove all objects from TaggerVector
for (Tagger currTagger : taggers) {
currTagger.delete();
}
Valgring(tool to check memory leak) says that the first exemple has no memory
leak but the second one generates memory leak (counted in mega bytes).
What version of the product are you using? On what operating system?
The latest version of crfpp: 0.57. Tested on OS X 10.8 and linux centos.
Please provide any additional information below.
The problem comes from the generated java code. In the class Model.java, the
method createTagger calls the constuctor of Tagger.java: Tagger(cPtr, false).
The last argument is cMemoryOwn. In our case, cMemoryOwn should be true and not
false to say that the memory will be managed on java side and not c++ side.
Indeed, we can notice that the delete method of Tagger.java checks that the
flag value is true to delete (else it does nothing): if (swigCMemOwn).
In fact SWIG makes a mistake while generating the java code of model. instead
of generating:
public Tagger createTagger() {
long cPtr = CRFPPJNI.Model_createTagger(swigCPtr, this);
return (cPtr == 0) ? null : new Tagger(cPtr, false);
}
it should generate:
public Tagger createTagger() {
long cPtr = CRFPPJNI.Model_createTagger(swigCPtr, this);
return (cPtr == 0) ? null : new Tagger(cPtr, true);
}
To sum up, the call to the tagger constructor in createTagger should be new
Tagger(cPtr, true) instead of new Tagger(cPtr, false).
Original issue reported on code.google.com by [email protected]
on 12 Oct 2012 at 3:54
ubgpu@ubgpu:/github/crfpp$ make -j/github/crfpp$
make all-am
make[1]: Entering directory /home/ubgpu/github/crfpp' /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o libcrfpp.lo libcrfpp.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o lbfgs.lo lbfgs.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o param.lo param.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o encoder.lo encoder.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o feature.lo feature.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o feature_cache.lo feature_cache.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o feature_index.lo feature_index.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o node.lo node.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o path.lo path.cpp /bin/bash ./libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o tagger.lo tagger.cpp g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o crf_learn.o crf_learn.cpp g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c -o crf_test.o crf_test.cpp crf_learn.cpp:9:21: fatal error: winmain.h: No such file or directory #include "winmain.h" ^ compilation terminated. crf_test.cpp:9:21: fatal error: winmain.h: No such file or directory #include "winmain.h" ^ compilation terminated. libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c param.cpp -fPIC -DPIC -o .libs/param.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c node.cpp -fPIC -DPIC -o .libs/node.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c tagger.cpp -fPIC -DPIC -o .libs/tagger.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature_index.cpp -fPIC -DPIC -o .libs/feature_index.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature.cpp -fPIC -DPIC -o .libs/feature.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c lbfgs.cpp -fPIC -DPIC -o .libs/lbfgs.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature_cache.cpp -fPIC -DPIC -o .libs/feature_cache.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c encoder.cpp -fPIC -DPIC -o .libs/encoder.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c libcrfpp.cpp -fPIC -DPIC -o .libs/libcrfpp.o libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c path.cpp -fPIC -DPIC -o .libs/path.o make[1]: *** [crf_test.o] Error 1 make[1]: *** Waiting for unfinished jobs.... make[1]: *** [crf_learn.o] Error 1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature_cache.cpp -o feature_cache.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c libcrfpp.cpp -o libcrfpp.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c path.cpp -o path.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c node.cpp -o node.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature.cpp -o feature.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c lbfgs.cpp -o lbfgs.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c param.cpp -o param.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c encoder.cpp -o encoder.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c feature_index.cpp -o feature_index.o >/dev/null 2>&1 libtool: compile: g++ -DHAVE_CONFIG_H -I. -O3 -Wall -c tagger.cpp -o tagger.o >/dev/null 2>&1 make[1]: Leaving directory
/home/ubgpu/github/crfpp'
make: *** [all] Error 2
ubgpu@ubgpu:
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.