import numpy as np
#大写的是常量
dir(np)
['ALLOW_THREADS',
'BUFSIZE',
'CLIP',
'DataSource',
'ERR_CALL',
'ERR_DEFAULT',
'ERR_IGNORE',
'ERR_LOG',
'ERR_PRINT',
'ERR_RAISE',
'ERR_WARN',
'FLOATING_POINT_SUPPORT',
'FPE_DIVIDEBYZERO',
'FPE_INVALID',
'FPE_OVERFLOW',
'FPE_UNDERFLOW',
'False_',
'Inf',
'Infinity',
'MAXDIMS',
'MAY_SHARE_BOUNDS',
'MAY_SHARE_EXACT',
'NAN',
'NINF',
'NZERO',
'NaN',
'PINF',
'PZERO',
'RAISE',
'RankWarning',
'SHIFT_DIVIDEBYZERO',
'SHIFT_INVALID',
'SHIFT_OVERFLOW',
'SHIFT_UNDERFLOW',
'ScalarType',
'True_',
'UFUNC_BUFSIZE_DEFAULT',
'UFUNC_PYVALS_NAME',
'WRAP',
'_CopyMode',
'_NoValue',
'_UFUNC_API',
'__NUMPY_SETUP__',
'__all__',
'__builtins__',
'__cached__',
'__config__',
'__deprecated_attrs__',
'__dir__',
'__doc__',
'__expired_functions__',
'__file__',
'__former_attrs__',
'__future_scalars__',
'__getattr__',
'__loader__',
'__name__',
'__package__',
'__path__',
'__spec__',
'__version__',
'_add_newdoc_ufunc',
'_builtins',
'_distributor_init',
'_financial_names',
'_get_promotion_state',
'_globals',
'_int_extended_msg',
'_mat',
'_no_nep50_warning',
'_pyinstaller_hooks_dir',
'_pytesttester',
'_set_promotion_state',
'_specific_msg',
'_typing',
'_using_numpy2_behavior',
'_utils',
'abs',
'absolute',
'add',
'add_docstring',
'add_newdoc',
'add_newdoc_ufunc',
'all',
'allclose',
'alltrue',
'amax',
'amin',
'angle',
'any',
'append',
'apply_along_axis',
'apply_over_axes',
'arange',
'arccos',
'arccosh',
'arcsin',
'arcsinh',
'arctan',
'arctan2',
'arctanh',
'argmax',
'argmin',
'argpartition',
'argsort',
'argwhere',
'around',
'array',
'array2string',
'array_equal',
'array_equiv',
'array_repr',
'array_split',
'array_str',
'asanyarray',
'asarray',
'asarray_chkfinite',
'ascontiguousarray',
'asfarray',
'asfortranarray',
'asmatrix',
'atleast_1d',
'atleast_2d',
'atleast_3d',
'average',
'bartlett',
'base_repr',
'binary_repr',
'bincount',
'bitwise_and',
'bitwise_not',
'bitwise_or',
'bitwise_xor',
'blackman',
'block',
'bmat',
'bool_',
'broadcast',
'broadcast_arrays',
'broadcast_shapes',
'broadcast_to',
'busday_count',
'busday_offset',
'busdaycalendar',
'byte',
'byte_bounds',
'bytes_',
'c_',
'can_cast',
'cast',
'cbrt',
'cdouble',
'ceil',
'cfloat',
'char',
'character',
'chararray',
'choose',
'clip',
'clongdouble',
'clongfloat',
'column_stack',
'common_type',
'compare_chararrays',
'compat',
'complex128',
'complex256',
'complex64',
'complex_',
'complexfloating',
'compress',
'concatenate',
'conj',
'conjugate',
'convolve',
'copy',
'copysign',
'copyto',
'corrcoef',
'correlate',
'cos',
'cosh',
'count_nonzero',
'cov',
'cross',
'csingle',
'ctypeslib',
'cumprod',
'cumproduct',
'cumsum',
'datetime64',
'datetime_as_string',
'datetime_data',
'deg2rad',
'degrees',
'delete',
'deprecate',
'deprecate_with_doc',
'diag',
'diag_indices',
'diag_indices_from',
'diagflat',
'diagonal',
'diff',
'digitize',
'disp',
'divide',
'divmod',
'dot',
'double',
'dsplit',
'dstack',
'dtype',
