msg-systems / holmes-extractor Goto Github PK

Trying to install this in Ubuntu 22.04, I get these types of errors. Any suggestions on how to fix this?
`error: command '/usr/bin/x86_64-linux-gnu-gcc' failed with exit code 1
[end of output]

    note: This error originates from a subprocess, and is likely not a problem with pip.
    ERROR: Failed building wheel for blis
    Running setup.py clean for blis
  Failed to build thinc preshed blis
  Installing collected packages: wrapt, wasabi, srsly, plac, murmurhash, cymem, wheel, tqdm, six, setuptools, preshed, numpy, Cython, thinc_gpu_ops, blis, thinc
    Running setup.py install for preshed: started
    Running setup.py install for preshed: finished with status 'error'
    error: subprocess-exited-with-error
  
    × Running setup.py install for preshed did not run successfully.
    │ exit code: 1
    ╰─> [14 lines of output]
        /home/sdspieg/.env/lib/python3.10/site-packages/setuptools/installer.py:27: SetuptoolsDeprecationWarning: setuptools.installer is deprecated. Requirements should be satisfied by a PEP 517 installer.
          warnings.warn(
        running install
        /home/sdspieg/.env/lib/python3.10/site-packages/setuptools/command/install.py:34: SetuptoolsDeprecationWarning: setup.py install is deprecated. Use build and pip and other standards-based tools.
          warnings.warn(
        running build
        running build_py
        running build_ext
        building 'preshed.maps' extension
        x86_64-linux-gnu-gcc -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O2 -Wall -g -fstack-protector-strong -Wformat -Werror=format-security -g -fwrapv -O2 -fPIC -I/usr/include/python3.10 -I/home/sdspieg/.env/include -I/usr/include/python3.10 -c preshed/maps.cpp -o build/temp.linux-x86_64-cpython-310/preshed/maps.o -O3 -Wno-strict-prototypes -Wno-unused-function
        cc1plus: warning: command-line option ‘-Wno-strict-prototypes’ is valid for C/ObjC but not for C++
        cc1plus: fatal error: preshed/maps.cpp: No such file or directory
        compilation terminated.
        error: command '/usr/bin/x86_64-linux-gnu-gcc' failed with exit code 1
        [end of output]
  
    note: This error originates from a subprocess, and is likely not a problem with pip.
  error: legacy-install-failure
  
  × Encountered error while trying to install package.
  ╰─> preshed
  
  note: This is an issue with the package mentioned above, not pip.
  hint: See above for output from the failure.
  [end of output]

`

Hi! Good morning from Singapore!

Could u kindly give examples from how you'd use the word_match.type=='entity' / 'derivation'?

Any plans to add Spanish as a language, or any ideas how I could get this to work for spanish?

@richardpaulhudson thanks for this precious contribution to the open source community, not only the library itself but the documentation and NLP insights are awesome. I was wondering if you would have any scientific literature suggestion about the following topics:

• searching over large corpora similar to what you do in the holmes demo https://holmes-demo.xt.msg.team/
• using a similar nlp-processing logic to build knowledge graphs from unstructured text

Many thanks again!

I am executing example codes from the git page. But I get a type error.

holmes_manager = holmes.Manager(model='de_core_news_lg', number_of_workers=1)

TypeError: init() got an unexpected keyword argument 'number_of_workers'

If the parameter is removed, I get OSError.

OSError: [E050] Can't find model 'de_core_news_lg'. It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory.

Am I following it correctly? Has anyone faced similar issues?

When I try to run the holmes extractor using this example, it reaches the end of all the parsing, spawns a lot of (?) worker threads during indexing, and then the program gets killed, presumably because it exceeded system resources.

What's going on, and how to fix?

Hi, @richardpaulhudson
I have tested this tool with code

import holmes_extractor as holmes
holmes_manager = holmes.Manager(model='en_core_web_lg')
holmes_manager.register_search_phrase('be all ear')
holmes_manager.start_chatbot_mode_console()

I am trying to do "phrase" search, and found out that, once the sentence contains a "be" verb, this will lead to detection of "be all ear".
I suppose this search phrase should be a full sentence, a phrase like stuff is not supported when you saying Ungrammatical expressions. (Cause I found it a little bit misleading of search phrase and search "phrase")

I am using holmes extractor with ontology-based matching and it is working fine for taxonomies of nouns (or instances of these), even for compund nouns.
The same is not working for verbs: if we have a verb like 'move' and a subclass of it like 'walk' in the ontology, then 'walk' is not recognized (if 'move' is used in the search phrase), only synonyms (owl:equivalentClass) are working.
So my questions are:

• is this the correct function or am I doing something wrong?

• and if the function is missing, are there plans to extend the taxonomy support also for verbs?

Thank you in advance.