Skip to content
This repository has been archived by the owner on Jan 24, 2024. It is now read-only.

Commit

Permalink
update the Chinese version README.
Browse files Browse the repository at this point in the history
  • Loading branch information
@lcy-seso
lcy-seso committed Jul 7, 2017
1 parent e646292 commit 6326528
Show file tree
Hide file tree
Showing 3 changed files with 110 additions and 118 deletions.
112 changes: 54 additions & 58 deletions 08.machine_translation/README.cn.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -185,16 +185,16 @@ is_generating = False
### 模型结构
1. 首先，定义了一些全局变量。

```python
dict_size = 30000 # 字典维度
source_dict_dim = dict_size # 源语言字典维度
target_dict_dim = dict_size # 目标语言字典维度
word_vector_dim = 512 # 词向量维度
encoder_size = 512 # 编码器中的GRU隐层大小
decoder_size = 512 # 解码器中的GRU隐层大小
beam_size = 3 # 柱宽度
max_length = 250 # 生成句子的最大长度
```
```python
dict_size = 30000 # 字典维度
source_dict_dim = dict_size # 源语言字典维度
target_dict_dim = dict_size # 目标语言字典维度
word_vector_dim = 512 # 词向量维度
encoder_size = 512 # 编码器中的GRU隐层大小
decoder_size = 512 # 解码器中的GRU隐层大小
beam_size = 3 # 柱宽度
max_length = 250 # 生成句子的最大长度
```

2. 其次，实现编码器框架。分为三步：

Expand All @@ -209,9 +209,7 @@ is_generating = False

```python
src_embedding = paddle.layer.embedding(
input=src_word_id,
size=word_vector_dim,
param_attr=paddle.attr.ParamAttr(name='_source_language_embedding'))
input=src_word_id, size=word_vector_dim)
```
- 用双向GRU编码源语言序列，拼接两个GRU的编码结果得到$\mathbf{h}$。

Expand All @@ -228,19 +226,22 @@ is_generating = False
- 对源语言序列编码后的结果（见2的最后一步），过一个前馈神经网络（Feed Forward Neural Network），得到其映射。

```python
encoded_proj = paddle.layer.mixed(
size=decoder_size,
input=paddle.layer.full_matrix_projection(encoded_vector))
encoded_proj = paddle.layer.fc(
act=paddle.activation.Linear(),
size=decoder_size,
bias_attr=False,
input=encoded_vector)
```

- 构造解码器RNN的初始状态。由于解码器需要预测时序目标序列，但在0时刻并没有初始值，所以我们希望对其进行初始化。这里采用的是将源语言序列逆序编码后的最后一个状态进行非线性映射，作为该初始值，即$c_0=h_T$。

```python
backward_first = paddle.layer.first_seq(input=src_backward)
decoder_boot = paddle.layer.mixed(
size=decoder_size,
act=paddle.activation.Tanh(),
input=paddle.layer.full_matrix_projection(backward_first))
decoder_boot = paddle.layer.fc(
size=decoder_size,
act=paddle.activation.Tanh(),
bias_attr=False,
input=backward_first)
```

- 定义解码阶段每一个时间步的RNN行为，即根据当前时刻的源语言上下文向量$c_i$、解码器隐层状态$z_i$和目标语言中第$i$个词$u_i$，来预测第$i+1$个词的概率$p_{i+1}$。
Expand All @@ -260,12 +261,13 @@ is_generating = False
encoded_proj=enc_proj,
decoder_state=decoder_mem)

decoder_inputs = paddle.layer.mixed(
decoder_inputs = paddle.layer.fc(
act=paddle.activation.Linear(),
size=decoder_size * 3,
input=[
paddle.layer.full_matrix_projection(input=context),
paddle.layer.full_matrix_projection(input=current_word)
])
bias_attr=False,
input=[context, current_word],
layer_attr=paddle.attr.ExtraLayerAttribute(
error_clipping_threshold=100.0))

gru_step = paddle.layer.gru_step(
name='gru_decoder',
Expand All @@ -285,8 +287,8 @@ is_generating = False

```python
decoder_group_name = "decoder_group"
group_input1 = paddle.layer.StaticInput(input=encoded_vector, is_seq=True)
group_input2 = paddle.layer.StaticInput(input=encoded_proj, is_seq=True)
group_input1 = paddle.layer.StaticInput(input=encoded_vector)
group_input2 = paddle.layer.StaticInput(input=encoded_proj)
group_inputs = [group_input1, group_input2]
```

