Question

我有一个皮肤病学数据库，其标准化已经完成。这是一个CSV文件。我需要打开文件并输入一个numpy数组。我的数据库有34列，大约350行。我的神经网络有3个隐藏层。这是我目前神经网络的python代码。有人可以帮助我在numpy数组/矩阵中输入吗？谢谢

这是我的代码：

import numpy as np
#input x
X = np.array([  ])
#input y
y = np.array([]).T

syn0 = 2*np.random.random((34,26)) - 1
syn1 = 2*np.random.random((26,18)) - 1
syn2 = 2*np.random.random((18,11)) - 1
syn3 = 2*np.random.random((11,6)) - 1
for j in xrange(350):
  l1 = 1/(1+np.exp(-(np.dot(X,syn0))))
  l2 = 1/(1+np.exp(-(np.dot(l1,syn1))))
  l3 = 1/(1+np.exp(-(np.dot(l2,syn2))))
  l4 = 1/(1+np.exp(-(np.dot(l3,syn3))))
  l4_delta = (y - l4)*(l4*(1-l4))
  l3_delta = l4_delta.dot(syn3.T) * (l3 * (1-l3))
  l2_delta = l3_delta.dot(syn2.T) * (l2 * (1-l2))
  l1_delta = l2_delta.dot(syn1.T) * (l1 * (1-l1))
  syn3 += l1.T.dot(l4_delta)
  syn2 += l1.T.dot(l3_delta)
  syn1 += l1.T.dot(l2_delta)
  syn0 += X.T.dot(l1_delta)

Answer 1

假设标签位于最后一列，

angular.module('monopolyApp')
  .controller('MainCtrl', function ($scope, $window, $timeout, localStorageService) {

我认为你不需要转换X = np.array([[float(cell) for cell in row[:-1]] for row in csv.reader(open(csv_filename))]) Y = np.array([float(row[-1]) for row in csv.reader(open(csv_filename))])，假设它是一维的。

Answer 2

我这样做了：

my_5_input_numbers.csv
0.3,0.5,0.6,0.7,1
0.4,0.6,0.7,0.8,0
0.5,0.7,0.8,0.9,1

使用numpy：

data_common=np.genfromtxt('my_5_input_numbers.csv',delimiter=',')
"""
data_common=array([[ 0.3,  0.5,  0.6,  0.7,  1. ],
       [ 0.4,  0.6,  0.7,  0.8,  0. ],
       [ 0.5,  0.7,  0.8,  0.9,  1. ]])
data_common.shape=(3,5)       
"""

X_train=data_common[:,-1]#X_train.shape=(3,4)
y_train=data_common[0:4,-1]#y_train.shape=(3,)
y_train=y_train.reshape(3,1)

Python神经网络编码

2 个答案: