課程五(Sequence Models)，第三週（Sequence models & Attention mechanism） —— 1.Programming assignments：Neural
Neural Machine Translation
Welcome to your first programming assignment for this week!html
You will build a Neural Machine Translation (NMT) model to translate human readable dates ("25th of June, 2009") into machine readable dates ("2009-06-25"). You will do this using an attention model, one of the most sophisticated sequence to sequence models.node
This notebook was produced together with NVIDIA's Deep Learning Institute.python
Let's load all the packages you will need for this assignmen.babel
【code】網絡
from keras.layers import Bidirectional, Concatenate, Permute, Dot, Input, LSTM, Multiply from keras.layers import RepeatVector, Dense, Activation, Lambda from keras.optimizers import Adam from keras.utils import to_categorical from keras.models import load_model, Model import keras.backend as K import numpy as np from faker import Faker import random from tqdm import tqdm from babel.dates import format_date from nmt_utils import * import matplotlib.pyplot as plt %matplotlib inline
1 - Translating human readable dates into machine readable dates
The model you will build here could be used to translate from one language to another, such as translating from English to Hindi. However, language translation requires massive datasets and usually takes days of training on GPUs. To give you a place to experiment with these models even without using massive datasets, we will instead use a simpler "date translation" task.app
The network will input a date written in a variety of possible formats (e.g. "the 29th of August 1958", "03/30/1968", "24 JUNE 1987") and translate them into standardized, machine readable dates (e.g. "1958-08-29", "1968-03-30", "1987-06-24"). We will have the network learn to output dates in the common machine-readable format YYYY-MM-DD.dom
1.1 - Dataset
We will train the model on a dataset of 10000 human readable dates and their equivalent, standardized, machine readable dates. Let's run the following cells to load the dataset and print some examples.ide
【code】函數
m = 10000 dataset, human_vocab, machine_vocab, inv_machine_vocab = load_dataset(m)
dataset[:10]
【result】oop
[('9 may 1998', '1998-05-09'),
('10.09.70', '1970-09-10'),
('4/28/90', '1990-04-28'),
('thursday january 26 1995', '1995-01-26'),
('monday march 7 1983', '1983-03-07'),
('sunday may 22 1988', '1988-05-22'),
('tuesday july 8 2008', '2008-07-08'),
('08 sep 1999', '1999-09-08'),
('1 jan 1981', '1981-01-01'),
('monday may 22 1995', '1995-05-22')]
You've loaded:
dataset: a list of tuples of (human readable date, machine readable date)human_vocab: a python dictionary mapping all characters used in the human readable dates to an integer-valued indexmachine_vocab: a python dictionary mapping all characters used in machine readable dates to an integer-valued index. These indices are not necessarily consistent withhuman_vocab.inv_machine_vocab: the inverse dictionary ofmachine_vocab, mapping from indices back to characters.
Let's preprocess the data and map the raw text data into the index values. We will also use Tx=30 (which we assume is the maximum length of the human readable date; if we get a longer input, we would have to truncate it) and Ty=10 (since "YYYY-MM-DD" is 10 characters long).
【中文翻譯】
您已加載:
dataset: (人類可讀日期, 機器可讀日期) 的元組列表
human_vocab: python 字典將人類可讀日期中使用的全部字符映射爲整數值索引
machine_vocab: python 字典將機器可讀日期中使用的全部字符映射到整數值索引。這些指數不必定與 human_vocab 一致。
inv_machine_vocab: machine_vocab 的逆字典, 從索引返回到字符的映射。
讓咱們對數據進行預處理, 並將原始文本數據映射到索引值中。咱們也將使用 Tx=30 (咱們假設是人類可讀日期的最大長度; 若是咱們獲得一個更長的輸入, 咱們將不得不截斷它) 和 Ty=10 (由於 "YYYY-MM-DD" 是10個字符長)。
【code】
Tx = 30
Ty = 10
X, Y, Xoh, Yoh = preprocess_data(dataset, human_vocab, machine_vocab, Tx, Ty)
print("X.shape:", X.shape)
print("Y.shape:", Y.shape)
print("Xoh.shape:", Xoh.shape)
print("Yoh.shape:", Yoh.shape)
【result】
X.shape: (10000, 30) Y.shape: (10000, 10) Xoh.shape: (10000, 30, 37) Yoh.shape: (10000, 10, 11)
You now have:
X: a processed version of the human readable dates in the training set, where each character is replaced by an index mapped to the character viahuman_vocab. Each date is further padded to Tx values with a special character (< pad >).X.shape = (m, Tx)Y: a processed version of the machine readable dates in the training set, where each character is replaced by the index it is mapped to inmachine_vocab. You should haveY.shape = (m, Ty).Xoh: one-hot version ofX, the "1" entry's index is mapped to the character thanks tohuman_vocab.Xoh.shape = (m, Tx, len(human_vocab))Yoh: one-hot version ofY, the "1" entry's index is mapped to the character thanks tomachine_vocab.Yoh.shape = (m, Tx, len(machine_vocab)). Here,len(machine_vocab) = 11since there are 11 characters ('-' as well as 0-9).