'dtypes',
'e',
'ediff1d',
'einsum',
'einsum_path',
'emath',
'empty',
'empty_like',
'equal',
'errstate',
'euler_gamma',
'exceptions',
'exp',
'exp2',
'expand_dims',
'expm1',
'extract',
'eye',
'fabs',
'fastCopyAndTranspose',
'fft',
'fill_diagonal',
'find_common_type',
'finfo',
'fix',
'flatiter',
'flatnonzero',
'flexible',
'flip',
'fliplr',
'flipud',
'float128',
'float16',
'float32',
'float64',
'float_',
'float_power',
'floating',
'floor',
'floor_divide',
'fmax',
'fmin',
'fmod',
'format_float_positional',
'format_float_scientific',
'format_parser',
'frexp',
'from_dlpack',
'frombuffer',
'fromfile',
'fromfunction',
'fromiter',
'frompyfunc',
'fromregex',
'fromstring',
'full',
'full_like',
'gcd',
'generic',
'genfromtxt',
'geomspace',
'get_array_wrap',
'get_include',
'get_printoptions',
'getbufsize',
'geterr',
'geterrcall',
'geterrobj',
'gradient',
'greater',
'greater_equal',
'half',
'hamming',
'hanning',
'heaviside',
'histogram',
'histogram2d',
'histogram_bin_edges',
'histogramdd',
'hsplit',
'hstack',
'hypot',
'i0',
'identity',
'iinfo',
'imag',
'in1d',
'index_exp',
'indices',
'inexact',
'inf',
'info',
'infty',
'inner',
'insert',
'int16',
'int32',
'int64',
'int8',
'int_',
'intc',
'integer',
'interp',
'intersect1d',
'intp',
'invert',
'is_busday',
'isclose',
'iscomplex',
'iscomplexobj',
'isfinite',
'isfortran',
'isin',
'isinf',
'isnan',
'isnat',
'isneginf',
'isposinf',
'isreal',
'isrealobj',
'isscalar',
'issctype',
'issubclass_',
'issubdtype',
'issubsctype',
'iterable',
'ix_',
'kaiser',
'kernel_version',
'kron',
'lcm',
'ldexp',
'left_shift',
'less',
'less_equal',
'lexsort',
'lib',
'linalg',
'linspace',
'little_endian',
'load',
'loadtxt',
'log',
'log10',
'log1p',
'log2',
'logaddexp',
'logaddexp2',
'logical_and',
'logical_not',
'logical_or',
'logical_xor',
'logspace',
'longcomplex',
'longdouble',
'longfloat',
'longlong',
'lookfor',
'ma',
'mask_indices',
'mat',
'matmul',
'matrix',
'max',
'maximum',
'maximum_sctype',
'may_share_memory',
'mean',
'median',
'memmap',
'meshgrid',
'mgrid',
'min',
'min_scalar_type',
'minimum',
'mintypecode',
'mod',
'modf',
'moveaxis',
'msort',
'multiply',
'nan',
'nan_to_num',
'nanargmax',
'nanargmin',
'nancumprod',
'nancumsum',
'nanmax',
'nanmean',
'nanmedian',
'nanmin',
'nanpercentile',
'nanprod',
'nanquantile',
'nanstd',
'nansum',
'nanvar',
'nbytes',
'ndarray',
'ndenumerate',
'ndim',
'ndindex',
'nditer',
'negative',
'nested_iters',
'newaxis',
'nextafter',
'nonzero',
'not_equal',
'numarray',
'number',
'obj2sctype',
'object_',
'ogrid',
'oldnumeric',
'ones',
'ones_like',
'outer',
'packbits',
'pad',
'partition',
'percentile',
'pi',
'piecewise',
'place',
'poly',
'poly1d',
'polyadd',
'polyder',
'polydiv',
'polyfit',
'polyint',
'polymul',
'polynomial',
'polysub',
'polyval',
'positive',
'power',
'printoptions',
'prod',
'product',
'promote_types',
'ptp',
'put',
'put_along_axis',
'putmask',
'quantile',
'r_',
'rad2deg',
'radians',
'random',
'ravel',
'ravel_multi_index',
'real',
'real_if_close',
'rec',
'recarray',
'recfromcsv',
'recfromtxt',
'reciprocal',
'record',
'remainder',
'repeat',
'require',
'reshape',
'resize',
'result_type',
'right_shift',
'rint',
'roll',
'rollaxis',
'roots',
'rot90',
'round',
'round_',
'row_stack',
's_',
'safe_eval',
'save',
'savetxt',
'savez',
'savez_compressed',
'sctype2char',
'sctypeDict',
'sctypes',
'searchsorted',
'select',
'set_numeric_ops',
'set_printoptions',
'set_string_function',
'setbufsize',
'setdiff1d',
'seterr',
'seterrcall',
'seterrobj',
'setxor1d',
'shape',
'shares_memory',
'short',
'show_config',
'show_runtime',
'sign',
'signbit',
'signedinteger',
'sin',
'sinc',
'single',
'singlecomplex',
'sinh',
'size',
'sometrue',
'sort',
'sort_complex',
'source',
'spacing',
'split',
'sqrt',
'square',
'squeeze',
'stack',
'std',
'str_',
'string_',
'subtract',
'sum',
'swapaxes',
'take',
'take_along_axis',
'tan',
'tanh',
'tensordot',
'test',
'testing',
'tile',
'timedelta64',
'trace',
'tracemalloc_domain',
'transpose',
'trapz',
'tri',
'tril',
'tril_indices',
'tril_indices_from',
'trim_zeros',
'triu',
'triu_indices',
'triu_indices_from',
'true_divide',
'trunc',
'typecodes',
'typename',
'ubyte',
'ufunc',
'uint',
'uint16',
'uint32',
'uint64',
'uint8',
'uintc',
'uintp',
'ulonglong',
'unicode_',
'union1d',
'unique',
'unpackbits',
'unravel_index',
'unsignedinteger',
'unwrap',
'ushort',
'vander',
'var',
'vdot',
'vectorize',
'version',
'void',
'vsplit',
'vstack',
'where',
'who',
'zeros',
'zeros_like']
len(dir(np))
594
|
'abs',
'absolute',
'add',
'add_docstring',
'add_newdoc',
'add_newdoc_ufunc',
'all',
'allclose',
'alltrue',
'amax',
'amin',
'angle',
'any',
'append',
'apply_along_axis',
'apply_over_axes',
'arange',
'arccos',
'arccosh',
'arcsin',
'arcsinh',
'arctan',
'arctan2',
'arctanh',
'argmax',
'argmin',
'argpartition',
'argsort',
'argwhere',
'around',
'array',
'array2string',
'array_equal',
'array_equiv',
'array_repr',
'array_split',
'array_str',
'asanyarray',
'asarray',
'asarray_chkfinite',
'ascontiguousarray',
'asfarray',
'asfortranarray',
'asmatrix',
'atleast_1d',
'atleast_2d',
'atleast_3d',
'average',
|
'bartlett', 'base_repr', 'binary_repr', 'bincount', 'bitwise_and', 'bitwise_not', 'bitwise_or', 'bitwise_xor', 'blackman', 'block', 'bmat', 'bool_', 'broadcast', 'broadcast_arrays', 'broadcast_shapes', 'broadcast_to', 'busday_count', 'busday_offset', 'busdaycalendar', 'byte', 'byte_bounds', 'bytes_', |
|
|
|
|
import numpy as np
print(np.ones(3)) # array([1., 1., 1.])
print(np.ones([3]))# array([1., 1., 1.])
print(np.zeros(3)) # array([0., 0., 0.])
print(np.empty([3,5])) # empty的值没有意义，但速度最快，连初始化都不需要
print(np.eye(N=3)) # 单位矩阵E
"""
array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]])
"""
# 任何矩阵与单位矩阵相乘后还是本身
print(np.linspace(1,15,15).reshape(3,5)@np.eye(5))
"""
array([[ 1., 2., 3., 4., 5.],
[ 6., 7., 8., 9., 10.],
[11., 12., 13., 14., 15.]])
