關(guān)于tf.nn.dynamic_rnn返回值詳解
函數(shù)原型
tf.nn.dynamic_rnn( cell, inputs, sequence_length=None, initial_state=None, dtype=None, parallel_iterations=None, swap_memory=False, time_major=False, scope=None )
實(shí)例講解:
import tensorflow as tf import numpy as np n_steps = 2 n_inputs = 3 n_neurons = 5 X = tf.placeholder(tf.float32, [None, n_steps, n_inputs]) basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons) seq_length = tf.placeholder(tf.int32, [None]) outputs, states = tf.nn.dynamic_rnn(basic_cell, X, dtype=tf.float32, sequence_length=seq_length) init = tf.global_variables_initializer() X_batch = np.array([ # step 0 step 1 [[0, 1, 2], [9, 8, 7]], # instance 1 [[3, 4, 5], [0, 0, 0]], # instance 2 (padded with zero vectors) [[6, 7, 8], [6, 5, 4]], # instance 3 [[9, 0, 1], [3, 2, 1]], # instance 4 ]) seq_length_batch = np.array([2, 1, 2, 2]) with tf.Session() as sess: init.run() outputs_val, states_val = sess.run( [outputs, states], feed_dict={X: X_batch, seq_length: seq_length_batch}) print("outputs_val.shape:", outputs_val.shape, "states_val.shape:", states_val.shape) print("outputs_val:", outputs_val, "states_val:", states_val)
log info:
outputs_val.shape: (4, 2, 5) states_val.shape: (4, 5) outputs_val: [[[ 0.53073734 -0.61281306 -0.5437517 0.7320347 -0.6109526 ] [ 0.99996936 0.99990636 -0.9867181 0.99726075 -0.99999976]] [[ 0.9931584 0.5877845 -0.9100412 0.988892 -0.9982337 ] [ 0. 0. 0. 0. 0. ]] [[ 0.99992317 0.96815354 -0.985101 0.9995968 -0.9999936 ] [ 0.99948144 0.9998127 -0.57493806 0.91015154 -0.99998355]] [[ 0.99999255 0.9998929 0.26732785 0.36024097 -0.99991137] [ 0.98875254 0.9922327 0.6505734 0.4732064 -0.9957567 ]]] states_val: [[ 0.99996936 0.99990636 -0.9867181 0.99726075 -0.99999976] [ 0.9931584 0.5877845 -0.9100412 0.988892 -0.9982337 ] [ 0.99948144 0.9998127 -0.57493806 0.91015154 -0.99998355] [ 0.98875254 0.9922327 0.6505734 0.4732064 -0.9957567 ]]
首先輸入X是一個(gè) [batch_size,step,input_size] = [4,2,3] 的tensor,注意我們這里調(diào)用的是BasicRNNCell,只有一層循環(huán)網(wǎng)絡(luò),outputs是最后一層每個(gè)step的輸出,它的結(jié)構(gòu)是[batch_size,step,n_neurons] = [4,2,5],states是每一層的最后那個(gè)step的輸出,由于本例中,我們的循環(huán)網(wǎng)絡(luò)只有一個(gè)隱藏層,所以它就代表這一層的最后那個(gè)step的輸出,因此它和step的大小是沒有關(guān)系的,我們的X有4個(gè)樣本組成,輸出神經(jīng)元大小n_neurons是5,因此states的結(jié)構(gòu)就是[batch_size,n_neurons] = [4,5],最后我們觀察數(shù)據(jù),states的每條數(shù)據(jù)正好就是outputs的最后一個(gè)step的輸出。
下面我們繼續(xù)講解多個(gè)隱藏層的情況,這里是三個(gè)隱藏層,注意我們這里仍然是調(diào)用BasicRNNCell
import tensorflow as tf import numpy as np n_steps = 2 n_inputs = 3 n_neurons = 5 n_layers = 3 X = tf.placeholder(tf.float32, [None, n_steps, n_inputs]) seq_length = tf.placeholder(tf.int32, [None]) layers = [tf.contrib.rnn.BasicRNNCell(num_units=n_neurons, activation=tf.nn.relu) for layer in range(n_layers)] multi_layer_cell = tf.contrib.rnn.MultiRNNCell(layers) outputs, states = tf.nn.dynamic_rnn(multi_layer_cell, X, dtype=tf.float32, sequence_length=seq_length) init = tf.global_variables_initializer() X_batch = np.array([ # step 0 step 1 [[0, 1, 2], [9, 8, 7]], # instance 