【中文翻譯】
您如今有:
X: 訓練集中的被人可讀的日期的一個通過處理的版本, 其中每一個字符都由經過 human_vocab 映射到該字符的索引替換。將每一個日期進一步由特殊字符 (< pad >)填充爲Tx 長度。X.shape = (m, Tx)
Y: 在訓練集中的被機器可讀的日期的一個通過處理的版本, 其中每一個字符被由映射到 machine_vocab 中的索引替換。Y.shape = (m, Ty)。
Xoh: one-hot版本的 X, "1 " 條目的索引映射到字符, 多虧了 human_vocab. Xoh.shape = (m, Tx, len(human_vocab))
Yoh: ne-hot版本的 Y, "1 " 條目的索引被映射到字符, 多虧了 machine_vocab. Yoh.shape = (m, Tx, len(machine_vocab))。這裏, len (machine_vocab) = 11, 由於有11個字符 ('-' 而且 0-9)。
Lets also look at some examples of preprocessed training examples. Feel free to play with index in the cell below to navigate the dataset and see how source/target dates are preprocessed.
【code】
index = 0
print("Source date:", dataset[index][0])
print("Target date:", dataset[index][1])
print()
print("Source after preprocessing (indices):", X[index])
print("Target after preprocessing (indices):", Y[index])
print()
print("Source after preprocessing (one-hot):", Xoh[index])
print("Target after preprocessing (one-hot):", Yoh[index])
【result】
Source date: 9 may 1998 Target date: 1998-05-09 Source after preprocessing (indices): [12 0 24 13 34 0 4 12 12 11 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36] Target after preprocessing (indices): [ 2 10 10 9 0 1 6 0 1 10] Source after preprocessing (one-hot): [[ 0. 0. 0. ..., 0. 0. 0.] [ 1. 0. 0. ..., 0. 0. 0.] [ 0. 0. 0. ..., 0. 0. 0.] ..., [ 0. 0. 0. ..., 0. 0. 1.] [ 0. 0. 0. ..., 0. 0. 1.] [ 0. 0. 0. ..., 0. 0. 1.]] Target after preprocessing (one-hot): [[ 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0.] [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.] [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.] [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.] [ 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [ 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [ 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.] [ 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [ 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]]
2 - Neural machine translation with attention
If you had to translate a book's paragraph from French to English, you would not read the whole paragraph, then close the book and translate. Even during the translation process, you would read/re-read and focus on the parts of the French paragraph corresponding to the parts of the English you are writing down.
The attention mechanism tells a Neural Machine Translation model where it should pay attention to at any step.
【中文翻譯】
2- 注意力機制的神經網絡的機器翻譯
若是你必須把一本書的段落從法語翻譯成英語, 你不會讀完整個段落, 而後把書合上並翻譯。即便在翻譯過程當中, 你也會閱讀/從新閱讀和關注與你寫下的英語部分相應的法語段落部分。
attention機制告訴一個神經網絡的機器翻譯模型, 在每一步它應該注意什麼。
2.1 - Attention mechanism
In this part, you will implement the attention mechanism presented in the lecture videos. Here is a figure to remind you how the model works. The diagram on the left shows the attention model. The diagram on the right shows what one "Attention" step does to calculate the attention variables α⟨t,t′⟩, which are used to compute the context variable context⟨t⟩ for each timestep in the output (t=1,…,Ty).
Figure 1: Neural machine translation with attention
Here are some properties of the model that you may notice:
-
There are two separate LSTMs in this model (see diagram on the left). Because the one at the bottom of the picture is a Bi-directional LSTM and comes before the attention mechanism, we will call it pre-attention Bi-LSTM. The LSTM at the top of the diagram comes after the attention mechanism, so we will call it the post-attention LSTM. The pre-attention Bi-LSTM goes through Tx time steps; the post-attention LSTM goes through Ty time steps.
-
The post-attention LSTM passes s⟨t⟩,c⟨t⟩ from one time step to the next. In the lecture videos, we were using only a basic RNN for the post-activation sequence model, so the state captured by the RNN output activations s⟨t⟩. But since we are using an LSTM here, the LSTM has both the output activation s⟨t⟩ and the hidden cell state c⟨t⟩. However, unlike previous text generation examples (such as Dinosaurus in week 1), in this model the post-activation LSTM at time t does will not take the specific generated y⟨t−1⟩ as input; it only takes s⟨t⟩ and c⟨t⟩ as input. We have designed the model this way, because (unlike language generation where adjacent characters are highly correlated) there isn't as strong a dependency between the previous character and the next character in a YYYY-MM-DD date.
-
We use
to represent the concatenation of the activations of both the forward-direction and backward-directions of the pre-attention Bi-LSTM. -
The diagram on the right uses a
RepeatVectornode to copy s⟨t−1⟩'s value Tx times, and thenConcatenationto concatenate s⟨t−1⟩ and a⟨t⟩ to compute e⟨t,t′⟩, which is then passed through a softmax to compute α⟨t,t′⟩. We'll explain how to useRepeatVectorandConcatenationin Keras below.
