Example fail while using GPUs

[icedpanda]

I was using en_core_web_trf with GPU enabled. I changed it to en_core_web_lg since the en_core_web_trf is not supported. However, this would give me the following error while with GPU enabled. It took me a while to figure out what was wrong.

Fail while using GPU

import spacy
import concise_concepts

from spacy import displacy

spacy.require_gpu(0) # this triggered the numpy error

data = {
    "fruit": ["apple", "pear", "orange"],
    "vegetable": ["broccoli", "spinach", "tomato"],
    "meat": ["beef", "pork", "fish", "lamb"]
}

text = """
    Heat the oil in a large pan and add the Onion, celery and carrots.
    Then, cook over a medium–low heat for 10 minutes, or until softened.
    Add the courgette, garlic, red peppers and oregano and cook for 2–3 minutes.
    Later, add some oranges and chickens."""

# use any model that has internal spacy embeddings
nlp = spacy.load('en_core_web_lg')
nlp.add_pipe("concise_concepts",
    config={"data": data}
)
doc = nlp(text)

options = {"colors": {"fruit": "darkorange", "vegetable": "limegreen", "meat": "salmon"},
           "ents": ["fruit", "vegetable", "meat"]}

displacy.render(doc, style="ent", options=options)

ERROR Messasge

TypeError: Implicit conversion to a NumPy array is not allowed. Please use `.get()` to construct a NumPy array explicitly.

[davidberenstein1957]

This is probably caused because the internal Gensim.KeyedVectors model tries to use the vector arrays from the spacy.load('en_core_web_lg'), which were loaded on GPU, due to spacy.require_gpu(0).

I've got an M1 mac, which doesn't have GPU support from spaCy yet. Could you try to create a PR, given these assumptions? Even though, I assume people usually wouldn't use en_core_web_lg on GPU.

[icedpanda]

This is the full traceback. I am not familiar with how Gensim works internally or whether it supports GPU or not.

I guess it would be sufficient to just add a note that concise-concepts is not GPU-supported yet as people usually wouldn't use en_core_web_log on GPU.

TypeError                                 Traceback (most recent call last)
Cell In [1], line 23
     21 # use any model that has internal spacy embeddings
     22 nlp = spacy.load('en_core_web_lg')
---> 23 nlp.add_pipe("concise_concepts",
     24     config={"data": data}
     25 )
     26 doc = nlp(text)
     28 options = {"colors": {"fruit": "darkorange", "vegetable": "limegreen", "meat": "salmon"},
     29            "ents": ["fruit", "vegetable", "meat"]}

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/spacy/language.py:795, in Language.add_pipe(self, factory_name, name, before, after, first, last, source, config, raw_config, validate)
    787     if not self.has_factory(factory_name):
    788         err = Errors.E002.format(
    789             name=factory_name,
    790             opts=", ".join(self.factory_names),
   (...)
    793             lang_code=self.lang,
    794         )
--> 795     pipe_component = self.create_pipe(
    796         factory_name,
    797         name=name,
    798         config=config,
    799         raw_config=raw_config,
    800         validate=validate,
    801     )
    802 pipe_index = self._get_pipe_index(before, after, first, last)
    803 self._pipe_meta[name] = self.get_factory_meta(factory_name)

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/spacy/language.py:674, in Language.create_pipe(self, factory_name, name, config, raw_config, validate)
    671 cfg = {factory_name: config}
    672 # We're calling the internal _fill here to avoid constructing the
    673 # registered functions twice
--> 674 resolved = registry.resolve(cfg, validate=validate)
    675 filled = registry.fill({"cfg": cfg[factory_name]}, validate=validate)["cfg"]
    676 filled = Config(filled)

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/confection/__init__.py:728, in registry.resolve(cls, config, schema, overrides, validate)
    719 @classmethod
    720 def resolve(
    721     cls,
   (...)
    726     validate: bool = True,
    727 ) -> Dict[str, Any]:
--> 728     resolved, _ = cls._make(
    729         config, schema=schema, overrides=overrides, validate=validate, resolve=True
    730     )
    731     return resolved

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/confection/__init__.py:777, in registry._make(cls, config, schema, overrides, resolve, validate)
    775 if not is_interpolated:
    776     config = Config(orig_config).interpolate()
--> 777 filled, _, resolved = cls._fill(
    778     config, schema, validate=validate, overrides=overrides, resolve=resolve
    779 )
    780 filled = Config(filled, section_order=section_order)
    781 # Check that overrides didn't include invalid properties not in config

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/confection/__init__.py:849, in registry._fill(cls, config, schema, validate, resolve, parent, overrides)
    846     getter = cls.get(reg_name, func_name)
    847     # We don't want to try/except this and raise our own error
    848     # here, because we want the traceback if the function fails.
--> 849     getter_result = getter(*args, **kwargs)
    850 else:
    851     # We're not resolving and calling the function, so replace
    852     # the getter_result with a Promise class
    853     getter_result = Promise(
    854         registry=reg_name, name=func_name, args=args, kwargs=kwargs
    855     )

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/concise_concepts/__init__.py:36, in make_concise_concepts(nlp, name, data, topn, model_path, word_delimiter, ent_score, exclude_pos, exclude_dep, include_compound_words, case_sensitive)
      9 @Language.factory(
     10     "concise_concepts",
     11     default_config={
   (...)
     34     case_sensitive: bool,
     35 ):
---> 36     return Conceptualizer(
     37         nlp=nlp,
     38         name=name,
     39         data=data,
     40         topn=topn,
     41         model_path=model_path,
     42         word_delimiter=word_delimiter,
     43         ent_score=ent_score,
     44         exclude_pos=exclude_pos,
     45         exclude_dep=exclude_dep,
     46         include_compound_words=include_compound_words,
     47         case_sensitive=case_sensitive,
     48     )

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/concise_concepts/conceptualizer/Conceptualizer.py:84, in Conceptualizer.__init__(self, nlp, name, data, topn, model_path, word_delimiter, ent_score, exclude_pos, exclude_dep, include_compound_words, case_sensitive)
     82 else:
     83     self.match_key = "LEMMA"
---> 84 self.run()
     85 self.data_upper = {k.upper(): v for k, v in data.items()}

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/concise_concepts/conceptualizer/Conceptualizer.py:89, in Conceptualizer.run(self)
     87 def run(self):
     88     self.determine_topn()
---> 89     self.set_gensim_model()
     90     self.verify_data()
     91     self.expand_concepts()

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/concise_concepts/conceptualizer/Conceptualizer.py:159, in Conceptualizer.set_gensim_model(self)
    155     vectorList.append(vector)
    157 self.kv = KeyedVectors(self.nlp.vocab.vectors_length)
--> 159 self.kv.add_vectors(wordList, vectorList)

File ~/miniconda3/envs/mkg/lib/python3.10/site-packages/gensim/models/keyedvectors.py:586, in KeyedVectors.add_vectors(self, keys, weights, extras, replace)
    584     weights = np.array(weights).reshape(1, -1)
    585 elif isinstance(weights, list):
--> 586     weights = np.array(weights)
    587 if extras is None:
    588     extras = {}

File cupy/_core/core.pyx:1397, in cupy._core.core.ndarray.__array__()

TypeError: Implicit conversion to a NumPy array is not allowed. Please use `.get()` to construct a NumPy array explicitly.