機器學習----python基礎

1. shape函式

import numpy as np
x = np.array([[1,2,5],[2,3,5],[3,4,5],[2,3,6]])
# 輸出陣列的行和列數
print x.shape  # (4, 3)
# 只輸出行數
print x.shape[0] # 4
# 只輸出列數
print x.shape[1] # 3

2. dot函式

    import numpy as np
    np.dot(x,y)
    # 1、矩陣乘法，例如np.dot(X, X.T)。
    # 2、點積，比如np.dot([1, 2, 3], [4, 5, 6]) = 1 * 4 + 2 * 5 + 3 * 6 = 32。
 

3. np.reshape

import numpy as np
  
a = np.array([[1,2,3], [4,5,6]])
np.reshape(a, 6)
# array([1, 2, 3, 4, 5, 6])
np.reshape(a, 6, order='F')
# array([1, 4, 2, 5, 3, 6])
np.reshape(a, (3,-1))       # the unspecified value is inferred to be 2
# array([[1, 2],
#        [3, 4],
#        [5, 6]])

4. np.sum

• 求和函式

import numpy as np

>>> a = np.sum([6,7,3])
>>> a
16

>>> a = np.sum([[0,4],[2,6]])
>>> a
12

>>> a = np.sum([[0,4],[2,6]],axis=0)
>>> a
array([ 2, 10])

>>> a = np.sum([[0,4],[2,6]],axis=1)
>>> a
array([4, 8])

>>> a = np.sum([[0,4],[2,6]],axis=-1)
>>> a
array([4, 8])

 

5. np.hstack

• np.hstack將引數元組的元素陣列按水平方向進行疊加

import numpy as np
 
arr1 = np.array([[1,3], [2,4] ])
arr2 = np.array([[1,4], [2,6] ])
res = np.hstack((arr1, arr2))
 
print (res)
 
#
# [[1 3 1 4]
#  [2 4 2 6]]
 
arr1 = [1,2,3]
arr2 = [4,5]
arr3 = [6,7]
res = np.hstack((arr1, arr2,arr3)) 
print (res)
 
 
#
[1 2 3 4 5 6 7]

6. 梯度下降演算法流程

• 隨機初始引數
• 確定學習率
• 求出損失函式對引數梯度
• 按照公式更新引數
• 重複3、4直到滿足終止條件（如：損失函式或引數更新變化值小於某個閾值，或者訓練次數達到設定閾值）