Python operator.iconcat方法代码示例

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def get_unique_terms():
    "retrieve all unique terms from termdata definitions"
    ts = functools.reduce(operator.iconcat, terms_by_type.values(), [])
    cs = functools.reduce(operator.iconcat, terms_by_country.values(), [])
    return set(ts + cs) 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def flatten(l):
    if isinstance(l, list):
        return functools.reduce(operator.iconcat, l, [])
    elif isinstance(l, dict):
        return flattenDict(l) 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def flatten(x: Sequence):
    """
    Flatten the provided (potentially nested) sequence

    Args:
        x (Sequence): Potentially nested sequence to flatten

    Returns:
        list: Flattened sequence
    """

    return functools.reduce(operator.iconcat, x, []) 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def test_inplace(self):
        class C(object):
            def __iadd__     (self, other): return "iadd"
            def __iand__     (self, other): return "iand"
            def __idiv__     (self, other): return "idiv"
            def __ifloordiv__(self, other): return "ifloordiv"
            def __ilshift__  (self, other): return "ilshift"
            def __imod__     (self, other): return "imod"
            def __imul__     (self, other): return "imul"
            def __ior__      (self, other): return "ior"
            def __ipow__     (self, other): return "ipow"
            def __irshift__  (self, other): return "irshift"
            def __isub__     (self, other): return "isub"
            def __itruediv__ (self, other): return "itruediv"
            def __ixor__     (self, other): return "ixor"
            def __getitem__(self, other): return 5  # so that C is a sequence
        c = C()
        self.assertEqual(operator.iadd     (c, 5), "iadd")
        self.assertEqual(operator.iand     (c, 5), "iand")
        self.assertEqual(operator.idiv     (c, 5), "idiv")
        self.assertEqual(operator.ifloordiv(c, 5), "ifloordiv")
        self.assertEqual(operator.ilshift  (c, 5), "ilshift")
        self.assertEqual(operator.imod     (c, 5), "imod")
        self.assertEqual(operator.imul     (c, 5), "imul")
        self.assertEqual(operator.ior      (c, 5), "ior")
        self.assertEqual(operator.ipow     (c, 5), "ipow")
        self.assertEqual(operator.irshift  (c, 5), "irshift")
        self.assertEqual(operator.isub     (c, 5), "isub")
        self.assertEqual(operator.itruediv (c, 5), "itruediv")
        self.assertEqual(operator.ixor     (c, 5), "ixor")
        self.assertEqual(operator.iconcat  (c, c), "iadd")
        self.assertEqual(operator.irepeat  (c, 5), "imul")
        self.assertEqual(operator.__iadd__     (c, 5), "iadd")
        self.assertEqual(operator.__iand__     (c, 5), "iand")
        self.assertEqual(operator.__idiv__     (c, 5), "idiv")
        self.assertEqual(operator.__ifloordiv__(c, 5), "ifloordiv")
        self.assertEqual(operator.__ilshift__  (c, 5), "ilshift")
        self.assertEqual(operator.__imod__     (c, 5), "imod")
        self.assertEqual(operator.__imul__     (c, 5), "imul")
        self.assertEqual(operator.__ior__      (c, 5), "ior")
        self.assertEqual(operator.__ipow__     (c, 5), "ipow")
        self.assertEqual(operator.__irshift__  (c, 5), "irshift")
        self.assertEqual(operator.__isub__     (c, 5), "isub")
        self.assertEqual(operator.__itruediv__ (c, 5), "itruediv")
        self.assertEqual(operator.__ixor__     (c, 5), "ixor")
        self.assertEqual(operator.__iconcat__  (c, c), "iadd")
        self.assertEqual(operator.__irepeat__  (c, 5), "imul") 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def rectify(dataset: str,
            xy_var_names: str = None,
            var_names: str = None,
            output_path: str = None,
            output_format: str = None,
            output_size: str = None,
            output_tile_size: str = None,
            output_point: str = None,
            output_res: float = None,
            delta: float = DEFAULT_DELTA,
            dry_run: bool = DEFAULT_DRY_RUN):
    """
    Rectify a dataset to WGS-84 using its per-pixel geo-locations.
    """

    input_path = dataset

    xy_var_names = parse_cli_sequence(xy_var_names,
                                      metavar='VARIABLES', num_items=2,
                                      item_plural_name='names')
    var_name_lists = [parse_cli_sequence(var_name_specifier,
                                         metavar='VARIABLES',
                                         item_plural_name='names')
                      for var_name_specifier in var_names]
    var_name_flat_list = functools.reduce(operator.iconcat, var_name_lists, [])

    output_size = parse_cli_sequence(output_size,
                                     metavar='SIZE', num_items=2, item_plural_name='sizes',
                                     item_parser=int, item_validator=assert_positive_int_item)
    output_tile_size = parse_cli_sequence(output_tile_size,
                                          metavar='TILE_SIZE', num_items=2, item_plural_name='tile sizes',
                                          item_parser=int, item_validator=assert_positive_int_item)
    output_point = parse_cli_sequence(output_point,
                                      metavar='POINT', num_items=2, item_plural_name='coordinates',
                                      item_parser=float)