"""
np.unique
from sklearn.datasets import make_classification
X,y = make_classification(n_samples=10000,n_features=2000,n_classes=2,)
import numpy as np
unique,count=np.unique(y,return_counts=True)
data_count=dict(zip(unique,count))
print(data_count)
{0: 4999, 1: 5001}
python collections.Counter
import collections
data_count2=collections.Counter(y)
data_count2
Counter({0: 4999, 1: 5001})
python list count
a = [1,2,2,3,3,3]
b = set(a)
for v in b:
print(v,":",a.count(v))
1 : 1
2 : 2
3 : 3
import numpy as np np.allclose(1e-8,0)
python有range生成一系列的数，numpy有arange,开头的a指np.array
arange([start,] stop[, step,], dtype=None, *, like=None) import numpy as np print(range(3)) # range(0, 3) print(np.array([0,1,2])) # array([0, 1, 2]) print(np.array(range(3))) # array([0, 1, 2]) print(np.arange(3)) # array([0, 1, 2])
#一个参数 默认起点0，步长为1 输出：[0 1 2] a = np.arange(3) #两个参数 默认步长为1 输出[3 4 5 6 7 8] a = np.arange(3, 9) #三个参数 起点为0，终点为3，步长为0.1 输出[ 0. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1. 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2. 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9] a = np.arange(0, 3, 0.1)
指定范围内取N个数，线性分布
## linspace
从1开始,到100结束(包含100),取10个浮点数
- 近似于等差数列
a1 = np.linspace(1, 100, 10)
print(type(a1))
print(len(a1))
print(a1)
class 'numpy.ndarray'
10
[ 1. 12. 23. 34. 45. 56. 67. 78. 89. 100.]
- 不指定数量时，默认生成50个
a1 = np.linspace(1,2)
print(type(a1))
print(len(a1))
print(a1)
class 'numpy.ndarray'
50
[1. 1.02040816 1.04081633 1.06122449 1.08163265 1.10204082
1.12244898 1.14285714 1.16326531 1.18367347 1.20408163 1.2244898
1.24489796 1.26530612 1.28571429 1.30612245 1.32653061 1.34693878
1.36734694 1.3877551 1.40816327 1.42857143 1.44897959 1.46938776
1.48979592 1.51020408 1.53061224 1.55102041 1.57142857 1.59183673
1.6122449 1.63265306 1.65306122 1.67346939 1.69387755 1.71428571
1.73469388 1.75510204 1.7755102 1.79591837 1.81632653 1.83673469
1.85714286 1.87755102 1.89795918 1.91836735 1.93877551 1.95918367
1.97959184 2. ]
import numpy as np
ar1 = np.linspace(0.001,0.999,10)
print(ar1)
[0.001 0.11188889 0.22277778 0.33366667 0.44455556 0.55544444
0.66633333 0.77722222 0.88811111 0.999 ]
指定范围内取N个数，均匀分布，随机
uniform(low=0.0, high=1.0, size=None)
[low, high) (includes low, but excludes high)
从半开区间，均匀分布的数据中，随机取出size个数据
np.random.uniform(low=-1,high=1,size=10)
array([-0.60967645, -0.40023626, 0.83646232, 0.93581929, 0.50901503,
-0.86735182, 0.64710941, 0.47334668, 0.77855487, -0.86533817])
round(n):保留n位小数
import numpy as np a=np.array([1.001,1.123]) a=a.mean() # 1.0619999999999998 numpy.float64 a.round(3) # 1.062