【中文翻譯】
下面是您可能注意到的模型的一些屬性:
- 此模型中有兩個單獨的 LSTMs (請參見左側的圖示)。由於在圖片底部的一個是雙向 LSTM, 並出如今attention機制以前, 因此咱們將它稱爲 "pre-attention Bi-LSTM"。在圖的頂端的LSTM 是在attention機制以後, 因此咱們將它稱爲post-attention LSTM。The pre-attention Bi-LSTM經過 Tx時間步驟;the post-attention LSTM 經過 Ty 時間步驟。
- The post-attention LSTM經過 s⟨t⟩,c⟨t⟩從一個時間步驟到下一個。在講座視頻中, 咱們只使用了the post-activation sequence model的基本 RNN, 所以 RNN 輸出激活 s⟨t⟩捕獲的狀態。可是, 因爲咱們在這裏使用的 LSTM, LSTM 有輸出激活 s⟨t⟩和隱藏的細胞狀態 c⟨t⟩。可是, 與之前的文本生成示例 (如1週中的 Dinosaurus) 不一樣, 在這個模型中, the post-activation LSTM在t時刻將不會被 y⟨t−1⟩ 做爲輸入;它只須要 s⟨t⟩ 和 c⟨t⟩做爲輸入。咱們用這種方式設計了這個模型, 由於 (與相鄰字符高度關聯的語言生成不一樣), 之前的字符和在一個日期內的下一個字符之間沒有太強的依賴性。
- 咱們使用來表示the pre-attention Bi-LSTM的向前方向和向後方向的激活的串聯。
- 右側的關係圖使用 RepeatVector 節點複製s⟨t−1⟩的值 Tx次, 而後串聯以鏈接 s⟨t−1⟩ 和 a⟨t⟩以計算 e⟨t,t′⟩, 而後經過 softmax 傳遞來計算α⟨t,t′⟩。咱們將在下面的 Keras中解釋如何使用 RepeatVector 和串聯。
Lets implement this model. You will start by implementing two functions: one_step_attention() and model().
1) one_step_attention(): At step t, given all the hidden states of the Bi-LSTM ([a<1>,a<2>,...,a<Tx>]) and the previous hidden state of the second LSTM (s<t−1>), one_step_attention() will compute the attention weights ([α<t,1>,α<t,2>,...,α<t,Tx>]) and output the context vector (see Figure 1 (right) for details):
Note that we are denoting the attention in this notebook context⟨t⟩. In the lecture videos, the context was denoted c⟨t⟩, but here we are calling it context⟨t⟩ to avoid confusion with the (post-attention) LSTM's internal memory cell variable, which is sometimes also denoted c⟨t⟩.
2) model(): Implements the entire model. It first runs the input through a Bi-LSTM to get back [a<1>,a<2>,...,a<Tx>]). Then, it calls one_step_attention() Ty times (for loop). At each iteration of this loop, it gives the computed context vector c<t> to the second LSTM, and runs the output of the LSTM through a dense layer with softmax activation to generate a prediction ŷ <t>.
【中文翻譯】
讓咱們實現這個模型。您將首先實現兩個函數: one_step_attention() 和 model()。
1) one_step_attention (): 在步驟 t 中, 考慮到 the Bi-LSTM的全部隱藏狀態 ([a<1>,a<2>,...,a<Tx>])以及第二個 LSTM (s) 的前一個隱藏狀態(s<t−1>)， one_step_attention () 將計算attention weights([α<t,1>,α<t,2>,...,α<t,Tx>])和輸出the context vector (請參見圖 1 (右) 以瞭解詳細信息):
注意, 咱們在這本筆記本中用context⟨t⟩ 表示the attention。在講座視頻中, the context被表示爲 c⟨t⟩, 但在這裏咱們稱之爲 context⟨t⟩, 以免混淆 (post-attention) LSTM 的內部內存單元格變量, 這有時也被表示爲 c⟨t⟩。
2) model(): 實現整個模型。它首先經過Bi-LSTM 運行輸入以返回[a<1>,a<2>,...,a<Tx>])。而後, 它調用one_step_attention() Ty次。（for 循環)。在該循環的每一個迭代中, 它將計算context vector c<t>，賦予第二個 LSTM, 並經過具備 softmax 激活的dense層運行 LSTM 的輸出, 以生成預測ŷ <t>.
Exercise: Implement one_step_attention(). The function model() will call the layers in one_step_attention() Ty using a for-loop, and it is important that all Ty copies have the same weights. I.e., it should not re-initiaiize the weights every time. In other words, all Ty steps should have shared weights. Here's how you can implement layers with shareable weights in Keras:
- Define the layer objects (as global variables for examples).
- Call these objects when propagating the input.
We have defined the layers you need as global variables. Please run the following cells to create them. Please check the Keras documentation to make sure you understand what these layers are: RepeatVector(), Concatenate(), Dense(), Activation(), Dot().
【中文翻譯】
練習: 實施 one_step_attention ()。函數 model() 將使用 for 循環調用 one_step_attention ()Ty 次， 而且全部的 Ty 拷貝都具備相同的權重是很重要的。也就是說, 它不該該每次從新初始化。換言之, 全部的 Ty 步驟都應該具備共享權重。下面是如何在 Keras 中實現可共享權重的層:
1.定義層對象 (做爲示例的全局變量)。
2.傳播輸入時調用這些對象。
咱們已將須要的層定義爲全局變量。請運行如下單元格以建立它們。請檢查 Keras 文檔, 以確保瞭解這些圖層的容: RepeatVector(), Concatenate(), Dense(), Activation(), Dot().
