Skip to content

M
models

PaddlePaddle / models
大约 1 年 前同步成功

通知 222
Star 6828
Fork 2962
M
models
提交 44a72d3d 编写于 作者: C caoying03
浏览文件
操作

clean codes and add comments.

上级 431f46fb
显示空白变更内容
内联 并排
Showing with 736 addition and 192 deletion
globally_normalized_reader/README.md
浏览文件 @ 44a72d3d
TBD # Globally Normalized Reader
This model implements the work in the following paper:
Jonathan Raiman and John Miller. Globally Normalized Reader. Empirical Methods in Natural Language Processing (EMNLP), 2017.
If you use the dataset/code in your research, please cite the above paper:
```text
@inproceedings{raiman2015gnr,
author={Raiman, Jonathan and Miller, John},
booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
title={Globally Normalized Reader},
year={2017},
}
```
You can also visit https://github.com/baidu-research/GloballyNormalizedReader to get more information.
# Installation
1. Please use [docker image](http://doc.paddlepaddle.org/develop/doc/getstarted/build_and_install/docker_install_en.html) to install the latest PaddlePaddle, by running:
```bash
docker pull paddledev/paddle
```
2. Download all necessary data by running:
```bash
cd data && ./download.sh
```
3. Featurize the data by running:
```
python featurize.py --datadir data --outdir featurized
```
# Training a Model
- Configurate the model by modifying `config.py` if needed, and then run:
```bash
python train.py 2>&1 | tee train.log
```
# Inferring by a Trained Model
- Infer by a trained model by running:
```bash
python infer.py \
--model_path models/pass_00000.tar.gz \
--data_dir data/featurized/ \
--batch_size 2 \
--use_gpu 0 \
--trainer_count 1 \
2>&1 | tee infer.log
```
globally_normalized_reader/basic_modules.py 100755 → 100644
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
import collections import collections
import paddle.v2 as paddle import paddle.v2 as paddle
...@@ -7,11 +8,43 @@ from paddle.v2.layer import parse_network ...@@ -7,11 +8,43 @@ from paddle.v2.layer import parse_network
__all__ = [ __all__ = [
"stacked_bidirectional_lstm", "stacked_bidirectional_lstm",
"stacked_bidirectional_lstm_by_nested_seq",
"lstm_by_nested_sequence", "lstm_by_nested_sequence",
] ]
def stacked_bidirectional_lstm(inputs, size, depth, drop_rate=0., prefix=""): def stacked_bidirectional_lstm(inputs,
hidden_dim,
depth,
drop_rate=0.,
prefix=""):
""" The stacked bi-directional LSTM.
In PaddlePaddle recurrent layers have two different implementations:
1. recurrent layer implemented by recurrent_group: any intermedia states a
recurent unit computes during one time step, such as hidden states,
input-to-hidden mapping, memory cells and so on, is accessable.
2. recurrent layer as a whole: only outputs of the recurrent layer are
accessable.
The second type (recurrent layer as a whole) is more computation efficient,
because recurrent_group is made up of many basic layers (including add,
element-wise multiplications, matrix multiplication and so on).
This function uses the second type to implement the stacked bi-directional
LSTM.
Arguments:
- inputs: The input layer to the bi-directional LSTM.
- hidden_dim: The dimension of the hidden state of the LSTM.
- depth: Depth of the stacked bi-directional LSTM.
- drop_rate: The drop rate to drop the LSTM output states.
- prefix: A string which will be appended to name of each layer
created in this function. Each layer in a network should
has a unique name. The prefix makes this fucntion can be
called multiple times.
"""
if not isinstance(inputs, collections.Sequence): if not isinstance(inputs, collections.Sequence):
inputs = [inputs] inputs = [inputs]
...@@ -20,7 +53,7 @@ def stacked_bidirectional_lstm(inputs, size, depth, drop_rate=0., prefix=""): ...@@ -20,7 +53,7 @@ def stacked_bidirectional_lstm(inputs, size, depth, drop_rate=0., prefix=""):
for i in range(depth): for i in range(depth):
input_proj = paddle.layer.mixed( input_proj = paddle.layer.mixed(
name="%s_in_proj_%0d_%s__" % (prefix, i, dirt), name="%s_in_proj_%0d_%s__" % (prefix, i, dirt),
size=size * 4, size=hidden_dim * 4,
bias_attr=paddle.attr.Param(initial_std=0.), bias_attr=paddle.attr.Param(initial_std=0.),
input=[paddle.layer.full_matrix_projection(lstm)] if i else [ input=[paddle.layer.full_matrix_projection(lstm)] if i else [
paddle.layer.full_matrix_projection(in_layer) paddle.layer.full_matrix_projection(in_layer)
...@@ -45,8 +78,8 @@ def stacked_bidirectional_lstm(inputs, size, depth, drop_rate=0., prefix=""): ...@@ -45,8 +78,8 @@ def stacked_bidirectional_lstm(inputs, size, depth, drop_rate=0., prefix=""):
def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False): def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False):
''' """This is a LSTM implemended by nested recurrent_group.