1 [[3, 4, 5], [0, 0, 0]], # instance 2 (padded with zero vectors) [[6, 7, 8], [6, 5, 4]], # instance 3 [[9, 0, 1], [3, 2, 1]], # instance 4 ]) seq_length_batch = np.array([2, 1, 2, 2]) with tf.Session() as sess: init.run() outputs_val, states_val = sess.run( [outputs, states], feed_dict={X: X_batch, seq_length: seq_length_batch}) print("outputs_val.shape:", outputs, "states_val.shape:", states) print("outputs_val:", outputs_val, "states_val:", states_val)
log info:
outputs_val.shape: Tensor("rnn/transpose_1:0", shape=(?, 2, 5), dtype=float32) states_val.shape: (<tf.Tensor 'rnn/while/Exit_3:0' shape=(?, 5) dtype=float32>, <tf.Tensor 'rnn/while/Exit_4:0' shape=(?, 5) dtype=float32>, <tf.Tensor 'rnn/while/Exit_5:0' shape=(?, 5) dtype=float32>) outputs_val: [[[0. 0. 0. 0. 0. ] [0. 0.18740742 0. 0.2997518 0. ]] [[0. 0.07222144 0. 0.11551574 0. ] [0. 0. 0. 0. 0. ]] [[0. 0.13463384 0. 0.21534224 0. ] [0.03702604 0.18443246 0. 0.34539366 0. ]] [[0. 0.54511094 0. 0.8718864 0. ] [0.5382122 0. 0.04396425 0.4040263 0. ]]] states_val: (array([[0. , 0.83723307, 0. , 0. , 2.8518028 ], [0. , 0.1996038 , 0. , 0. , 1.5456247 ], [0. , 1.1372368 , 0. , 0. , 0.832613 ], [0. , 0.7904129 , 2.4675028 , 0. , 0.36980057]], dtype=float32), array([[0.6524607 , 0. , 0. , 0. , 0. ], [0.25143963, 0. , 0. , 0. , 0. ], [0.5010576 , 0. , 0. , 0. , 0. ], [0. , 0.3166597 , 0.4545995 , 0. , 0. ]], dtype=float32), array([[0. , 0.18740742, 0. , 0.2997518 , 0. ], [0. , 0.07222144, 0. , 0.11551574, 0. ], [0.03702604, 0.18443246, 0. , 0.34539366, 0. ], [0.5382122 , 0. , 0.04396425, 0.4040263 , 0. ]], dtype=float32))
我們說過,outputs是最后一層的輸出,即 [batch_size,step,n_neurons] = [4,2,5]
states是每一層的最后一個(gè)step的輸出,即三個(gè)結(jié)構(gòu)為 [batch_size,n_neurons] = [4,5] 的tensor
繼續(xù)觀察數(shù)據(jù),states中的最后一個(gè)array,正好是outputs的最后那個(gè)step的輸出
下面我們繼續(xù)講當(dāng)由BasicLSTMCell構(gòu)造單元工廠的時(shí)候,只講多層的情況,我們只需要將上面的BasicRNNCell替換成BasicLSTMCell就行了,打印信息如下:
outputs_val.shape: Tensor("rnn/transpose_1:0", shape=(?, 2, 5), dtype=float32) states_val.shape: (LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_3:0' shape=(?, 5) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_4:0' shape=(?, 5) dtype=float32>), LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_5:0' shape=(?, 5) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_6:0' shape=(?, 5) dtype=float32>), LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_7:0' shape=(?, 5) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_8:0' shape=(?, 5) dtype=float32>)) outputs_val: [[[1.2949290e-04 0.0000000e+00 2.7623639e-04 0.0000000e+00 0.0000000e+00] [9.4675866e-05 0.0000000e+00 2.0214770e-04 0.0000000e+00 0.0000000e+00]] [[4.3100454e-06 4.2123037e-07 1.4312843e-06 0.0000000e+00 0.0000000e+00] [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]] [[0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00] [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]] [[0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00] [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]]] states_val: (LSTMStateTuple( c=array([[0. , 0. , 0.04676079, 0.04284539, 0. ], [0. , 0. , 0.0115245 , 0. , 0. ], [0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. ]], dtype=float32), h=array([[0. , 0. , 0.00035096, 0.04284406, 0. ], [0. , 0. , 0.00142574, 0. , 0. ], [0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. ]], dtype=float32)), LSTMStateTuple( c=array([[0.0000000e+00, 1.0477135e-02, 4.9871090e-03, 8.2785974e-04, 0.0000000e+00], [0.0000000e+00, 2.3306280e-04, 0.0000000e+00, 9.9445322e-05, 5.9535629e-05], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00]], dtype=float32), h=array([[0.00000000e+00, 5.23016974e-03, 2.47756205e-03, 4.11730434e-04, 0.00000000e+00], [0.00000000e+00, 1.16522635e-04, 0.00000000e+00, 4.97301044e-05, 2.97713632e-05], [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00], [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]], dtype=float32)), LSTMStateTuple( c=array([[1.8937115e-04, 0.0000000e+00, 4.0442235e-04, 0.0000000e+00, 0.0000000e+00], [8.6200516e-06, 8.4243663e-07, 2.8625946e-06, 0.0000000e+00, 0.0000000e+00], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00]], dtype=float32), h=array([[9.4675866e-05, 0.0000000e+00, 2.0214770e-04, 0.0000000e+00, 0.0000000e+00], [4.3100454e-06, 4.2123037e-07, 1.4312843e-06, 0.0000000e+00, 0.0000000e+00], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00], [0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00, 0.0000000e+00]], dtype=float32)))
我們先看看LSTM單元的結(jié)構(gòu)
如果您不查看框內(nèi)的內(nèi)容,LSTM單元看起來與常規(guī)單元格完全相同,除了它的狀態(tài)分為兩個(gè)向量:h(t)和c(t)。你可以將h(t)視為短期狀態(tài),將c(t)視為長期狀態(tài)。
因此我們的states包含三個(gè)LSTMStateTuple,每一個(gè)表示每一層的最后一個(gè)step的輸出,這個(gè)輸出有兩個(gè)信息,一個(gè)是h表示短期記憶信息,一個(gè)是c表示長期記憶信息。維度都是[batch_size,n_neurons] = [4,5],states的最后一個(gè)LSTMStateTuple中的h就是outputs的最后一個(gè)step的輸出
以上這篇關(guān)于tf.nn.dynamic_rnn返回值詳解就是小編分享給大家的全部內(nèi)容了,希望能給大家一個(gè)參考,也希望大家多多支持腳本之家。
相關(guān)文章
Python 3.6 讀取并操作文件內(nèi)容的實(shí)例
下面小編就為大家分享一篇Python 3.6 讀取并操作文件內(nèi)容的實(shí)例,具有很好的參考價(jià)值,希望對(duì)大家有所幫助。一起跟隨小編過來看看吧2018-04-04Python + selenium + requests實(shí)現(xiàn)12306全自動(dòng)搶票及驗(yàn)證碼破解加自動(dòng)點(diǎn)擊功能
這篇文章主要介紹了Python + selenium + requests實(shí)現(xiàn)12306全自動(dòng)搶票及驗(yàn)證碼破解加自動(dòng)點(diǎn)擊功能,需要的朋友可以參考下2018-11-11python中matplotlib調(diào)整圖例位置的方法實(shí)例
在matplotlib中,一般圖例默認(rèn)是在圖表內(nèi)部的,如果要放置到圖例外面,需要對(duì)坐標(biāo)進(jìn)行指定,下面這篇文章主要給大家介紹了關(guān)于python中matplotlib調(diào)整圖例位置的相關(guān)資料,需要的朋友可以參考下2022-06-06使用Python編寫一個(gè)簡單的tic-tac-toe游戲的教程
這篇文章主要介紹了使用Python編寫一個(gè)簡單的tic-tac-toe游戲的教程,有利于Python初學(xué)者進(jìn)行上手實(shí)踐,需要的朋友可以參考下2015-04-04pytorch 模擬關(guān)系擬合——回歸實(shí)例
今天小編就為大家分享一篇pytorch 模擬關(guān)系擬合——回歸實(shí)例,具有很好的參考價(jià)值,希望對(duì)大家有所幫助。一起跟隨小編過來看看吧2020-01-01Python面向?qū)ο笏枷肱c應(yīng)用入門教程【類與對(duì)象】
這篇文章主要介紹了Python面向?qū)ο笏枷肱c應(yīng)用,較為詳細(xì)的分析了Python面向?qū)ο笏枷肱c原理,并結(jié)合實(shí)例形式分析了類與對(duì)象相關(guān)定義、用法及操作注意事項(xiàng),需要的朋友可以參考下2019-04-04Python實(shí)現(xiàn)全角半角轉(zhuǎn)換的方法
這篇文章主要介紹了Python實(shí)現(xiàn)全角半角轉(zhuǎn)換的方法,很實(shí)用的方法,需要的朋友可以參考下2014-08-08