【code】
# Defined shared layers as global variables repeator = RepeatVector(Tx) concatenator = Concatenate(axis=-1) densor1 = Dense(10, activation = "tanh") densor2 = Dense(1, activation = "relu") activator = Activation(softmax, name='attention_weights') # We are using a custom softmax(axis = 1) loaded in this notebook dotor = Dot(axes = 1)
Now you can use these layers to implement one_step_attention(). In order to propagate a Keras tensor object X through one of these layers, use layer(X) (or layer([X,Y]) if it requires multiple inputs.), e.g. densor(X) will propagate X through the Dense(1) layer defined above.
【code】
# GRADED FUNCTION: one_step_attention
def one_step_attention(a, s_prev):
"""
Performs one step of attention: Outputs a context vector computed as a dot product of the attention weights
"alphas" and the hidden states "a" of the Bi-LSTM.
Arguments:
a -- hidden state output of the Bi-LSTM, numpy-array of shape (m, Tx, 2*n_a)
s_prev -- previous hidden state of the (post-attention) LSTM, numpy-array of shape (m, n_s)
Returns:
context -- context vector, input of the next (post-attetion) LSTM cell
"""
### START CODE HERE ###
# Use repeator to repeat s_prev to be of shape (m, Tx, n_s) so that you can concatenate it with all hidden states "a" (≈ 1 line)
s_prev = repeator(s_prev)
# Use concatenator to concatenate a and s_prev on the last axis (≈ 1 line)
concat = concatenator([a,s_prev])
# Use densor1 to propagate concat through a small fully-connected neural network to compute the "intermediate energies" variable e. (≈1 lines)
e = densor1(concat)
# Use densor2 to propagate e through a small fully-connected neural network to compute the "energies" variable energies. (≈1 lines)
energies = densor2(e)
# Use "activator" on "energies" to compute the attention weights "alphas" (≈ 1 line)
alphas = activator(energies)
# Use dotor together with "alphas" and "a" to compute the context vector to be given to the next (post-attention) LSTM-cell (≈ 1 line)
context = dotor([ alphas,a])
### END CODE HERE ###
return context
You will be able to check the expected output of one_step_attention() after you've coded the model() function.
Exercise: Implement model() as explained in figure 2 and the text above. Again, we have defined global layers that will share weights to be used in model().
【code】
n_a = 32 n_s = 64 post_activation_LSTM_cell = LSTM(n_s, return_state = True) output_layer = Dense(len(machine_vocab), activation=softmax)
Now you can use these layers Ty times in a for loop to generate the outputs, and their parameters will not be reinitialized. You will have to carry out the following steps:
- Propagate the input into a Bidirectional LSTM
-
Iterate for t=0,…,Ty−1:
- Call
one_step_attention()on [α<t,1>,α<t,2>,...,α<t,Tx>] and s<t−1> to get the context vector context<t>. - Give context<t> to the post-attention LSTM cell. Remember pass in the previous hidden-state s⟨t−1⟩ and cell-states c⟨t−1⟩ of this LSTM using
initial_state= [previous hidden state, previous cell state]. Get back the new hidden state s<t> and the new cell state c<t>. - Apply a softmax layer to s<t>, get the output.
- Save the output by adding it to the list of outputs.
- Call
-
Create your Keras model instance, it should have three inputs ("inputs", s<0> and c<0>) and output the list of "outputs".
【code】
# GRADED FUNCTION: model
def model(Tx, Ty, n_a, n_s, human_vocab_size, machine_vocab_size):
"""
Arguments:
Tx -- length of the input sequence
Ty -- length of the output sequence
n_a -- hidden state size of the Bi-LSTM
n_s -- hidden state size of the post-attention LSTM
human_vocab_size -- size of the python dictionary "human_vocab"
machine_vocab_size -- size of the python dictionary "machine_vocab"
Returns:
model -- Keras model instance
"""
# Define the inputs of your model with a shape (Tx,)
# Define s0 and c0, initial hidden state for the decoder LSTM of shape (n_s,)
X = Input(shape=(Tx, human_vocab_size))
s0 = Input(shape=(n_s,), name='s0')
c0 = Input(shape=(n_s,), name='c0')
s = s0
c = c0
# Initialize empty list of outputs
outputs = []
### START CODE HERE ###
# Step 1: Define your pre-attention Bi-LSTM. Remember to use return_sequences=True. (≈ 1 line)
a = Bidirectional(LSTM(n_a, return_sequences = True), input_shape = (m, Tx, n_a*2))(X)
# Step 2: Iterate for Ty steps
for t in range(Ty):
# Step 2.A: Perform one step of the attention mechanism to get back the context vector at step t (≈ 1 line)
context = one_step_attention(a, s)
# Step 2.B: Apply the post-attention LSTM cell to the "context" vector.
# Don't forget to pass: initial_state = [hidden state, cell state] (≈ 1 line)
s, _, c = post_activation_LSTM_cell(context,initial_state = [s, c])
# Step 2.C: Apply Dense layer to the hidden state output of the post-attention LSTM (≈ 1 line)
out = output_layer(s)
# Step 2.D: Append "out" to the "outputs" list (≈ 1 line)
outputs.append(out)
# Step 3: Create model instance taking three inputs and returning the list of outputs. (≈ 1 line)
model = Model([X, s0, c0], outputs = outputs)
### END CODE HERE ###
return model
Run the following cell to create your model.
【code】
model = model(Tx, Ty, n_a, n_s, len(human_vocab), len(machine_vocab))
Let's get a summary of the model to check if it matches the expected output.
【code】
model.summary()
【result】
____________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
====================================================================================================
input_6 (InputLayer) (None, 30, 37) 0
____________________________________________________________________________________________________
s0 (InputLayer) (None, 64) 0
____________________________________________________________________________________________________
bidirectional_6 (Bidirectional) (None, 30, 64) 17920 input_6[0][0]
____________________________________________________________________________________________________
repeat_vector_1 (RepeatVector) (None, 30, 64) 0 s0[0][0]
lstm_9[0][0]
lstm_9[1][0]
lstm_9[2][0]
lstm_9[3][0]
lstm_9[4][0]
lstm_9[5][0]
lstm_9[6][0]
lstm_9[7][0]
lstm_9[8][0]
____________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 30, 128) 0 bidirectional_6[0][0]
repeat_vector_1[2][0]
bidirectional_6[0][0]
repeat_vector_1[3][0]
bidirectional_6[0][0]
repeat_vector_1[4][0]
bidirectional_6[0][0]
repeat_vector_1[5][0]
bidirectional_6[0][0]
repeat_vector_1[6][0]
bidirectional_6[0][0]
repeat_vector_1[7][0]
bidirectional_6[0][0]
repeat_vector_1[8][0]
bidirectional_6[0][0]
repeat_vector_1[9][0]
bidirectional_6[0][0]
repeat_vector_1[10][0]
bidirectional_6[0][0]
repeat_vector_1[11][0]
____________________________________________________________________________________________________
dense_1 (Dense) (None, 30, 10) 1290 concatenate_1[0][0]
concatenate_1[1][0]
concatenate_1[2][0]
concatenate_1[3][0]
concatenate_1[4][0]
concatenate_1[5][0]
concatenate_1[6][0]
concatenate_1[7][0]
concatenate_1[8][0]
concatenate_1[9][0]
____________________________________________________________________________________________________
dense_2 (Dense) (None, 30, 1) 11 dense_1[0][0]
dense_1[1][0]
dense_1[2][0]
dense_1[3][0]
dense_1[4][0]
dense_1[5][0]
dense_1[6][0]
dense_1[7][0]
dense_1[8][0]
dense_1[9][0]
____________________________________________________________________________________________________
attention_weights (Activation) (None, 30, 1) 0 dense_2[0][0]
dense_2[1][0]
dense_2[2][0]
dense_2[3][0]
dense_2[4][0]
dense_2[5][0]
dense_2[6][0]
dense_2[7][0]
dense_2[8][0]
dense_2[9][0]
____________________________________________________________________________________________________
dot_1 (Dot) (None, 1, 64) 0 attention_weights[0][0]
bidirectional_6[0][0]
attention_weights[1][0]
bidirectional_6[0][0]
attention_weights[2][0]
bidirectional_6[0][0]
attention_weights[3][0]
bidirectional_6[0][0]
attention_weights[4][0]
bidirectional_6[0][0]
attention_weights[5][0]
bidirectional_6[0][0]
attention_weights[6][0]
bidirectional_6[0][0]
attention_weights[7][0]
bidirectional_6[0][0]
attention_weights[8][0]
bidirectional_6[0][0]
attention_weights[9][0]
bidirectional_6[0][0]
____________________________________________________________________________________________________
c0 (InputLayer) (None, 64) 0
____________________________________________________________________________________________________
lstm_9 (LSTM) [(None, 64), (None, 6 33024 dot_1[0][0]
s0[0][0]
c0[0][0]
dot_1[1][0]
lstm_9[0][0]
lstm_9[0][2]
dot_1[2][0]
lstm_9[1][0]
lstm_9[1][2]
dot_1[3][0]
lstm_9[2][0]
lstm_9[2][2]
dot_1[4][0]
lstm_9[3][0]
lstm_9[3][2]
dot_1[5][0]
lstm_9[4][0]
lstm_9[4][2]
dot_1[6][0]
lstm_9[5][0]
lstm_9[5][2]
dot_1[7][0]
lstm_9[6][0]
lstm_9[6][2]
dot_1[8][0]
lstm_9[7][0]
lstm_9[7][2]
dot_1[9][0]
lstm_9[8][0]
lstm_9[8][2]
____________________________________________________________________________________________________
dense_6 (Dense) (None, 11) 715 lstm_9[0][0]
lstm_9[1][0]
lstm_9[2][0]
lstm_9[3][0]
lstm_9[4][0]
lstm_9[5][0]
lstm_9[6][0]
lstm_9[7][0]
lstm_9[8][0]
lstm_9[9][0]
====================================================================================================
Total params: 52,960
Trainable params: 52,960
Non-trainable params: 0
____________________________________________________________________________________________________
【Expected Output】
Here is the summary you should see
Total params: 52,960 Trainable params: 52,960 Non-trainable params: 0 bidirectional_1's output shape (None, 30, 64) repeat_vector_1's output shape (None, 30, 64) concatenate_1's output shape (None, 30, 128) attention_weights's output shape (None, 30, 1) dot_1's output shape (None, 1, 64) dense_3's output shape (None, 11)
As usual, after creating your model in Keras, you need to compile it and define what loss, optimizer and metrics your are want to use. Compile your model using categorical_crossentropy loss, a custom Adam optimizer (learning rate = 0.005, β1=0.9, β2=0.999, decay = 0.01) and ['accuracy'] metrics:
【code】
### START CODE HERE ### (≈2 lines) opt = Adam(lr = 0.005, beta_1 = 0.9, beta_2 = 0.999,decay = 0.01) model.compile(loss = 'categorical_crossentropy', optimizer = opt,metrics = ['accuracy']) ### END CODE HERE ###
The last step is to define all your inputs and outputs to fit the model:
- You already have X of shape (m=10000,Tx=30) containing the training examples.