Expand All @@ -301,7 +303,7 @@ is_generating = False
if not is_generating:
trg_embedding = paddle.layer.embedding(
input=paddle.layer.data(
name='target_language_word',
name='target_language_word',
type=paddle.data_type.integer_value_sequence(target_dict_dim)),
size=word_vector_dim,
param_attr=paddle.attr.ParamAttr(name='_target_language_embedding'))
Expand Down Expand Up @@ -330,14 +332,13 @@ is_generating = False

```python
if is_generating:
# In generation, the decoder predicts a next target word based on
# the encoded source sequence and the last generated target word.
# In generation, the decoder predicts a next target word based on
# the encoded source sequence and the previous generated target word.

# The encoded source sequence (encoder's output) must be specified by
# StaticInput, which is a read-only memory.
# Embedding of the last generated word is automatically gotten by
# GeneratedInputs, which is initialized by a start mark, such as <s>,
# and must be included in generation.
# The encoded source sequence (encoder's output) must be specified by
# StaticInput, which is a read-only memory.
# Embedding of the previous generated word is automatically retrieved
# by GeneratedInputs initialized by a start mark <s>.

trg_embedding = paddle.layer.GeneratedInput(
size=target_dict_dim,
Expand Down Expand Up @@ -468,36 +469,31 @@ is_generating = False

```python
if is_generating:
# get the dictionary
# load the dictionary
src_dict, trg_dict = paddle.dataset.wmt14.get_dict(dict_size)

# the delimited element of generated sequences is -1,
# the first element of each generated sequence is the sequence length
seq_list = []
seq = []
for w in beam_result[1]:
if w != -1:
seq.append(w)
else:
seq_list.append(' '.join([trg_dict.get(w) for w in seq[1:]]))
seq = []

prob = beam_result[0]
for i in xrange(gen_num):
print "\n*******************************************************\n"
print "src:", ' '.join(
[src_dict.get(w) for w in gen_data[i][0]]), "\n"
gen_sen_idx = np.where(beam_result[1] == -1)[0]
assert len(gen_sen_idx) == len(gen_data) * beam_size

# -1 is the delimiter of generated sequences.
# the first element of each generated sequence its length.
start_pos, end_pos = 1, 0
for i, sample in enumerate(gen_data):
print(" ".join([src_dict[w] for w in sample[0][1:-1]]))
for j in xrange(beam_size):
print "prob = %f:" % (prob[i][j]), seq_list[i * beam_size + j]
end_pos = gen_sen_idx[i * beam_size + j]
print("%.4f\t%s" % (beam_result[0][i][j], " ".join(
trg_dict[w] for w in beam_result[1][start_pos:end_pos])))
start_pos = end_pos + 2
print("\n")
```

生成开始后，可以观察到输出的日志如下：
```text
src: <s> Les <unk> se <unk> au sujet de la largeur des sièges alors que de grosses commandes sont en jeu <e>
prob = -19.019573: The <unk> will be rotated about the width of the seats , while large orders are at stake . <e>
prob = -19.113066: The <unk> will be rotated about the width of the seats , while large commands are at stake . <e>
prob = -19.512890: The <unk> will be rotated about the width of the seats , while large commands are at play . <e>
Les <unk> se <unk> au sujet de la largeur des sièges alors que de grosses commandes sont en jeu
-19.0196 The <unk> will be rotated about the width of the seats , while large orders are at stake . <e>
-19.1131 The <unk> will be rotated about the width of the seats , while large commands are at stake . <e>
-19.5129 The <unk> will be rotated about the width of the seats , while large commands are at play . <e>
```

## 总结
Expand Down
112 changes: 54 additions & 58 deletions 08.machine_translation/index.cn.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -227,16 +227,16 @@
### 模型结构
1. 首先，定义了一些全局变量。

