我的数据位于dtype float64的ndarray中。
我的变量x和y如下所示：
>>print x
[[ 2.00000000e+00 1.12400000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.48800000e+03]
[ 1.00000000e+00 7.00000000e+02]
[ 2.00000000e+00 1.24900000e+03]
[ 1.00000000e+00 8.05000000e+02]
[ 2.00000000e+00 1.36000000e+03]
[ 2.00000000e+00 1.12100000e+03]
[ 1.00000000e+00 8.05000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.50800000e+03]
[ 4.00000000e+00 3.41400000e+03]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 1.15700000e+03]
[ 1.00000000e+00 8.55000000e+02]
[ 1.00000000e+00 7.30000000e+02]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 1.21500000e+03]
[ 2.00000000e+00 1.38500000e+03]
[ 3.00000000e+00 1.29300000e+03]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 1.48800000e+03]
[ 2.00000000e+00 1.20000000e+03]
[ 3.00000000e+00 1.22500000e+03]
[ 1.00000000e+00 8.15000000e+02]
[ 3.00000000e+00 1.24700000e+03]
[ 2.00000000e+00 1.15600000e+03]
[ 1.00000000e+00 8.27000000e+02]
[ 1.00000000e+00 7.00000000e+02]
[ 2.00000000e+00 1.20000000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 7.64000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 8.30000000e+02]
[ 3.00000000e+00 1.22500000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 8.16000000e+02]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 1.03000000e+03]
[ 3.00000000e+00 1.24700000e+03]
[ 2.00000000e+00 1.06200000e+03]
[ 1.00000000e+00 6.57000000e+02]
[ 1.00000000e+00 7.73000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.31300000e+03]
[ 2.00000000e+00 8.00000000e+02]
[ 1.00000000e+00 7.50000000e+02]
[ 2.00000000e+00 1.21700000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 4.00000000e+00 2.76300000e+03]
[ 2.00000000e+00 1.15700000e+03]
[ 2.00000000e+00 1.12100000e+03]
[ 2.00000000e+00 1.20000000e+03]
[ 3.00000000e+00 1.48100000e+03]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 8.00000000e+02]
[ 3.00000000e+00 1.61600000e+03]
[ 2.00000000e+00 1.38500000e+03]
[ 2.00000000e+00 1.50000000e+03]
[ 2.00000000e+00 1.38500000e+03]
[ 2.00000000e+00 1.14800000e+03]
[ 1.00000000e+00 8.59000000e+02]
[ 2.00000000e+00 1.38500000e+03]
[ 3.00000000e+00 1.55800000e+03]
[ 2.00000000e+00 1.47000000e+03]
[ 1.00000000e+00 7.77000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.21000000e+03]
[ 3.00000000e+00 1.30100000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 3.00000000e+00 1.22500000e+03]
[ 2.00000000e+00 1.15600000e+03]
[ 1.00000000e+00 8.05000000e+02]
[ 1.00000000e+00 7.34000000e+02]
[ 2.00000000e+00 9.65000000e+02]
[ 1.00000000e+00 8.30000000e+02]
[ 3.00000000e+00 1.22500000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 3.00000000e+00 1.42100000e+03]
[ 1.00000000e+00 7.50000000e+02]
[ 3.00000000e+00 1.78900000e+03]
[ 2.00000000e+00 1.12100000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 8.05000000e+02]
[ 3.00000000e+00 1.20000000e+03]
[ 4.00000000e+00 2.76400000e+03]
[ 2.00000000e+00 1.01500000e+03]
[ 2.00000000e+00 1.84400000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.09100000e+03]
[ 1.00000000e+00 8.70000000e+02]
[ 1.00000000e+00 8.30000000e+02]
[ 2.00000000e+00 1.12100000e+03]
[ 2.00000000e+00 1.21400000e+03]
[ 2.00000000e+00 9.26000000e+02]
[ 2.00000000e+00 1.09700000e+03]
[ 1.00000000e+00 6.25000000e+02]
[ 1.00000000e+00 6.25000000e+02]
[ 1.00000000e+00 7.50000000e+02]
[ 2.00000000e+00 1.15600000e+03]
[ 2.00000000e+00 1.48800000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.06000000e+03]
[ 5.00000000e+00 3.66200000e+03]
[ 2.00000000e+00 1.03000000e+03]