- You need to create
s0andc0to initialize yourpost_activation_LSTM_cellwith 0s. - Given the
model()you coded, you need the "outputs" to be a list of 11 elements of shape (m, T_y). So that:outputs[i][0], ..., outputs[i][Ty]represent the true labels (characters) corresponding to the ithith training example (X[i]). More generally,outputs[i][j]is the true label of the jth character in the ithithtraining example.
【code】
s0 = np.zeros((m, n_s)) c0 = np.zeros((m, n_s)) outputs = list(Yoh.swapaxes(0,1))
Let's now fit the model and run it for one epoch.
【code】
model.fit([Xoh, s0, c0], outputs, epochs=1, batch_size=100)
【result】
Epoch 1/1 10000/10000 [==============================] - 46s - loss: 16.5956 - dense_6_loss_1: 1.2932 - dense_6_loss_2: 1.0162 - dense_6_loss_3: 1.7596 - dense_6_loss_4: 2.6613
- dense_6_loss_5: 0.7426 - dense_6_loss_6: 1.3393 - dense_6_loss_7: 2.6329 - dense_6_loss_8: 0.8744 - dense_6_loss_9: 1.7038 - dense_6_loss_10: 2.5724
- dense_6_acc_1: 0.4418 - dense_6_acc_2: 0.6646 - dense_6_acc_3: 0.2957 - dense_6_acc_4: 0.0854 - dense_6_acc_5: 0.9590 - dense_6_acc_6: 0.3268
- dense_6_acc_7: 0.0647 - dense_6_acc_8: 0.9349 - dense_6_acc_9: 0.2069 - dense_6_acc_10: 0.1007 <keras.callbacks.History at 0x7fdd32eb9b38>
While training you can see the loss as well as the accuracy on each of the 10 positions of the output. The table below gives you an example of what the accuracies could be if the batch had 2 examples:
Thus, dense_2_acc_8: 0.89 means that you are predicting the 7th character of the output correctly 89% of the time in the current batch of data.
We have run this model for longer, and saved the weights. Run the next cell to load our weights. (By training a model for several minutes, you should be able to obtain a model of similar accuracy, but loading our model will save you time.)
【code】
model.load_weights('models/model.h5')
You can now see the results on new examples.
【code】
EXAMPLES = ['3 May 1979', '5 April 09', '21th of August 2016', 'Tue 10 Jul 2007', 'Saturday May 9 2018', 'March 3 2001', 'March 3rd 2001', '1 March 2001']
for example in EXAMPLES:
source = string_to_int(example, Tx, human_vocab)
source = np.array(list(map(lambda x: to_categorical(x, num_classes=len(human_vocab)), source))).swapaxes(0,1)
prediction = model.predict([source, s0, c0])
prediction = np.argmax(prediction, axis = -1)
output = [inv_machine_vocab[int(i)] for i in prediction]
print("source:", example)
print("output:", ''.join(output))
【result】
source: 3 May 1979 output: 1977-07-07 source: 5 April 09 output: 1975-05-07 source: 21th of August 2016 output: 2000-05-05 source: Tue 10 Jul 2007 output: 2000-00-20 source: Saturday May 9 2018 output: 1905-02-05 source: March 3 2001 output: 2000-05-20 source: March 3rd 2001 output: 2000-00-20 source: 1 March 2001 output: 2000-00-20
You can also change these examples to test with your own examples. The next part will give you a better sense on what the attention mechanism is doing--i.e., what part of the input the network is paying attention to when generating a particular output character.
3 - Visualizing Attention (Optional / Ungraded)
Since the problem has a fixed output length of 10, it is also possible to carry out this task using 10 different softmax units to generate the 10 characters of the output. But one advantage of the attention model is that each part of the output (say the month) knows it needs to depend only on a small part of the input (the characters in the input giving the month). We can visualize what part of the output is looking at what part of the input.