This is a LSTM implemended by nested recurrent_group.
Paragraph is a nature nested sequence: Paragraph is a nature nested sequence:
1. each paragraph is a sequence of sentence. 1. each paragraph is a sequence of sentence.
2. each sentence is a sequence of words. 2. each sentence is a sequence of words.
...@@ -60,7 +93,14 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False): ...@@ -60,7 +93,14 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False):
5. Consequently, this function is just equivalent to concatenate all 5. Consequently, this function is just equivalent to concatenate all
sentences in a paragraph into one (long) sentence, and use one LSTM to sentences in a paragraph into one (long) sentence, and use one LSTM to
encode this new long sentence. encode this new long sentence.
'''
Arguments:
- input_layer: The input layer to the bi-directional LSTM.
- hidden_dim: The dimension of the hidden state of the LSTM.
- name: The name of the bi-directional LSTM.
- reverse: The boolean parameter indicating whether to prcess
the input sequence by the reverse order.
"""
def lstm_outer_step(lstm_group_input, hidden_dim, reverse, name=''): def lstm_outer_step(lstm_group_input, hidden_dim, reverse, name=''):
outer_memory = paddle.layer.memory( outer_memory = paddle.layer.memory(
...@@ -71,9 +111,8 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False): ...@@ -71,9 +111,8 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False):
name="__inner_state_%s__" % name, name="__inner_state_%s__" % name,
size=hidden_dim, size=hidden_dim,
boot_layer=outer_memory) boot_layer=outer_memory)
input_proj = paddle.layer.fc(size=hidden_dim * 4, input_proj = paddle.layer.fc(
bias_attr=False, size=hidden_dim * 4, bias_attr=False, input=input_layer)
input=input_layer)
return paddle.networks.lstmemory_unit( return paddle.networks.lstmemory_unit(
input=input_proj, input=input_proj,
name="__inner_state_%s__" % name, name="__inner_state_%s__" % name,
...@@ -111,7 +150,27 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False): ...@@ -111,7 +150,27 @@ def lstm_by_nested_sequence(input_layer, hidden_dim, name="", reverse=False):
reverse=reverse) reverse=reverse)
def stacked_bi_lstm_by_nested_seq(input_layer, depth, hidden_dim, prefix=""): def stacked_bidirectional_lstm_by_nested_seq(input_layer,
depth,
hidden_dim,
prefix=""):
""" The stacked bi-directional LSTM to process a nested sequence.
The modules defined in this function is exactly equivalent to
that defined in stacked_bidirectional_lstm, the only difference is the
bi-directional LSTM defined in this function implemented by recurrent_group
in PaddlePaddle, and receive a nested sequence as its input.
Arguments:
- inputs: The input layer to the bi-directional LSTM.
- hidden_dim: The dimension of the hidden state of the LSTM.
- depth: Depth of the stacked bi-directional LSTM.
- prefix: A string which will be appended to name of each layer
created in this function. Each layer in a network should
has a unique name. The prefix makes this fucntion can be
called multiple times.
"""
lstm_final_outs = [] lstm_final_outs = []
for dirt in ["fwd", "bwd"]: for dirt in ["fwd", "bwd"]:
for i in range(depth): for i in range(depth):
...@@ -122,16 +181,3 @@ def stacked_bi_lstm_by_nested_seq(input_layer, depth, hidden_dim, prefix=""): ...@@ -122,16 +181,3 @@ def stacked_bi_lstm_by_nested_seq(input_layer, depth, hidden_dim, prefix=""):
reverse=(dirt == "bwd")) reverse=(dirt == "bwd"))
lstm_final_outs.append(lstm_out) lstm_final_outs.append(lstm_out)
return paddle.layer.concat(input=lstm_final_outs) return paddle.layer.concat(input=lstm_final_outs)
if __name__ == "__main__":
vocab_size = 1024
emb_dim = 128
embedding = paddle.layer.embedding(
input=paddle.layer.data(
name="word",
type=paddle.data_type.integer_value_sub_sequence(vocab_size)),
size=emb_dim)
print(parse_network(
stacked_bi_lstm_by_nested_seq(
input_layer=embedding, depth=3, hidden_dim=128, prefix="__lstm")))
globally_normalized_reader/beam_decoding.py
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
import pdb
import numpy as np import numpy as np
__all__ = ["BeamDecoding"] __all__ = ["BeamDecoding"]
class BeamDecoding(object): class BeamDecoding(object):
"""
Decode outputs of the PaddlePaddle layers into readable answers.