|
np.set_printoptions(precision=None, threshold=None, linewidth=None, suppress=None, formatter=None)
np.set_printoptions()
用于控制Python中小数的显示精度。
np.set_printoptions(precision=None, threshold=None, linewidth=None, suppress=None, formatter=None)
1.precision：控制输出结果的精度(即小数点后的位数)，默认值为8
2.threshold：当数组元素总数过大时，设置显示的数字位数，其余用省略号代替(当数组元素总数大于设置值，控制输出值得个数为6个，当数组元素小于或者等于设置值得时候，全部显示)，当设置值为sys.maxsize(需要导入sys库)，则会输出所有元素
3.linewidth：每行字符的数目，其余的数值会换到下一行
4.suppress：小数是否需要以科学计数法的形式输出
5.formatter：自定义输出规则
|
import numpy as np
np.set_printoptions(precision=4)
val=[1.2345678]
np.array(val)
array([1.2346])
注意四位有效小数，最后一位是6,6进5+1,相当于四舍五入了
|
import numpy as np np.set_printoptions(precision=5,suppress=False) val=[1.2345678] np.array(val) array([1.23457]) |
|
threshold默认不低于3
import numpy as np
np.set_printoptions(precision=3,suppress=False,threshold=2)
val1=[1,2,3,4,5,6,7,8,9,0]
np.array(val1)
array([1, 2, 3, ..., 8, 9, 0])
np.random.randn(32,12)
array([[-0.4794, 1.7589, 0.3096, ..., 1.889 , -0.1393, -0.7241],
[-0.0544, -0.484 , -1.5246, ..., -0.8761, 1.2215, 1.8836],
[ 0.8409, -1.1877, -1.886 , ..., -0.14 , -0.7232, -1.0742],
...,
[ 1.3005, -1.4814, 1.0046, ..., 1.7119, 0.3857, -1.099 ],
[ 1.1772, -1.6911, 0.5059, ..., 0.4867, -0.1444, 0.3945],
[-0.1769, -1.4905, 0.025 , ..., 0.267 , 0.2176, -1.0928]])
如果不设置threshold
import numpy as np
# np.set_printoptions(precision=4,suppress=False,threshold=2)
val1=[1,2,3,4,5,6,7,8,9,0]
np.array(val1)
array([1, 2, 3, 4, 5, 6, 7, 8, 9, 0])
np.random.randn(32,12)
array([[ 1.31096198, 0.78054052, -0.40963379, 0.23273274, -0.10150605,
1.88460194, 0.35433371, 0.58183019, -0.57228141, -0.77110854,
1.03263305, -0.89838403],
[ 0.84391413, 0.78073747, 1.04194755, -0.44456158, 2.12592094,
-0.75856683, 0.24688149, -1.25829492, 0.77219029, -1.85642349,
1.01285344, 2.06686249],
[-0.47637921, -0.21378507, 1.47798267, 0.40207861, -0.9119715 ,
-0.33286264, 0.12025972, 0.97894507, -0.73496818, -0.23886736,
0.25050232, 0.50630583],
[ 1.15780458, -0.08534266, 0.28685981, -0.30310841, -1.14923578,
0.45098689, -0.801712 , 0.73749301, 0.95536514, -1.3927399 ,
-0.56234157, -0.11920498],
[ 0.77520323, 0.20221095, 0.77120729, -0.0667725 , -0.15135993,
0.2243866 , -1.71608205, 2.16473421, -1.20444339, -0.61015631,
0.22587885, -0.71552811],
[-1.44029834, 0.8743934 , 1.2970308 , 0.92969892, 0.82170051,
-1.82683443, -0.97449604, -0.56134529, -1.1953847 , 0.09939801,
0.04745533, -0.33863088],
[-1.98972731, -0.55352395, -0.03600633, 0.88468785, -0.67922446,
-0.28280658, -0.21982757, 0.81073329, -0.437203 , 0.01027904,