```python
dict_size = 30000 # 字典维度
source_dict_dim = dict_size # 源语言字典维度
target_dict_dim = dict_size # 目标语言字典维度
word_vector_dim = 512 # 词向量维度
encoder_size = 512 # 编码器中的GRU隐层大小
decoder_size = 512 # 解码器中的GRU隐层大小
beam_size = 3 # 柱宽度
max_length = 250 # 生成句子的最大长度
```
```python
dict_size = 30000 # 字典维度
source_dict_dim = dict_size # 源语言字典维度
target_dict_dim = dict_size # 目标语言字典维度
word_vector_dim = 512 # 词向量维度
encoder_size = 512 # 编码器中的GRU隐层大小
decoder_size = 512 # 解码器中的GRU隐层大小
beam_size = 3 # 柱宽度
max_length = 250 # 生成句子的最大长度
```

2. 其次，实现编码器框架。分为三步：

Expand All @@ -251,9 +251,7 @@

```python
src_embedding = paddle.layer.embedding(
input=src_word_id,
size=word_vector_dim,
param_attr=paddle.attr.ParamAttr(name='_source_language_embedding'))
input=src_word_id, size=word_vector_dim)
```
- 用双向GRU编码源语言序列，拼接两个GRU的编码结果得到$\mathbf{h}$。

Expand All @@ -270,19 +268,22 @@
- 对源语言序列编码后的结果（见2的最后一步），过一个前馈神经网络（Feed Forward Neural Network），得到其映射。

```python
encoded_proj = paddle.layer.mixed(
size=decoder_size,
input=paddle.layer.full_matrix_projection(encoded_vector))
encoded_proj = paddle.layer.fc(
act=paddle.activation.Linear(),
size=decoder_size,
bias_attr=False,
input=encoded_vector)
```

- 构造解码器RNN的初始状态。由于解码器需要预测时序目标序列，但在0时刻并没有初始值，所以我们希望对其进行初始化。这里采用的是将源语言序列逆序编码后的最后一个状态进行非线性映射，作为该初始值，即$c_0=h_T$。

```python
backward_first = paddle.layer.first_seq(input=src_backward)
decoder_boot = paddle.layer.mixed(
size=decoder_size,
act=paddle.activation.Tanh(),
input=paddle.layer.full_matrix_projection(backward_first))
decoder_boot = paddle.layer.fc(
size=decoder_size,
act=paddle.activation.Tanh(),
bias_attr=False,
input=backward_first)
```

- 定义解码阶段每一个时间步的RNN行为，即根据当前时刻的源语言上下文向量$c_i$、解码器隐层状态$z_i$和目标语言中第$i$个词$u_i$，来预测第$i+1$个词的概率$p_{i+1}$。
Expand All @@ -302,12 +303,13 @@
encoded_proj=enc_proj,
decoder_state=decoder_mem)

decoder_inputs = paddle.layer.mixed(
decoder_inputs = paddle.layer.fc(
act=paddle.activation.Linear(),
size=decoder_size * 3,
input=[
paddle.layer.full_matrix_projection(input=context),
paddle.layer.full_matrix_projection(input=current_word)
])
bias_attr=False,
input=[context, current_word],
layer_attr=paddle.attr.ExtraLayerAttribute(
error_clipping_threshold=100.0))

gru_step = paddle.layer.gru_step(
name='gru_decoder',
Expand All @@ -327,8 +329,8 @@