"""
def __init__(self, documents, sentence_scores, selected_sentences, def __init__(self, documents, sentence_scores, selected_sentences,
start_scores, selected_starts, end_scores, selected_ends): start_scores, selected_starts, end_scores, selected_ends):
""" The constructor.
Arguments:
- documents: The one-hot input of the document words.
- sentence_scores: The score for each sentece in a document.
- selected_sentences: The top k seleceted sentence. This is the
output of the paddle.layer.kmax_seq_score
layer in the model.
- start_scores: The score for each words in the selected
sentence indicating whether the it is start
of the answer.
- selected_starts: The top k selected start spans. This is the
output of the paddle.layer.kmax_seq_score
layer in the model.
- end_scores: The score for each words in the sub-sequence
which is from the selecetd starts till end of
the selected sentence.
- selected_ends: The top k selected end spans. This is the
output of the paddle.layer.kmax_seq_score
layer in the model.
"""
self.documents = documents self.documents = documents
self.sentence_scores = sentence_scores self.sentence_scores = sentence_scores
...@@ -19,13 +46,14 @@ class BeamDecoding(object): ...@@ -19,13 +46,14 @@ class BeamDecoding(object):
self.end_scores = end_scores self.end_scores = end_scores
self.selected_ends = selected_ends self.selected_ends = selected_ends
"""
# sequence start position information for the three step search sequence start position information for the three step search
# beam1 is to search the sequence index beam1 is to search the sequence index
"""
self.beam1_seq_start_positions = [] self.beam1_seq_start_positions = []
# beam2 is to search the start answer span """beam2 is to search the start answer span"""
self.beam2_seq_start_positions = [] self.beam2_seq_start_positions = []
# beam3 is to search the end answer span """beam3 is to search the end answer span """
self.beam3_seq_start_positions = [] self.beam3_seq_start_positions = []
self.ans_per_sample_in_a_batch = [0] self.ans_per_sample_in_a_batch = [0]
...@@ -34,6 +62,11 @@ class BeamDecoding(object): ...@@ -34,6 +62,11 @@ class BeamDecoding(object):
self.final_ans = [[] for i in range(len(documents))] self.final_ans = [[] for i in range(len(documents))]
def _build_beam1_seq_info(self): def _build_beam1_seq_info(self):
"""
The internal function to calculate the offset of each test sequence
in a batch for the first beam in searching the answer sentence.
"""
self.beam1_seq_start_positions.append([0]) self.beam1_seq_start_positions.append([0])
for idx, one_doc in enumerate(self.documents): for idx, one_doc in enumerate(self.documents):
for sentence in one_doc: for sentence in one_doc:
...@@ -45,15 +78,20 @@ class BeamDecoding(object): ...@@ -45,15 +78,20 @@ class BeamDecoding(object):
[self.beam1_seq_start_positions[-1][-1]]) [self.beam1_seq_start_positions[-1][-1]])
def _build_beam2_seq_info(self): def _build_beam2_seq_info(self):
"""
The internal function to calculate the offset of each test sequence
in a batch for the second beam in searching the start spans.
"""
seq_num, beam_size = self.selected_sentences.shape seq_num, beam_size = self.selected_sentences.shape
self.beam2_seq_start_positions.append([0]) self.beam2_seq_start_positions.append([0])
for i in range(seq_num): for i in range(seq_num):
for j in range(beam_size): for j in range(beam_size):
selected_id = int(self.selected_sentences[i][j]) selected_id = int(self.selected_sentences[i][j])
if selected_id == -1: break if selected_id == -1: break
seq_len = self.beam1_seq_start_positions[i][ seq_len = self.beam1_seq_start_positions[
selected_id + 1] - self.beam1_seq_start_positions[i][ i][selected_id +
selected_id] 1] - self.beam1_seq_start_positions[i][selected_id]
self.beam2_seq_start_positions[-1].append( self.beam2_seq_start_positions[-1].append(
self.beam2_seq_start_positions[-1][-1] + seq_len) self.beam2_seq_start_positions[-1][-1] + seq_len)
...@@ -62,6 +100,11 @@ class BeamDecoding(object): ...@@ -62,6 +100,11 @@ class BeamDecoding(object):
[self.beam2_seq_start_positions[-1][-1]]) [self.beam2_seq_start_positions[-1][-1]])
def _build_beam3_seq_info(self): def _build_beam3_seq_info(self):
"""
The internal function to calculate the offset of each test sequence
in a batch for the third beam in searching the end spans.