0.38735405, 1.18708735],
[ 0.8603094 , -0.08396988, -0.73883093, -0.79855211, 1.00562191,
0.31154704, 0.03984569, 0.91169203, -2.22721951, -0.67505607,
0.12693552, 1.02194256],
[-0.21013117, 0.19916637, 0.29599217, 1.18469863, -0.51609669,
-0.80919176, -0.18930233, -0.57996685, -0.75708229, 0.57157711,
0.98432768, 1.75348857],
[ 0.41805862, 0.05898554, -1.58387403, -0.21579417, -1.22139546,
0.91596961, -0.81760997, 1.46059549, -0.1663579 , 0.18273167,
-0.90985894, 0.76214167],
[ 0.76408739, -0.79867742, 1.61184349, -0.50377978, -0.55804087,
0.57939787, 0.87574714, 1.29942416, -1.97320811, -0.49030193,
0.53141084, 0.35189542],
[-0.37105578, 0.10168585, -0.16420312, 0.96103647, -0.59469692,
-0.49589185, 1.58945027, 0.92076437, 0.92450181, -1.21256344,
-2.03462813, 1.08089276],
[-0.93407269, -0.47150967, -0.87131997, -1.91801268, -1.14335032,
0.98411407, -0.35696701, 0.20963831, 1.81055827, -0.38785915,
-0.25418608, 2.07452367],
[-0.49608547, 0.33450767, -1.3014056 , -1.52946272, 1.32558334,
1.41568975, 1.41339455, 0.80110783, 0.4616043 , 0.64017306,
-0.31558792, 1.78494037],
[ 0.67592165, 0.90067825, 0.07187182, -1.20817453, 0.50172272,
1.18440157, 0.80802945, -0.70223411, 0.15666584, 0.67119167,
0.25968969, 0.08471809],
[ 1.55923875, -2.03317548, -1.62797668, 0.32402003, 0.70732674,
-0.44145688, -0.26678467, 0.60273043, -0.14171507, 1.47339477,
1.29916823, 0.41381557],
[ 0.67091644, -1.13449214, 0.88084109, -0.86644771, 0.9574995 ,
-0.49179203, 0.59506458, 1.68567831, 1.45062971, 1.12630867,
-0.49457516, 0.09114502],
[ 0.3225183 , 0.47806111, -2.36724118, 0.93357899, -0.89987033,
-0.06345555, -1.83587324, 1.12663143, -1.47788753, 0.7524638 ,
1.07597296, 0.45719949],
[-2.06272979, 0.77196334, 0.09220272, 0.28972571, 0.15619067,
-1.07744337, 0.41897236, 0.29222976, 2.94242355, -0.61419158,
-0.19096359, -1.36344336],
[ 0.97305062, 1.10512088, -1.1480367 , 0.71308534, 0.33248084,
-0.9166076 , -0.29528799, -1.42392493, -0.04919407, -0.01888874,
0.08178188, -0.28983312],
[ 1.22496706, -0.47912479, 0.44151834, -1.01027493, 1.05074605,
-1.73803353, -2.9378151 , -1.03139439, -0.91806935, -1.58093067,
-0.12704403, 0.64359906],
[ 0.18644383, 0.35878495, 0.46146176, -1.75039558, 1.15147572,
-0.79001927, 0.06842818, -0.92784629, 0.53966766, -1.15624849,
-0.56785447, -0.09107513],
[ 1.64521859, -0.35208162, -0.24719867, 0.73485631, -0.7193255 ,
0.18748171, 0.40802318, 0.15651978, -1.7070667 , -0.19710219,
-0.52689884, 0.12777919],
[-0.00589913, 0.11645322, -0.69165855, -1.3728932 , -1.72926142,