[ 2.00000000e+00 1.17000000e+03]
[ 1.00000000e+00 7.64000000e+02]
[ 3.00000000e+00 1.34000000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 4.00000000e+00 3.54900000e+03]
[ 3.00000000e+00 1.00000000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 8.00000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 2.00000000e+00 1.09100000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 3.00000000e+00 1.38500000e+03]
[ 1.00000000e+00 6.09000000e+02]
[ 1.00000000e+00 6.09000000e+02]
[ 5.00000000e+00 3.66200000e+03]
[ 4.00000000e+00 1.76900000e+03]
[ 2.00000000e+00 1.14400000e+03]
[ 2.00000000e+00 1.09100000e+03]
[ 2.00000000e+00 1.09100000e+03]
[ 2.00000000e+00 1.12100000e+03]
[ 3.00000000e+00 1.47000000e+03]
[ 3.00000000e+00 1.58000000e+03]
[ 1.00000000e+00 9.60000000e+02]
[ 2.00000000e+00 1.01500000e+03]
[ 3.00000000e+00 1.44500000e+03]
[ 2.00000000e+00 1.06400000e+03]
[ 2.00000000e+00 1.09100000e+03]
[ 1.00000000e+00 7.50000000e+02]
[ 2.00000000e+00 1.09100000e+03]
[ 3.00000000e+00 1.80000000e+03]
[ 2.00000000e+00 1.25400000e+03]
[ 2.00000000e+00 1.09100000e+03]
[ 1.00000000e+00 8.79000000e+02]
[ 2.00000000e+00 1.50800000e+03]
[ 1.00000000e+00 8.43000000e+02]
[ 4.00000000e+00 2.10800000e+03]
[ 2.00000000e+00 1.20900000e+03]
[ 2.00000000e+00 1.50000000e+03]
[ 1.00000000e+00 7.50000000e+02]
[ 2.00000000e+00 1.46100000e+03]
[ 2.00000000e+00 8.50000000e+02]
[ 3.00000000e+00 1.50000000e+03]
[ 2.00000000e+00 9.50000000e+02]
[ 3.00000000e+00 1.34000000e+03]
[ 1.00000000e+00 7.30000000e+02]
[ 2.00000000e+00 1.14100000e+03]
[ 3.00000000e+00 1.12400000e+03]
[ 2.00000000e+00 1.12100000e+03]
[ 3.00000000e+00 1.22500000e+03]
[ 2.00000000e+00 1.00000000e+03]
[ 2.00000000e+00 1.31300000e+03]]
>>print y.flatten()
[ 1775. 1106. 1930. 1267. 1350. 1250. 1500. 1690. 1300. 1110.
1178. 2200. 4500. 1985. 2045. 1195. 1100. 1985. 2269. 1550.
2168. 2055. 1930. 1668. 1728. 1300. 1890. 1985. 1833. 1207.
1741. 1090. 1050. 1188. 1308. 1745. 1200. 1230. 1680. 2070.
1450. 1980. 1400. 1542. 1593. 1138. 2363. 850. 1050. 2137.
1211. 2750. 2045. 1677. 1500. 2200. 2070. 775. 2100. 1500.
1700. 1500. 1900. 1757. 1500. 2810. 1500. 1275. 1166. 1400.
2569. 1256. 1633. 2070. 1290. 1150. 1435. 1344. 1628. 1166.
2007. 1675. 2200. 1477. 1256. 1350. 1495. 2750. 1550. 2499.
1186. 2098. 1372. 1384. 1567. 1650. 1375. 1350. 1075. 1200.
1756. 1985. 1755. 1212. 1374. 3750. 1450. 1350. 1100. 1700.
1166. 1212. 1202. 3950. 1250. 1054. 1241. 1100. 1256. 2098.
1202. 1695. 1256. 1256. 3750. 2300. 1900. 2098. 2098. 1527.
1450. 1700. 1381. 1600. 2000. 2021. 2098. 1663. 2098. 2000.
2331. 2098. 1395. 2200. 1400. 2350. 2284. 1625. 1692. 1650.
1339. 1800. 1428. 1700. 1100. 1518. 1700. 1492. 1590. 1300.
2398.]
这个数据看起来与example类似，但是当我打印出我的结果时，回归似乎没有像我预期的那样运行（我预计会有1x2的beta版矩阵）。
我的结果如下：
('Coefficients: \n', array([[ 7.03950002e+00, -2.69281738e-02],
[ 7.03950002e+00, -2.69281738e-02],
[ -6.98978455e+00, 4.54840941e-03],
[ 1.44445066e+00, -1.75824530e-02],
[ 8.11638781e-02, -9.85091887e-03],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.65529232e-03, -4.10775159e-03],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ -4.85824760e+00, 2.00707943e-03],
[ 1.16874660e+00, -1.54523463e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ -1.66777791e+01, 1.59406094e-02],
[ 7.50461656e-01, -1.44174098e-02],
[ 1.27817712e+00, -1.62305159e-02],
[ -2.40249139e+00, -2.19857700e-03],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.27817712e+00, -1.62305159e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ 1.44445066e+00, -1.75824530e-02],
[ 3.20263392e+00, -9.50805754e-03],
[ 1.87802711e+00, -2.23108770e-02],
[ -2.40249139e+00, -2.19857700e-03],
[ -6.01730918e+00, 5.05808885e-03],
[ -7.18885421e+00, 6.34980889e-03],
[ -1.13882965e+00, -9.90208457e-05],
[ -8.39829744e+00, 8.30050908e-03],
[ 4.06246539e+00, -1.24502573e-02],
[ -8.39829763e+00, 8.30050917e-03],
[ -4.58224954e-01, 4.95062738e-04],
[ -1.74875944e+01, 1.68592654e-02],
[ 7.17565680e-01, -1.52851151e-03],
[ 3.01633761e+00, -9.16054410e-03],
[ -5.76658998e+01, 5.87982295e-02],
[ -8.39829954e+00, 8.30051003e-03],
[ -2.30256592e+01, 2.09775103e-02],
[ -6.52633498e-01, 5.82137199e-04],
[ 2.35503800e+00, -6.55862039e-03],
[ 2.11785612e+00, -5.89089182e-03],
[ 1.50150684e+00, -2.08582766e-03],
[ 4.82563106e-01, -6.81511087e-04],
[ 4.82562903e-01, -6.81510981e-04],
[ 4.91817942e+00, -8.22393628e-03],
[ -1.19659445e+00, 1.99936701e-03],
[ 4.29564525e-01, -4.87141011e-04],
[ 4.82563156e-01, -6.81511112e-04],
[ -4.06573886e-01, -6.12386202e-03],
[ -8.70466734e-02, 3.60870537e-04],
[ 4.82562847e-01, -6.81510951e-04],
[ 4.66515975e+00, -7.35139691e-03],
[ -5.91285911e+00, 4.80207972e-03],
[ 1.34044916e+00, -3.36860893e-03],
[ -2.21927262e+00, 3.25733406e-03],
[ 4.66515975e+00, -7.35139691e-03],
[ 4.66515975e+00, -7.35139691e-03],
[ -9.45979683e-01, 1.62357444e-03],
[ -6.73126888e+00, 3.86607763e-03],
[ -4.16308006e-02, 2.63002814e-04],
[ -6.73126888e+00, 3.86607763e-03],
[ 4.82562487e-01, -6.81510763e-04],
[ -9.86380890e+00, 9.58052088e-03],
[ 4.82562171e-01, -6.81510598e-04],
[ -6.73126888e+00, 3.86607762e-03],
[ 2.97700421e+00, -3.05777520e-03],
[ 2.24255263e+00, -2.17012884e-03],
[ 4.66515975e+00, -7.35139691e-03],
[ -4.16308006e-02, 2.63002814e-04],
[ 4.66515975e+00, -7.35139691e-03],
[ -2.14831264e+00, 3.16340765e-03],
[ -6.73126889e+00, 3.86607763e-03],
[ -1.71124469e+01, 1.64491331e-02],
[ 4.76675011e-01, -6.46787057e-04],
[ 2.28010471e+00, -1.89360254e-03],
[ -8.41385808e+00, 8.21857876e-03],
[ 4.33623144e+00, -6.44620020e-03],
[ 3.34391606e-01, -3.75409118e-04],
[ -2.25369937e+00, 3.30993121e-03],
[ 4.33623145e+00, -6.44620020e-03],
[ 1.84108480e-01, -1.06071424e-04],
[ 1.89804414e+00, -1.22769239e-03],
[ 2.80111814e+00, -2.67770187e-03],
[ 6.64662880e-01, -1.38196985e-03],
[ 2.81998334e-01, -2.50519561e-04],
[ 4.43329199e-01, -3.89208176e-04],
[ 2.49834048e-01, -2.55971878e-04],
[ 7.34670884e-01, -1.28326041e-03],
[ 1.86862116e+00, -1.22488295e-03],
[ -1.51028451e-01, 4.28126717e-04],
[ 3.66378641e+00, -5.17492867e-03],
[ 2.21897534e-01, 5.88682056e-04],
[ -1.27898988e-01, 4.05770885e-04],
[ -9.34311436e-02, 3.49189918e-04],
[ 1.23480481e+01, -1.06012344e-02],
[ -3.04172929e-01, 7.13615607e-04],
[ 6.33853873e+00, -4.36122413e-03],
[ -7.18817629e-01, 1.31581160e-03],
[ -7.18817629e-01, 1.31581160e-03],
[ -6.76849419e+01, 7.80179333e-02],
[ 3.61205395e+00, -4.73585605e-03],
[ 1.24893688e+00, 5.47184400e-04],
[ 2.13731015e+00, -1.59511439e-03],
[ 7.80322748e+00, 6.41670434e-05],
[ -1.25727569e+01, 1.80771247e-02],
[ -4.06421193e+00, 7.16662649e-03]]))
我的模型调用按以下方式完成：
from sklearn import linear_model
logreg.fit(x, y.flatten())
我的任何形状都是：
print y.shape,x.shape
(161, 1) (161, 2)
我一定是犯了一个愚蠢的错误，任何意见都非常感激。
答案 0 :(得分：1)
我认为您的问题不在于代码，而在于您用于数据的模型。
如果我只使用您的数据，我会得到相同的结果。你了解逻辑回归的作用吗？它根据连续输入数据预测一个类别。如果查看发布的示例，iris.target数据是0到2（包括0和2）的整数列表，每个整数对应一个不同的iris类型。例如0对应于setsoa类型。
您的y数据似乎不是标签数据。此外，期望1x2不符合分类设置。运行您链接到的示例，然后执行
logreg.coef_.shape
>>(3L, 2L)
你应该期望表格中每个类别的系数中有一行（n-cat，n-features），所以1x2期待一个类别，这没有意义--1类别只是“所有数据”。
您会获得大量行，因为您的每个y数据都被视为一个新类别。
我想也许你想要线性回归而不是逻辑回归。