【中文翻譯】
因爲該問題具備固定的輸出長度 10, 所以可使用10個不一樣的 softmax 單元來執行此任務, 以生成輸出的10個字符。但注意模型的一個優勢是, 輸出的每一個部分 (好比月份) 都知道它只須要依賴輸入的一小部分 (輸入月份的字符)。咱們能夠可視化輸出的哪一部分正在查看輸入的哪一部分。
Consider the task of translating "Saturday 9 May 2018" to "2018-05-09". If we visualize the computed α⟨t,t′⟩ we get this:
Notice how the output ignores the "Saturday" portion of the input. None of the output timesteps are paying much attention to that portion of the input. We see also that 9 has been translated as 09 and May has been correctly translated into 05, with the output paying attention to the parts of the input it needs to to make the translation. The year mostly requires it to pay attention to the input's "18" in order to generate "2018."
【中文翻譯】
注意輸出如何忽略輸入的 "Saturday " 部分。全部輸出 時間步都不太注意輸入的那一部分。咱們還看到, 9 已被翻譯爲 09, 並May已正確翻譯成 05。年份須要它注意輸入的 "18 " 才能生成 "2018"。
3.1 - Getting the activations from the network
Lets now visualize the attention values in your network. We'll propagate an example through the network, then visualize the values of α⟨t,t′⟩.
To figure out where the attention values are located, let's start by printing a summary of the model .
【code】
model.summary()
【result】
___________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
====================================================================================================
input_6 (InputLayer) (None, 30, 37) 0
____________________________________________________________________________________________________
s0 (InputLayer) (None, 64) 0
____________________________________________________________________________________________________
bidirectional_6 (Bidirectional) (None, 30, 64) 17920 input_6[0][0]
____________________________________________________________________________________________________
repeat_vector_1 (RepeatVector) (None, 30, 64) 0 s0[0][0]
lstm_9[0][0]
lstm_9[1][0]
lstm_9[2][0]
lstm_9[3][0]
lstm_9[4][0]
lstm_9[5][0]
lstm_9[6][0]
lstm_9[7][0]
lstm_9[8][0]
____________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 30, 128) 0 bidirectional_6[0][0]
repeat_vector_1[2][0]
bidirectional_6[0][0]
repeat_vector_1[3][0]
bidirectional_6[0][0]
repeat_vector_1[4][0]
bidirectional_6[0][0]
repeat_vector_1[5][0]
bidirectional_6[0][0]
repeat_vector_1[6][0]
bidirectional_6[0][0]
repeat_vector_1[7][0]
bidirectional_6[0][0]
repeat_vector_1[8][0]
bidirectional_6[0][0]
repeat_vector_1[9][0]
bidirectional_6[0][0]
repeat_vector_1[10][0]
bidirectional_6[0][0]
repeat_vector_1[11][0]
____________________________________________________________________________________________________
dense_1 (Dense) (None, 30, 10) 1290 concatenate_1[0][0]
concatenate_1[1][0]
concatenate_1[2][0]
concatenate_1[3][0]
concatenate_1[4][0]
concatenate_1[5][0]
concatenate_1[6][0]
concatenate_1[7][0]
concatenate_1[8][0]
concatenate_1[9][0]
____________________________________________________________________________________________________
dense_2 (Dense) (None, 30, 1) 11 dense_1[0][0]
dense_1[1][0]
dense_1[2][0]
dense_1[3][0]
dense_1[4][0]
dense_1[5][0]
dense_1[6][0]
dense_1[7][0]
dense_1[8][0]
dense_1[9][0]
____________________________________________________________________________________________________
attention_weights (Activation) (None, 30, 1) 0 dense_2[0][0]
dense_2[1][0]
dense_2[2][0]
dense_2[3][0]
dense_2[4][0]
dense_2[5][0]
dense_2[6][0]
dense_2[7][0]
dense_2[8][0]
dense_2[9][0]
____________________________________________________________________________________________________
dot_1 (Dot) (None, 1, 64) 0 attention_weights[0][0]
bidirectional_6[0][0]
attention_weights[1][0]
bidirectional_6[0][0]
attention_weights[2][0]
bidirectional_6[0][0]
attention_weights[3][0]
bidirectional_6[0][0]
attention_weights[4][0]
bidirectional_6[0][0]
attention_weights[5][0]
bidirectional_6[0][0]
attention_weights[6][0]
bidirectional_6[0][0]
attention_weights[7][0]
bidirectional_6[0][0]
attention_weights[8][0]
bidirectional_6[0][0]
attention_weights[9][0]
bidirectional_6[0][0]
____________________________________________________________________________________________________
c0 (InputLayer) (None, 64) 0
____________________________________________________________________________________________________
lstm_9 (LSTM) [(None, 64), (None, 6 33024 dot_1[0][0]
s0[0][0]
c0[0][0]
dot_1[1][0]
lstm_9[0][0]
lstm_9[0][2]
dot_1[2][0]
lstm_9[1][0]
lstm_9[1][2]
dot_1[3][0]
lstm_9[2][0]
lstm_9[2][2]
dot_1[4][0]
lstm_9[3][0]
lstm_9[3][2]
dot_1[5][0]
lstm_9[4][0]
lstm_9[4][2]
dot_1[6][0]
lstm_9[5][0]
lstm_9[5][2]
dot_1[7][0]
lstm_9[6][0]
lstm_9[6][2]
dot_1[8][0]
lstm_9[7][0]
lstm_9[7][2]
dot_1[9][0]
lstm_9[8][0]
lstm_9[8][2]
____________________________________________________________________________________________________
dense_6 (Dense) (None, 11) 715 lstm_9[0][0]
lstm_9[1][0]
lstm_9[2][0]
lstm_9[3][0]
lstm_9[4][0]
lstm_9[5][0]
lstm_9[6][0]
lstm_9[7][0]
lstm_9[8][0]
lstm_9[9][0]
====================================================================================================
Total params: 52,960
Trainable params: 52,960
Non-trainable params: 0
____________________________________________________________________________________________________
Navigate through the output of model.summary() above. You can see that the layer named attention_weights outputs the alphas of shape (m, 30, 1) before dot_2 computes the context vector for every time step t=0,…,Ty−1. Lets get the activations from this layer.