0.0853667 , 0.078379 , -0.21203821, -0.33286296, 2.36753064,
-1.12380217, -0.03725729],
[ 1.63224637, 0.8916754 , 1.02162072, -0.77177397, 0.10909745,
0.93361119, 0.49809446, -0.36998432, -0.15940681, -1.2318212 ,
0.54767701, -1.03294341],
[-0.71157578, -0.40025839, 0.18814183, 1.61617908, 0.67473142,
0.76452083, -1.45217186, 0.73087727, 0.5605336 , 0.77584341,
-2.04797342, 0.72173009],
[-0.75710551, -0.48514138, 1.18142324, 1.45099828, 0.61900149,
-1.50008651, 0.01960995, -0.25247218, 0.16883396, 0.36491727,
-0.19654314, -0.1775503 ],
[-0.05020209, 0.10460235, 0.40680731, 1.37479371, -0.29574042,
-0.01259258, -1.99811303, 1.91410299, -0.23692018, 0.11205839,
0.59292661, 0.44417237],
[-0.02894275, -0.65325663, 0.67701216, 0.27121676, 0.14872388,
-0.58924875, -0.54020538, -1.21573666, -0.61309256, -0.23269741,
0.04588593, 0.56945742],
[ 0.28749359, 0.56990275, 1.37091707, -1.08851728, -0.13620002,
-0.40595672, -0.1140739 , 0.41187812, -1.11993698, 0.68289479,
0.42379275, -0.64094676],
[-1.60017974, -0.84723902, -0.02600453, -0.38100938, 1.34531363,
-1.00090349, -2.28562568, 0.13931103, -0.25101233, 1.36768321,
-0.40857943, -0.04582211],
[ 0.7461974 , 0.11564297, 0.20692693, 0.15816297, -1.29618134,
0.05000082, 0.79307865, -1.80586943, 0.52601396, 0.74423465,
0.30464028, -0.59836129]])
|
|
数组排序
import numpy as np np.random.seed(73) a=np.random.randint(low=1,high=10,size=5) print(a) # [7 3 1 9 4] a.sort() print(a) # [1 3 4 7 9]
argsort索引排序
import numpy as np a=np.arange(start=2,stop=-1,step=-1) print(a) # array([2, 1, 0]) # 索引排序，默认按值的升序排列 print(a.argsort()) # array([2, 1, 0]) print(a[[2, 1, 0]]) # array([0, 1, 2])
aa=np.array([2,1,3])
aa[::-1]
array([3, 1, 2])
aa[aa.argsort()]
array([1, 2, 3])
aa[aa.argsort()][::-1]
array([3, 2, 1])
- numpy.concatenate((a1,a2,...), axis=0)函数。能够一次完成多个数组的拼接。其中a1,a2,...是数组类型的参数 - axis=0行方向上拼接 - axis=1列方向上拼接 - concatenate()效率更高，适合大规模的数据拼接
### a = np.unique(A) - 对于一维数组或者列表，unique函数去除其中重复的元素，并按元素由小到大返回一个新的无元素重复的元组或者列表 import numpy as np A = [1, 2, 2, 5, 3, 4, 3] a = np.unique(A) B= (1, 2, 2, 5, 3, 4, 3) b= np.unique(B) C= ['fgfh','asd','fgfh','asdfds','wrh'] c= np.unique(C) print(a) print(b) print(c) [1 2 3 4 5] [1 2 3 4 5] ['asd' 'asdfds' 'fgfh' 'wrh'] ### return_index - return_index=True - 表示返回新列表元素在旧列表中首次出现的位置，并以列表形式储存在s中。 import numpy as np A = [1, 2, 2, 5, 3, 4, 3] a, s= np.unique(A, return_index=True) print(a) print(s) [1 2 3 4 5] [0 1 4 5 3] ### return_inverse - a, s,p = np.unique(A, return_index=True, return_inverse=True) - a 取出了不同的类别,并做了排序 - s 是索引列表，其索引元素是A的,只不过是首次出现 - return_inverse=True 返回与旧列表元素个数相同的一个索引列表，其元素为a中元素的索引 - 从A到p完成了索引编码 - 本质上取出原标签的所有类别，并从0,1,2,...,n进行编码 a, s,p = np.unique(A, return_index=True, return_inverse=True) print(a) print(s) print(p) - a: 从小到大排列的不重复的元素列表 - s: a列表中的元素，首次在原始列表A中出现的索引位置 - p: 索引编码，原始列表的中元素在不重复列表a中的索引形成的列表 [1 2 3 4 5] [0 1 4 5 3] [0 1 1 4 2 3 2]
返回符合条件的索引
