How to use larray - 10 common examples

# we do not include PyQt because it is an indirect dependency. It might
        # be a good idea to include those too, but the list will get long in
        # that case.
        print("""
python {py}
larray {la}
numexpr {ne}
numpy {np}
pandas {pd}
pytables {pt}
pyyaml {yml}
bcolz {bc}
bottleneck {bn}
matplotlib {mpl}
vitables {vt}
""".format(py=py_version, la=la.__version__, ne=ne.__version__,
           np=np.__version__, pd=pd.__version__, pt=pt.__version__,
           yml=yaml.__version__,
           bc=bcolz_version, bn=bn_version, mpl=mpl_version, vt=vt_version,
           ))
        parser.exit()

def gettype(value):
    if isinstance(value, (np.ndarray, la.LArray)):
        type_ = value.dtype.type
    elif isinstance(value, (tuple, list)):
        type_ = type(value[0])
    else:
        type_ = type(value)
    return normalize_type(type_)

% (fpath,
                           len(str_table),
                           num_possible_values,
                           ' * '.join(str(len(values))
                                      for values in possible_values)))

    # TODO: compare time with numpy built-in conversion:
    # if dtype is None, numpy tries to detect the best type itself
    # which it does a good job of if the values are already numeric values
    # if dtype is provided, numpy does a good job to convert from string
    # values.
    if celltype is None:
        celltype = detect_column_type(str_table)
    data = convert_1darray(str_table, celltype)
    array = np.array(data, dtype=celltype)
    return la.LArray(array.reshape(shape), header, possible_values)

# convert to a 1d array. We don't simply use data = np.array(data),
        # because if data is a list of ndarray (for example if we use
        # groupby(a, expr=id), *and* all the ndarrays have the same length,
        # the result is a 2d array instead of an array of ndarrays like we
        # need (at this point).
        arr = np.empty(len(data), dtype=type(data[0]))
        arr[:] = data
        data = arr

        # and reshape it
        data = data.reshape(len_pvalues)
        axes = [la.Axis(axis_labels, axis_name)
                for axis_name, axis_labels in zip(labels, possible_values)]
        # FIXME: also handle totals
        return la.LArray(data, axes)
        # return la.LArray(data, labels, possible_values,

args = [np.asarray(args)]
            else:
                length = len(args[0])
                if any(len(a) != length for a in args):
                    raise ValueError("when plotting multiple arrays, they must "
                                     "have compatible axes")
        if len(args) == 1:
            data = args[0]
            if not isinstance(data, (np.ndarray, la.LArray)):
                data = np.asarray(data)
            if ndim(data) == ndim_req:
                # move the last axis first so that the last dimension is stacked
                axes = list(range(data.ndim))
                data = data.transpose(axes[-1], *axes[:-1])
            elif ndim(data) == ndim_req - 1:
                if isinstance(data, la.LArray):
                    # add dummy axis and move it as the first axis
                    data = data.expand(la.Axis(1, '__dummy__')).transpose('__dummy__')
                else:
                    data = data[np.newaxis]
            else:
                raise dimerror
        elif all(ndim(a) == ndim_req - 1 for a in args):
            data = args
        else:
            raise dimerror
        return np.asarray(data), aslabeledarray(data).axes

def store_result(self, result, context):
        if isinstance(result, (np.ndarray, la.LArray)):
            res_type = result.dtype.type
        else:
            res_type = type(result)

        if self.temporary:
            target = self.entity.temp_variables
        else:
            # we cannot store/cache self.entity.array[self.name] because the
            # array object can change (eg when enlarging it due to births)
            target = self.entity.array
            result = np.asarray(result)
            # TODO: assert type for temporary variables too
            target_type_idx = type_to_idx[target[self.name].dtype.type]
            res_type_idx = type_to_idx[res_type]
            if res_type_idx > target_type_idx:
                raise Exception(

def prepare(self, args, kwargs):
        ndim_req = self.ndim_req
        dimerror = ValueError("%s() only works on %d or %d dimensional data"
                              % (self.funcname, ndim_req - 1, ndim_req))
        if self.check_length and len(args) > 1:
            if all(np.isscalar(a) for a in args):
                args = [np.asarray(args)]
            else:
                length = len(args[0])
                if any(len(a) != length for a in args):
                    raise ValueError("when plotting multiple arrays, they must "
                                     "have compatible axes")
        if len(args) == 1:
            data = args[0]
            if not isinstance(data, (np.ndarray, la.LArray)):
                data = np.asarray(data)
            if ndim(data) == ndim_req:
                # move the last axis first so that the last dimension is stacked
                axes = list(range(data.ndim))
                data = data.transpose(axes[-1], *axes[:-1])
            elif ndim(data) == ndim_req - 1:
                if isinstance(data, la.LArray):
                    # add dummy axis and move it as the first axis
                    data = data.expand(la.Axis(1, '__dummy__')).transpose('__dummy__')
                else:
                    data = data[np.newaxis]
            else:
                raise dimerror
        elif all(ndim(a) == ndim_req - 1 for a in args):
            data = args
        else:

def compute(self, context, a, size=None, replace=True, p=None):
        if isinstance(a, la.LArray):
            assert p is None
            outcomes_axis = a.axes['outcomes']
            outcomes = outcomes_axis.labels
            other_axes = a.axes - outcomes_axis

            if other_axes:
                a = index_array_by_variables(a, context, other_axes)
                p = np.asarray(a.transpose('outcomes'))
            else:
                p = np.asarray(a)
            a = outcomes

        if isinstance(p, (list, np.ndarray)) and len(p) and not np.isscalar(p[0]):
            assert len(p) == len(a)
            assert all(len(px) == size for px in p)
            assert len(a) >= 2

def aslabeledarray(data):
    sequence = (tuple, list)
    if isinstance(data, la.LArray):
        return data
    elif (isinstance(data, sequence) and len(data) and
          isinstance(data[0], la.LArray)):
        # XXX: use la.stack?
        # TODO: check that all arrays have the same axes
        axes = [la.Axis(len(data))] + list(data[0].axes)
        return la.LArray(data, axes)
    else:
        return la.LArray(data)

# present in the input file but where not asked for by the
                # modeller. They are not accessible anyway.
                array = add_and_drop_fields(array, global_type)
            attrs = global_data.attrs
            dim_names = getattr(attrs, 'dimensions', None)
            if dim_names is not None:
                # we serialise dim_names as a numpy array so that it is
                # stored as a native hdf type and not a pickle but we
                # prefer to work with simple lists
                # also files serialized using Python2 are "bytes" not "str"
                dim_names = [str(dim_name) for dim_name in dim_names]
                pvalues = [getattr(attrs, 'dim%d_pvalues' % i)
                           for i in range(len(dim_names))]
                axes = [la.Axis(labels, axis_name)
                        for axis_name, labels in zip(dim_names, pvalues)]
                array = la.LArray(array, axes)
            globals_data[name] = array

        input_entities = input_root.entities

        entities_tables = {}
        print(" * indexing tables")
        for ent_name, entity in entities.items():
            print("    -", ent_name, "...", end=' ')

            table = getattr(input_entities, ent_name)
            assert_valid_type(table, list(entity.fields.in_input.name_types))

            rows_per_period, id_to_rownum_per_period = \
                timed(index_table, table)
            indexed_table = IndexedTable(table, rows_per_period,
                                         id_to_rownum_per_period)