Python fasttext.load_model() Examples
The following are 17
code examples of fasttext.load_model().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
fasttext
, or try the search function
.
Example #1
Source File: evaluator.py From NLP_Toolkit with Apache License 2.0 | 6 votes |
def __init__(self): resource_package = __name__ yelp_acc_path = 'acc_yelp.bin' yelp_ppl_path = 'ppl_yelp.binary' yelp_ref0_path = 'yelp.refs.0' yelp_ref1_path = 'yelp.refs.1' yelp_acc_file = pkg_resources.resource_stream(resource_package, yelp_acc_path) yelp_ppl_file = pkg_resources.resource_stream(resource_package, yelp_ppl_path) yelp_ref0_file = pkg_resources.resource_stream(resource_package, yelp_ref0_path) yelp_ref1_file = pkg_resources.resource_stream(resource_package, yelp_ref1_path) self.yelp_ref = [] with open(yelp_ref0_file.name, 'r') as fin: self.yelp_ref.append(fin.readlines()) with open(yelp_ref1_file.name, 'r') as fin: self.yelp_ref.append(fin.readlines()) self.classifier_yelp = fasttext.load_model(yelp_acc_file.name) self.yelp_ppl_model = kenlm.Model(yelp_ppl_file.name)
Example #2
Source File: field.py From deepmatcher with BSD 3-Clause "New" or "Revised" License | 6 votes |
def cache(self, name, cache, url=None): path = os.path.join(cache, name) if not os.path.isfile(path) and url: logger.info('Downloading vectors from {}'.format(url)) if not os.path.exists(cache): os.makedirs(cache) if not os.path.isfile(self.destination): if 'drive.google.com' in url: download_from_url(url, self.destination) else: urlretrieve(url, self.destination) logger.info('Extracting vectors into {}'.format(cache)) ext = os.path.splitext(self.destination)[1][1:] if ext == 'zip': with zipfile.ZipFile(self.destination, "r") as zf: zf.extractall(cache) elif ext == 'gz': with tarfile.open(self.destination, 'r:gz') as tar: tar.extractall(path=cache) if not os.path.isfile(path): raise RuntimeError('no vectors found at {}'.format(path)) self.model = fasttext.load_model(path) self.dim = len(self['a'])
Example #3
Source File: input_embedding.py From dstc8-meta-dialog with MIT License | 5 votes |
def fasttext(self): if self._fasttext is None and self.fasttext_model_file: LOG.info("Loading fasttext embeddings from %s", self.fasttext_model_file) self._fasttext = fasttext.load_model(self.fasttext_model_file) return self._fasttext
Example #4
Source File: FastTextClassifier.py From sklearn-fasttext with BSD 3-Clause "New" or "Revised" License | 5 votes |
def loadpretrained(self,X): 'returns the model with pretrained weights' self.classifier=ft.load_model(X,label_prefix=self.lpr)
Example #5
Source File: embedding_inferable.py From intent_classifier with Apache License 2.0 | 5 votes |
def load(self, embedding_fname, embedding_url=None, *args, **kwargs): """ Method initializes dict of embeddings from file Args: fname: file name Returns: Nothing """ if not embedding_fname: raise RuntimeError('Please, provide path to model') fasttext_model_file = embedding_fname if not Path(fasttext_model_file).is_file(): emb_path = embedding_url if not emb_path: raise RuntimeError('Fasttext model file does not exist locally. URL does not contain fasttext model file') embedding_fname = Path(fasttext_model_file).name try: download(dest_file_path=fasttext_model_file, source_url=embedding_url) except Exception as e: raise RuntimeError('Looks like the `EMBEDDINGS_URL` variable is set incorrectly', e) if self.module == "fastText": import fastText self.fasttext_model = fastText.load_model(fasttext_model_file) if self.module == "fasttext": import fasttext self.fasttext_model = fasttext.load_model(fasttext_model_file) return
Example #6
Source File: embedding.py From interact with MIT License | 5 votes |
def read_embedding_df_fasttext_format(filepath): """Read embedding from fasttext format.""" model = load_fasttext(filepath) return pd.DataFrame({ word: model.get_word_vector(word) for word in model.get_words() }).T
Example #7
Source File: explainer.py From fine-grained-sentiment with MIT License | 5 votes |
def __init__(self, path_to_model: str) -> None: "Input fastText trained sentiment model" import fasttext self.classifier = fasttext.load_model(path_to_model)
Example #8
Source File: fasttext_embedder.py From DeepPavlov with Apache License 2.0 | 5 votes |
def load(self) -> None: """ Load fastText binary model from self.load_path """ log.info(f"[loading fastText embeddings from `{self.load_path}`]") self.model = fasttext.load_model(str(self.load_path)) self.dim = self.model.get_dimension()
Example #9