```python
decoder_group_name = "decoder_group"
group_input1 = paddle.layer.StaticInput(input=encoded_vector, is_seq=True)
group_input2 = paddle.layer.StaticInput(input=encoded_proj, is_seq=True)
group_input1 = paddle.layer.StaticInput(input=encoded_vector)
group_input2 = paddle.layer.StaticInput(input=encoded_proj)
group_inputs = [group_input1, group_input2]
```

Expand All @@ -343,7 +345,7 @@
if not is_generating:
trg_embedding = paddle.layer.embedding(
input=paddle.layer.data(
name='target_language_word',
name='target_language_word',
type=paddle.data_type.integer_value_sequence(target_dict_dim)),
size=word_vector_dim,
param_attr=paddle.attr.ParamAttr(name='_target_language_embedding'))
Expand Down Expand Up @@ -372,14 +374,13 @@

```python
if is_generating:
# In generation, the decoder predicts a next target word based on
# the encoded source sequence and the last generated target word.
# In generation, the decoder predicts a next target word based on
# the encoded source sequence and the previous generated target word.

# The encoded source sequence (encoder's output) must be specified by
# StaticInput, which is a read-only memory.
# Embedding of the last generated word is automatically gotten by
# GeneratedInputs, which is initialized by a start mark, such as <s>,
# and must be included in generation.
# The encoded source sequence (encoder's output) must be specified by
# StaticInput, which is a read-only memory.
# Embedding of the previous generated word is automatically retrieved
# by GeneratedInputs initialized by a start mark <s>.

trg_embedding = paddle.layer.GeneratedInput(
size=target_dict_dim,
Expand Down Expand Up @@ -510,36 +511,31 @@

```python
if is_generating:
# get the dictionary
# load the dictionary
src_dict, trg_dict = paddle.dataset.wmt14.get_dict(dict_size)

# the delimited element of generated sequences is -1,
# the first element of each generated sequence is the sequence length
seq_list = []
seq = []
for w in beam_result[1]:
if w != -1:
seq.append(w)
else:
seq_list.append(' '.join([trg_dict.get(w) for w in seq[1:]]))
seq = []

prob = beam_result[0]
for i in xrange(gen_num):
print "\n*******************************************************\n"
print "src:", ' '.join(
[src_dict.get(w) for w in gen_data[i][0]]), "\n"
gen_sen_idx = np.where(beam_result[1] == -1)[0]
assert len(gen_sen_idx) == len(gen_data) * beam_size

# -1 is the delimiter of generated sequences.
# the first element of each generated sequence its length.
start_pos, end_pos = 1, 0
for i, sample in enumerate(gen_data):
print(" ".join([src_dict[w] for w in sample[0][1:-1]]))
for j in xrange(beam_size):
print "prob = %f:" % (prob[i][j]), seq_list[i * beam_size + j]
end_pos = gen_sen_idx[i * beam_size + j]
print("%.4f\t%s" % (beam_result[0][i][j], " ".join(
trg_dict[w] for w in beam_result[1][start_pos:end_pos])))
start_pos = end_pos + 2
print("\n")
```

生成开始后，可以观察到输出的日志如下：
```text
src: <s> Les <unk> se <unk> au sujet de la largeur des sièges alors que de grosses commandes sont en jeu <e>

prob = -19.019573: The <unk> will be rotated about the width of the seats , while large orders are at stake . <e>
prob = -19.113066: The <unk> will be rotated about the width of the seats , while large commands are at stake . <e>
prob = -19.512890: The <unk> will be rotated about the width of the seats , while large commands are at play . <e>
Les <unk> se <unk> au sujet de la largeur des sièges alors que de grosses commandes sont en jeu
-19.0196 The <unk> will be rotated about the width of the seats , while large orders are at stake . <e>
-19.1131 The <unk> will be rotated about the width of the seats , while large commands are at stake . <e>
-19.5129 The <unk> will be rotated about the width of the seats , while large commands are at play . <e>
```

## 总结
Expand Down
4 changes: 2 additions & 2 deletions 08.machine_translation/train.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -136,8 +136,8 @@ def gru_decoder_with_attention(enc_vec, enc_proj, current_word):


def main():
paddle.init(use_gpu=True, trainer_count=1)
is_generating = True
paddle.init(use_gpu=False, trainer_count=1)
is_generating = False

# source and target dict dim.
dict_size = 30000
Expand Down

0 comments on commit 6326528

Please sign in to comment.