nsaphra / amrica Goto Github PK

I am trying to use your repo, unfortunately for installing dependency I faced with an error

ERROR: Command errored out with exit status 1:
     command: 'C:\Users\moha\Anaconda3\envs\argenvv01\python.exe' -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\\Users\\moha\\AppData\\Local\\Temp\\pip-install-ev67ewnx\\argparse-config\\setup.py'"'"'; __file__='"'"'C:\\Users\\moha\\AppData\\Local\\Temp\\pip-install-ev67ewnx\\argparse-config\\setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' egg_info --egg-base 'C:\Users\moha\AppData\Local\Temp\pip-pip-egg-info-6k9vd6ut'
         cwd: C:\Users\moha\AppData\Local\Temp\pip-install-ev67ewnx\argparse-config\
    Complete output (7 lines):
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "C:\Users\moha\AppData\Local\Temp\pip-install-ev67ewnx\argparse-config\setup.py", line 2, in <module>
        from argparse_config import __version__
      File "C:\Users\moha\AppData\Local\Temp\pip-install-ev67ewnx\argparse-config\argparse_config\__init__.py", line 4, in <module>
        import ConfigParser
    ModuleNotFoundError: No module named 'ConfigParser'
    ----------------------------------------
ERROR: Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.

as the repo is for 4 years ago, is there any way that i use it?

hi, i am a coder from china, my hands can't type keyboard from last year, when i type keyboard or touch phone my figure hurt.
no doctor knows what happened.
i want to know how do you treat your hand?

I run

./disagree.py -b -s en-test.amr -t es-test.amr --align_tgt2src es-en.dict.A3.final --align_src2tgt en-es.dict.A3.final -o amr-visualize/

where

• en-test.amr is an English AMR sentence representation
• es-test.amr its Spanish counterpart
• tgt2src the Giza alignments target to source
• src2tgt the Giza alignments source to target
  -o the directory to store the image graph

and it gives me this error:

./disagree.py -b -s en-test.amr -t es-test.amr --align_tgt2src es-en.dict.A3.final --align_src2tgt en-es.dict.A3.final -o amr-visualize/
Traceback (most recent call last):
File "./disagree.py", line 314, in
xlang_main(args)
File "./disagree.py", line 242, in xlang_main
smatchgraphs = hilight_disagreement([tgt_amr], src_amr, args.num_restarts, aligner=aligner, gold_aligned_fh=gold_aligned_fh)
File "./disagree.py", line 63, in hilight_disagreement
aligner.set_amrs(a, gold_amr)
File "/home/ragerri/local/AMRICA/compare_smatch/amr_alignment.py", line 45, in set_amrs
self.get_nbest_alignments(self.src2tgt_fh), self.get_nbest_alignments(self.tgt2src_fh))
File "/home/ragerri/local/AMRICA/compare_smatch/amr_alignment.py", line 263, in align_sent2sent_union
src2tgt_align = align_sent2sent(tgt_toks, src_toks, src2tgt)
File "/home/ragerri/local/AMRICA/compare_smatch/amr_alignment.py", line 250, in align_sent2sent
tok_align[tgtind][srcind] += score
IndexError: list index out of range

I attach the 4 relevant files I use. With other alignments and amr examples it works.

es-en-sample.zip

Hi,
Thanks a lot for the AMRICA tool - it has been really helpful in comparing two AMR parses. I wanted to report a KeyError exception I got while I was using your tool. I will skip the details of what and why about the sample.amr input file.

Following is my sample input that raises KeyError when I run disagree.py:

$prompt$> cat tst/sample.amr
# ::id story-mc160.train.37_15-question-4-option-A_rte
# ::annotator story
# ::snt Ted the walrus also remembered that he had earlier built a time machine.
(x5 / remember
        :ARG0 (x1 / person
                :name (n / name
                        :op1 "Ted"
                        :op2 "the"
                        :op3 "walrus"))
        :mod (x4 / also)
        :ARG1 (x10 / build-01
                :time (x9 / early)
                :ARG1 (x13 / machine
                        :mod (x12 / time))))

# ::id story-mc160.train.37_15-question-4-option-A_rte
# ::annotator rte
# ::snt Mickey the mouse was Ted the Walrus before he was a giant walrus.
(x1 / person
        :name (n / name
                :op1 "Mickey"
                :op2 "the"
                :op3 "mouse"
                :op4 "was"
                :op5 "Ted"
                :op6 "the"
                :op7 "Walrus")
        :domain (x3 / mouse)
        :time (x8 / before
                :op1 (x13 / walrus
                        :mod (x12 / giant))))

When I run AMRICA on the above input I get following error:

$propmpt$> ./disagree.py -i tst/sample.amr -o tst/out
Traceback (most recent call last):
  File "./disagree.py", line 317, in <module>
    monolingual_main(args)
  File "./disagree.py", line 179, in monolingual_main
    amr_graphs = get_disagreement_graphs(smatchgraphs, unmatch_dead_nodes=(gold_aligned_fh == None))
  File "./disagree.py", line 90, in get_disagreement_graphs
    for (g, score) in smatchgraphs]
  File "/usr0/home/sudhi/git/AMRICA/compare_smatch/smatch_graph.py", line 58, in smatch2graph
    self.add_rel1(reln, v, const)
  File "/usr0/home/sudhi/git/AMRICA/compare_smatch/smatch_graph.py", line 154, in add_rel1
    self.unmatched_rel1[(self.gold_ind[var], const_match)].remove(reln)
KeyError: (2, u'the')

I have couple more examples of this error that I can provide. This is not a big issue for me (I am going to skip this input), but I wanted to report the issue and help with it in any way I can.

Thanks!

Hello there,
First of all thanks for this.
There is a little problem with :op order
The order is not kept in the output graph which makes reading long names very difficult.
also the ":" is removed from time concepts.