【中文翻譯】
在 model.summary()的輸出中導航。您能夠看到名爲 attention_weights 的層在 dot_1 計算每一個時間步驟 t=0,..., Ty−1的上下文向量以前, 輸出形狀爲 (m、30、1)的alphas。讓咱們從這個層獲得激活。
The function attention_map() pulls out the attention values from your model and plots them.
【code】
attention_map = plot_attention_map(model, human_vocab, inv_machine_vocab, "Tuesday 09 Oct 1993", num = 7, n_s = 64)
【result】
On the generated plot you can observe the values of the attention weights for each character of the predicted output. Examine this plot and check that where the network is paying attention makes sense to you.
In the date translation application, you will observe that most of the time attention helps predict the year, and hasn't much impact on predicting the day/month.
【中文翻譯】
在生成的圖形上, 您能夠觀察預測輸出的每一個字符的注意權重值。檢查這個情節並檢查網絡關注的地方是否對你有意義。
在日期翻譯應用程序中, 您將觀察到大多數時間的關注有助於預測年份, 而且對預測日/月沒有太大影響。
Congratulations!
You have come to the end of this assignment
Here's what you should remember from this notebook:
- Machine translation models can be used to map from one sequence to another. They are useful not just for translating human languages (like French->English) but also for tasks like date format translation.
- An attention mechanism allows a network to focus on the most relevant parts of the input when producing a specific part of the output.
- A network using an attention mechanism can translate from inputs of length Tx to outputs of length Ty , where Tx and Ty can be different.
- You can visualize attention weights α⟨t,t′⟩ to see what the network is paying attention to while generating each output.
Congratulations on finishing this assignment! You are now able to implement an attention model and use it to learn complex mappings from one sequence to another.
【中文翻譯】
祝賀！
你已經完成了這個任務！
如下是您在本筆記本中應該記住的內容:
機器翻譯模型可用於從一個序列映射到另外一個序列。它們不只用於翻譯人類語言 (如法語-英語), 並且對於諸如日期格式翻譯之類的任務也頗有用。
注意機制容許網絡在生成特定部分輸出時, 將焦點放在輸入的最相關部分。
使用注意機制的網絡能夠從長度 Tx 的輸入轉換爲長度 Ty 的輸出, 在那裏Tx 和 Ty 能夠不一樣。
您能夠可視化注意權重α⟨t, t′⟩,在生成每一個輸出時查看網絡所關注的內容。
恭喜你完成了任務!如今, 您能夠實現一個注意模型, 並使用學習複雜映射，它從一個序列映射到另外一個序列。
- 1. 課程五(Sequence Models)，第三週（Sequence models & Attention mechanism） —— 2.Programming assignments：Trigger
- 2. Sequence Models(Week3)---Sequence models & Attention mechanism
- 3. 5. Sequence Models - Wee 3 - Sequence models & Attention mechanism
- 4. Sequence Models-第三週
- 5. [coursera/SequenceModels/week3]Sequence models & Attention mechanism (summary&question)
- 6. Sequence Models-第二週
- 7. Sequence Models-第一週
- 8. 吳恩達Deeplearning.ai 第五課 Sequence Model 第一週------Why sequence Models
- 9. 021Sequence models & Attention mechanism
- 10. Sequence Models
- 更多相關文章...
- • XML Schema sequence 元素 - XML Schema 教程
- • XSL-FO page-sequence 對象 - XSL-FO 教程
- • Git五分鐘教程
- • 使用Rxjava計算圓周率
-
每一个你不满意的现在，都有一个你没有努力的曾经。
- 1. 課程五(Sequence Models)，第三週（Sequence models & Attention mechanism） —— 2.Programming assignments：Trigger
- 2. Sequence Models(Week3)---Sequence models & Attention mechanism
- 3. 5. Sequence Models - Wee 3 - Sequence models & Attention mechanism
- 4. Sequence Models-第三週
- 5. [coursera/SequenceModels/week3]Sequence models & Attention mechanism (summary&question)
- 6. Sequence Models-第二週
- 7. Sequence Models-第一週
- 8. 吳恩達Deeplearning.ai 第五課 Sequence Model 第一週------Why sequence Models
- 9. 021Sequence models & Attention mechanism
- 10. Sequence Models