import numpy as np a = np.array([0,1,2,3]) #一维矩阵 b = np.where(a > 1) print(b) # (array([2, 3]),)
真假条件转换
import numpy as np
a = np.linspace(start=-1,stop=1,num=10)
"""
array([-1. , -0.77777778, -0.55555556, -0.33333333, -0.11111111,
0.11111111, 0.33333333, 0.55555556, 0.77777778, 1. ])
"""
b = np.where(a > 0, 1, 0)
# [0 0 0 0 0 1 1 1 1 1]
print(b)
reshape
### reshape
- 将原数组中的所有元素串起来，按新的维度进行组合
```python
import numpy as np
a = np.array( [
[
[[1],[255]],
[[0],[100] ],
]
])
print("1",a)
print("2",a.shape)
b = a.reshape(2,2)
print("3",b)
```
```
1 [[[[ 1]
[255]]
[[ 0]
[100]]]]
2 (1, 2, 2, 1)
3 [[ 1 255]
[ 0 100]]
```
reshape(-1, 1)
### reshape(-1, 1)
- 这里的-1被理解为unspecified value，意思是未指定，未给定的。
- 为-1时，该维度由程序自动决定
- 即将矩阵的shape转换为n行一列
```python
x = np.array(range(1, 10)).reshape(-1, 1)
print(x)
```
```
[[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]]
```
```python
x = np.array(range(2, 10)).reshape(2, -1)
print(x)
```
```
[[2 3 4 5]
[6 7 8 9]]
```
```python
# 原始数据的标签，是1维数据，每个样本向量对应一个标量
# 将标签变成列向量
y_train = y_train.reshape((-1,1))
y_test = y_test.reshape((-1,1))
```
X.reshape(-1)
X.reshape(-1) 将n维矩阵转1维矩阵，即向量
np.newaxis
将np.newaxis放在哪一维表示在哪一维上增加一个维度 import numpy as np arr=np.random.randn(2,3) print(arr.shape) # (2, 3) a=arr[np.newaxis,:] print(a.shape) # (1, 2, 3) a=arr[:,np.newaxis] print(a.shape) # (2, 1, 3) a=arr[:,:,np.newaxis] print(a.shape) # (2, 3, 1)
reshape
原来的shape为(n1,n2,n3) 新加一个维度,reshape(n1,1,n2,n3) 这个1写在哪个位置就表示新增维度放在哪里
- 矩阵转置,按相同的顺序交换两个序列数据的维度 - 参数为原数组shape的索引下标 import numpy as np a = np.arange(15).reshape(3,5) print(a) [[ 0 1 2 3 4] [ 5 6 7 8 9] [10 11 12 13 14]] import numpy as np a = np.arange(15).reshape(3,5) print(a.transpose(1, 0)) [[ 0 5 10] [ 1 6 11] [ 2 7 12] [ 3 8 13] [ 4 9 14]]
import numpy as np
print(np.log(np.e)) # 1.0
print(np.log2(8)) # 3.0
# 感受一下对数与十进制的变化速度
import numpy as np
a=np.array([1,10,100,1000,10000,100000,1000000,10000000]) # 个，十，百，千，万，十万，百万，千万
print(np.log(a).astype(np.float32))
"""
array([ 0. , 2.3025851, 4.6051702, 6.9077554, 9.2103405,
11.512925 , 13.815511 , 16.118095 ], dtype=float32)
"""
print(np.log2(a).astype(np.float32))
"""
array([ 0. , 3.321928, 6.643856, 9.965784, 13.287712, 16.60964 ,
19.931568, 23.253496], dtype=float32)
"""
numpy数组拼接方法介绍(concatenate)