Source File: textfeatures.py From spice-hate_speech_detection with MIT License | 5 votes |
def load_model(self, filename=''): # Define path to the feature extractor model filename if (len(filename) > 0) and os.path.exists(filename): self.filename = filename if not os.path.exists(self.filename): print('Feature file %s does not exist' % self.filename) return -1 print('Loading model %s' % self.filename) if self.method == 'fasttext': self.model = fasttext.load_model(self.filename) elif self.method == 'bow': self.model = joblib.load(self.filename)
Example #10
Source File: textfeatures.py From spice-hate_speech_detection with MIT License | 5 votes |
def __init__(self, method='fasttext', filename=''): self.method = method self.filename = filename self.model = None # If a filename is given, try to load the model if os.path.exists(self.filename): self.load_model()
Example #11
Source File: infer.py From FARM with Apache License 2.0 | 5 votes |
def load(cls, load_dir, batch_size=4, gpu=False): import fasttext if os.path.isfile(load_dir): return cls(model=fasttext.load_model(load_dir)) else: logger.error(f"Fasttext model file does not exist at: {load_dir}")
Example #12
Source File: word_eval.py From embedding with MIT License | 5 votes |
def __init__(self, vecs_txt_fname, vecs_bin_fname=None, method="word2vec", dim=100, tokenizer_name="mecab"): self.tokenizer = get_tokenizer(tokenizer_name) self.tokenizer_name = tokenizer_name self.dim = dim self.method = method self.dictionary, self.words, self.vecs = self.load_vectors(vecs_txt_fname, method) if "fasttext" in method: self.model = load_ft_model(vecs_bin_fname)
Example #13
Source File: fasttext_classifier.py From nlp-journey with Apache License 2.0 | 5 votes |
def load(self, model_path): """ 加载训练好的模型 :param model_path: 训练好的模型路径 :return: """ if os.path.exists(self.model_path + '.bin'): return fasttext.load_model(model_path + '.bin') else: return None
Example #14
Source File: fasttext_model.py From nlp-journey with Apache License 2.0 | 5 votes |
def load(self): if os.path.exists(self.model_path): return fasttext.load_model(self.model_path) else: return None
Example #15
Source File: models.py From dostoevsky with MIT License | 5 votes |
def get_compiled_model(self): return load_fasttext_model(self.MODEL_PATH)
Example #16
Source File: classifiers.py From fine-grained-sentiment with MIT License | 4 votes |
def __init__(self, model_file: str=None) -> None: super().__init__() # pip install fasttext import fasttext try: self.model = fasttext.load_model(model_file) except ValueError: raise Exception("Please specify a valid trained FastText model file (.bin or .ftz extension)'{}'." .format(model_file))
Example #17
Source File: fastchess.py From fastchess with GNU General Public License v3.0 | 4 votes |
def __init__(self, path): ft = self.ft = fasttext.load_model(path) vectors = (ft.get_output_matrix() @ ft.get_input_matrix().T).T rows, _cols = vectors.shape # Add counts and evals vectors = np.hstack([ np.ones(rows).reshape(rows, 1), vectors]) # maybe its an occ model? self.occ = False # Start with bias. No bias for eval. bias = np.hstack([[0], vectors[0]]) # Parse remaining words piece_to_vec = defaultdict(lambda: 0) castling = {} for w, v in zip(ft.words[1:], vectors[1:]): sq = getattr(chess, w[:2].upper()) if w.endswith('-Occ'): self.occ = True for color in chess.COLORS: for piece_type in chess.PIECE_TYPES: piece_to_vec[piece_type, color, sq] += np.hstack([[0], v]) elif w.endswith('-C'): e = pst.castling[sq] castling[sq] = np.hstack([[e], v]) else: p = chess.Piece.from_symbol(w[2]) e = pst.piece[p.piece_type-1] * (1 if p.color else -1) e += pst.pst[0 if p.color else 1][p.piece_type-1][sq] #print(w[2], p, e) piece_to_vec[p.piece_type, p.color, sq] += np.hstack([[e], v]) # Convert to two-colours # We keep a record of the board from both perspectives piece_to_vec2 = {} for (piece_type, color, sq), v in piece_to_vec.items(): inv = piece_to_vec[piece_type, not color, chess.square_mirror(sq)] piece_to_vec2[piece_type, color, sq] = np.vstack([v, inv]) self.bias = np.vstack([bias, bias]) self.piece_to_vec = piece_to_vec2 self.castling = {sq: np.vstack([v, castling[chess.square_mirror(sq)]]) for sq, v in castling.items()} # Parse labels self.moves = [chess.Move.from_uci(label_uci[len('__label__'):]) for label_uci in ft.labels] # Adding 2 to the move ids, since the first entry will be the count, # and the second entry will be the evaluation self.move_to_id = {move: i + 2 for i, move in enumerate(self.moves)}