    # noinspection PyBroadException
    _rectify(input_path,
             xy_var_names,
             None if len(var_name_flat_list) == 0 else var_name_flat_list,
             output_path,
             output_format,
             output_size,
             output_tile_size,
             output_point,
             output_res,
             delta,
             dry_run=dry_run,
             monitor=print)

    return 0 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def calculate_n(model, batch_vectorizer):
    """
    Calculate all necessary statistics from batch. This may take some time.
    """
    doc2token = {}
    for batch_id in range(len(batch_vectorizer._batches_list)):
        batch_name = batch_vectorizer._batches_list[batch_id]._filename
        batch = artm.messages.Batch()
        with open(batch_name, "rb") as f:
            batch.ParseFromString(f.read())

        for item_id in range(len(batch.item)):
            item = batch.item[item_id]
            theta_item_id = getattr(item, model.theta_columns_naming)

            doc2token[theta_item_id] = {'tokens': [], 'weights': []}
            for token_id, token_weight in zip(item.token_id, item.token_weight):
                doc2token[theta_item_id]['tokens'].append(batch.token[token_id])
                doc2token[theta_item_id]['weights'].append(token_weight)

    previous_num_document_passes = model._num_document_passes
    model._num_document_passes = 10
    ptdw = model.transform(batch_vectorizer=batch_vectorizer, theta_matrix_type='dense_ptdw')
    model._num_document_passes = previous_num_document_passes

    docs = ptdw.columns
    docs_unique = OrderedDict.fromkeys(docs).keys()

    tokens = [doc2token[doc_id]['tokens'] for doc_id in docs_unique]
    tokens = functools.reduce(operator.iconcat, tokens, [])

    ndw = np.concatenate([np.array(doc2token[doc_id]['weights']) for doc_id in docs_unique])
    ndw = np.tile(ndw, (ptdw.shape[0], 1))

    ptdw.columns = pd.MultiIndex.from_arrays([docs, tokens], names=('doc', 'token'))
    ntdw = ptdw * ndw

    ntd = ntdw.groupby(level=0, axis=1).sum()

    nwt = ntdw.groupby(level=1, axis=1).sum().T

    nt = nwt.sum(axis=0)

    return ntdw, ntd, nwt, nt 

# 需要导入模块: import operator [as 别名]
# 或者: from operator import iconcat [as 别名]
def to_dict(self, parsed_results):
        """This function standandlize the parsed results. It first reduces the
        collected parsed results into lists of tuples; then loads them as
        pandas.DataFrames whose columns represent p_keys.

        Parameters
        ----------
        parsed_results : iterable
            An iterable that contains parsed results generated by Parser.

        Returns
        ----------
        A dict, the keys of which are consistent with database tables; and
        the values of which are pandas.DataFrame.
        """
        tkeys = PMETA.keys()
        reduced_results = reduce(lambda x, y: {k: iconcat(x[k], y[k])
                                               for k in tkeys},
                                 parsed_results)
        dfs = {
            k: pd.DataFrame(reduced_results[k], columns=PMETA[k]['p_keys'])
            for k in tkeys
        }
        # drop duplicates mainly based on unique keys
        for k in tkeys:
            if k == 'full_user' or k == 'mentioned_user':
                dfs[k] = dfs[k].sort_values('updated_at', ascending=False)
            #
            # !IMPORTANT (ESPECIALLY FOR `ass_tweet` table)
            # Causion:
            # (1) The default missing values for pandas.DataFrame is
            # np.NAN, which is not compatible with SQL insertion in SQLAlchemy.
            # Thus a replace operation need to take.
            # (2) When missing values occurs, the dtype of a DataFrame would
            # be 'float' (either float32 or float64), which could truncate
            # the large numbers. Since version 24, pandas provide new data type
            # Int64 (CAPITAL I). Thus we need to convert it to this data type.
            #
            if k == 'ass_tweet':
                # replace np.NAN as None
                dfs[k] = dfs[k].astype('Int64')
                dfs[k].replace({pd.np.nan: None}, inplace=True)
            dfs[k] = dfs[k].drop_duplicates(PMETA[k]['pu_keys'], keep='first')
        return dfs