深度學習 - 認識 one hot 數據預處理

深度學習 - 認識 one hot 數據預處理

one hot 數據預處理，是將訓練&測試數據向量化的方法之一，見以下編碼實例：

import numpy as np

def vectorize_sequences(sequences, dimension=10000):
    results = np.zeros((len(sequences), dimension))
    for i, sequence in enumerate(sequences):
        results[i, sequence] = 1.    
    return results

以IMDb為例:
In[3]: train_data.shape
Out[3]: (25000,)
訓練資料是一個25000組的0維向量組成

In[4]:train_data

Out[4]:

for i in train_data[0]:
    print('%s,' %i,end="")

印出 train_data[0] 實際的樣子如下:
1,14,22,16,43,530,973,1622,1385,65,458,4468,66,3941,4,173,36,256,5,25,100,43,838,112,50,670,2,9,35,.......

train_data_0 =train_data[0]
train_data_0.sort()
for i in train_data_0[:128]:
    print('%s,' %int(i) ,end="")

將train_data[0] 排序後列印出來，可以看到 2,4,5,6,7,8...重複。1,2,2,2,2,2,2,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,6,6,6,7,7,8,8,8,9,12,12,12,12,12,12,13,13,13,14,14,14,15,15,15,15,16,16,16,16,16,16,16,16,16,16,16,17,17,17,18,18,18,18,.........

以上這些幫助我們接下來要了解one hot 編碼要做的事情。

np.zeros((len(sequences), dimension))

此例的 one hot 編碼創造了一個10000維向量為0的矩陣 [0, 0, 0, 0,......0,......0] 總共10000個0。
然後將index 有值的地方輸出結果為1
見以下使用範例:
>>>seq = ['one', 'two', 'three'] >>>for i, element in enumerate(seq): ... print(i, seq[i])

    for i, sequence in enumerate(sequences):
        results[i, sequence] = 1.    

這邊是將index有值的部分，輸出1


將訓練data 帶進one hot 向量化函數後印出來看看發生了什麼事?

x_train = vectorize_sequences(train_data)
x_test = vectorize_sequences(test_data)

for i in x_train[0][:128]:
    print('%s,' %int(i) ,end="")

得到以下數據 :
0,1,1,0,1,1,1,1,1,1,0,0,1,1,1,1,1,1,1,1,0,1,1,0,0,1,1,0,1,0,1,0,1,1,0,1,1,0,1,1,0,0,0,1,0,0,1,0,1,0,1,1,1,0,0,0,1,0,0,0,0,0,1,0,0,1,1,0,0,0,0,1,0,0,0,0,1,1,0,0,0,0,1,0,0,0,0,1,1,0,0,0,1,0,0,0,0,0,1,0,1,0,0,1,1,0,1,1,0,0,0,0,1,1,0,0,0,1,0,0,0,0,0,0,1,0,0,0,

比對一下train_data & x_train就可以了解。

可以看到3,10,11的地方沒有數據，則為0向量。