zhouzhou0929的博客

import numpy as np# 定义softmax函数，softmax:输出0-1之间的数，其和为1，可以用来表征判定是某个类别的概率def softmax(x): # x为一维数据时 if x.ndim == 1: return np.exp(x-np.max(x))/np.sum(np.exp(x-np.max(x))) # x为二维数据时 elif x.ndim == 2: val_num = np.ze

import numpy as np# 定义softmax函数，softmax:输出0-1之间的数，其和为1，可以用来表征判定是某个类别的概率def softmax(x): # x为一维数据时 if x.ndim == 1: return np.exp(x-np.max(x))/np.sum(np.exp(x-np.max(x))) # x为二维数据时 elif x.ndim == 2: val_num = np.ze

import numpy as np# 定义func2函数def func2(x): return x[0]**2 + x[1]**2# 求func2函数在点[3.0, 4.0]处的梯度def numerical_diff(f, x): h = 1e-4 grap = np.zeros_like(x) for idx in range(x.size): tmp_val = x[idx] x[idx] = tmp_val + h

# 画出模型在训练集和验证集上，关于训练集大小的性能函数（采用均方误差）# 1.导入需要使用的包from sklearn.metrics import mean_squared_errorfrom sklearn.model_selection import train_test_splitfrom sklearn.linear_model import LinearRegressionimport matplotlib.pyplot as pltimport numpy as np# 2

import numpy as npfrom sklearn.preprocessing import PolynomialFeaturesfrom sklearn.linear_model import LinearRegressionimport matplotlib.pyplot as pltm = 100X = 6 * np.random.rand(m, 1) - 3y = 0.5 * X**2 + X + 2 + np.random.randn(m, 1)plf = Polyno

import numpy as np# 设定x,y对应的值，即二维空间对应的100个点x = np.random.rand(100, 1)y = 4 + (3*x) + (0.2 * np.random.rand(100, 1))m = 100# 对每一行添加x0=1x_b = np.c_[np.ones(shape=(100, 1)), x]# 设定迭代次数n_iteration = 200# 设定学习率函数def learning_scheture(t): retur

import numpy as np# 设定x,y对应的值，即二维空间对应的100个点x = np.random.rand(100, 1)y = 4 + (3*x) + (0.2 * np.random.rand(100, 1))# 对每一行添加x0=1x_b = np.c_[np.ones(shape=(100, 1)), x]# 设定步长eta = 0.1# 设定迭代次数n_iteration = 1000# 利用迭代求解成本函数最小时的theta值# 设定任意初始theta值，