您好,登錄后才能下訂單哦!
這篇文章主要為大家展示了“Tensorflow tf.dynamic_partition矩陣拆分的示例分析”,內(nèi)容簡(jiǎn)而易懂,條理清晰,希望能夠幫助大家解決疑惑,下面讓小編帶領(lǐng)大家一起研究并學(xué)習(xí)一下“Tensorflow tf.dynamic_partition矩陣拆分的示例分析”這篇文章吧。
先給出一個(gè)樣例看看
import tensorflow as tf raw = tf.constant([1, 2, 3, 4, 5, 6, 6, 5, 4, 3, 2, 1]) ''' 拆成 [1,2] [3,4] [5,6] [6,5] [4,3] [2,1] ''' result_1 = tf.dynamic_partition(tf.reshape(raw, [6,2]),[0, 1, 2, 3, 4, 5], 6) ''' 拆成 [1, 2, 3, 4, 5, 6] [6, 5, 4, 3, 2, 1] ''' result_2 = tf.dynamic_partition(tf.reshape(raw, [2, 6]), [0, 1], 2) ''' 拆成 [1] [2] [3] [4] [5] [6] [6] [5] [4] [3] [2] [1] ''' result_3 = tf.dynamic_partition(tf.reshape(raw, [12, 1]), [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11], 12) with tf.Session() as sess: print(sess.run(result_1)) print(sess.run(result_2)) print(sess.run(result_3))
結(jié)果
[array([[1, 2]]), array([[3, 4]]), array([[5, 6]]), array([[6, 5]]), array([[4, 3]]), array([[2, 1]])] [array([[1, 2, 3, 4, 5, 6]]), array([[6, 5, 4, 3, 2, 1]])] [array([[1]]), array([[2]]), array([[3]]), array([[4]]), array([[5]]), array([[6]]), array([[6]]), array([[5]]), array([[4]]), array([[3]]), array([[2]]), array([[1]])]
再給出一個(gè)樣例
Py3代碼:
# one-hot 函數(shù)的樣例 import tensorflow as tf label = tf.placeholder(tf.int32,[None]) # 直接把 輸入的序列進(jìn)行One-Hot的結(jié)果 one_hot = tf.one_hot(label, 3, 1, 0) # 進(jìn)行轉(zhuǎn)置 one_hot_new = tf.transpose(one_hot, perm=[1,0]) one_hot_new = tf.cast(one_hot_new, tf.float32) # one_hot_new[2] = one_hot_new[2] * 1.5 # 按照每一維的大小進(jìn)行拆分 one_hot_new_1 = tf.dynamic_partition(one_hot_new, [0, 1, 1], 2)[0] one_hot_new_2 = tf.dynamic_partition(one_hot_new, [1, 0, 1], 2)[0] one_hot_new_3 = tf.dynamic_partition(one_hot_new, [1, 1, 0], 2)[0] # 按照每一維大小進(jìn)行拆分 one_hot_1 = tf.dynamic_partition(one_hot_new, [0, 1, 2], 3)[0] one_hot_2 = tf.dynamic_partition(one_hot_new, [0, 1, 2], 3)[1] one_hot_3 = tf.dynamic_partition(one_hot_new, [0, 1, 2], 3)[2] # one_hot_new_3 = tf.dynamic_partition(one_hot_new, [0, 0, 1], 2)[2] # 拼接以上兩維得到原來(lái)的結(jié)果 one_hot_new = tf.concat([one_hot_new_1, one_hot_new_2], axis=0) if __name__ == '__main__': with tf.Session() as sess: sess.run(tf.global_variables_initializer()) one_hot_out, one_hot_new_out, one_hot_new_1_out, one_hot_new_2_out, one_hot_new_3_out, one_hot_1_out, one_hot_2_out, one_hot_3_out = sess.run([one_hot, one_hot_new, one_hot_new_1, one_hot_new_2, one_hot_new_3, one_hot_1, one_hot_2, one_hot_3], feed_dict={label: [0, 1, 1, 2, 2, 0, 0, 1, 2, 2, 0, 2]}) print("原始的One-hot結(jié)果:") print(one_hot_out, end='\n\n') print("以上的結(jié)果.T:") print("方法一拆分:") print(one_hot_new_out, end='\n\n') print("拆分(1)維:") print(one_hot_new_1_out, end='\n\n') print("拆分 (2)維:") print(one_hot_new_2_out, end='\n\n') print("拆分 (3)維:") print(one_hot_new_3_out, end='\n\n') print("方法二拆分:") print("拆分(1)維:") print(one_hot_1_out, end='\n\n') print("拆分 (2)維:") print(one_hot_2_out, end='\n\n') print("拆分 (3)維:") print(one_hot_3_out, end='\n\n')
控制臺(tái)輸出:
原始的One-hot結(jié)果: [[1 0 0] [0 1 0] [0 1 0] [0 0 1] [0 0 1] [1 0 0] [1 0 0] [0 1 0] [0 0 1] [0 0 1] [1 0 0] [0 0 1]] 以上的結(jié)果.T: 方法一拆分: [[ 1. 0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0.] [ 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0.]] 拆分(1)維: [[ 1. 0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0.]] 拆分 (2)維: [[ 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0.]] 拆分 (3)維: [[ 0. 0. 0. 1. 1. 0. 0. 0. 1. 1. 0. 1.]] 方法二拆分: 拆分(1)維: [[ 1. 0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0.]] 拆分 (2)維: [[ 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0.]] 拆分 (3)維: [[ 0. 0. 0. 1. 1. 0. 0. 0. 1. 1. 0. 1.]]
以上是“Tensorflow tf.dynamic_partition矩陣拆分的示例分析”這篇文章的所有內(nèi)容,感謝各位的閱讀!相信大家都有了一定的了解,希望分享的內(nèi)容對(duì)大家有所幫助,如果還想學(xué)習(xí)更多知識(shí),歡迎關(guān)注億速云行業(yè)資訊頻道!
免責(zé)聲明:本站發(fā)布的內(nèi)容(圖片、視頻和文字)以原創(chuàng)、轉(zhuǎn)載和分享為主,文章觀點(diǎn)不代表本網(wǎng)站立場(chǎng),如果涉及侵權(quán)請(qǐng)聯(lián)系站長(zhǎng)郵箱:is@yisu.com進(jìn)行舉報(bào),并提供相關(guān)證據(jù),一經(jīng)查實(shí),將立刻刪除涉嫌侵權(quán)內(nèi)容。