"""
seq_num_in_a_batch = len(self.documents) seq_num_in_a_batch = len(self.documents)
seq_id = 0 seq_id = 0
...@@ -71,9 +114,9 @@ class BeamDecoding(object): ...@@ -71,9 +114,9 @@ class BeamDecoding(object):
self.beam3_seq_start_positions.append([0]) self.beam3_seq_start_positions.append([0])
sub_seq_num, beam_size = self.selected_starts.shape sub_seq_num, beam_size = self.selected_starts.shape
for i in range(sub_seq_num): for i in range(sub_seq_num):
seq_len = self.beam2_seq_start_positions[seq_id][ seq_len = self.beam2_seq_start_positions[
sub_seq_id + 1] - self.beam2_seq_start_positions[seq_id][ seq_id][sub_seq_id +
sub_seq_id] 1] - self.beam2_seq_start_positions[seq_id][sub_seq_id]
for j in range(beam_size): for j in range(beam_size):
start_id = int(self.selected_starts[i][j]) start_id = int(self.selected_starts[i][j])
if start_id == -1: break if start_id == -1: break
...@@ -88,17 +131,27 @@ class BeamDecoding(object): ...@@ -88,17 +131,27 @@ class BeamDecoding(object):
[self.beam3_seq_start_positions[-1][-1]]) [self.beam3_seq_start_positions[-1][-1]])
sub_seq_id = 0 sub_seq_id = 0
seq_id += 1 seq_id += 1
sub_seq_count = len(self.beam2_seq_start_positions[ sub_seq_count = len(
seq_id]) - 1 self.beam2_seq_start_positions[seq_id]) - 1
assert ( assert (
self.beam3_seq_start_positions[-1][-1] == self.end_scores.shape[0]) self.beam3_seq_start_positions[-1][-1] == self.end_scores.shape[0])
def _build_seq_info_for_each_beam(self): def _build_seq_info_for_each_beam(self):
"""
The internal function to calculate the offset of each test sequence
in a batch for beams expanded at all the three search steps.
"""
self._build_beam1_seq_info() self._build_beam1_seq_info()
self._build_beam2_seq_info() self._build_beam2_seq_info()
self._build_beam3_seq_info() self._build_beam3_seq_info()
def _cal_ans_per_sample_in_a_batch(self): def _cal_ans_per_sample_in_a_batch(self):
"""
The internal function to calculate there are how many candidate answers
for each of the test sequemce in a batch.
"""
start_row = 0 start_row = 0
for seq in self.beam3_seq_start_positions: for seq in self.beam3_seq_start_positions:
end_row = start_row + len(seq) - 1 end_row = start_row + len(seq) - 1
...@@ -109,6 +162,13 @@ class BeamDecoding(object): ...@@ -109,6 +162,13 @@ class BeamDecoding(object):
start_row = end_row start_row = end_row
def _get_valid_seleceted_ids(slef, mat): def _get_valid_seleceted_ids(slef, mat):
"""
The internal function to post-process the output matrix of
paddle.layer.kmax_seq_score layer. This function takes off the special
dilimeter -1 away and flattens the original two-dimensional output
matrix into a python list.
"""
flattened = [] flattened = []
height, width = mat.shape height, width = mat.shape
for i in range(height): for i in range(height):
...@@ -118,6 +178,11 @@ class BeamDecoding(object): ...@@ -118,6 +178,11 @@ class BeamDecoding(object):
return flattened return flattened
def decoding(self): def decoding(self):
"""
The internal function to decode forward results of the GNR network into
readable answers.
"""
self._build_seq_info_for_each_beam() self._build_seq_info_for_each_beam()
self._cal_ans_per_sample_in_a_batch() self._cal_ans_per_sample_in_a_batch()
...@@ -134,11 +199,16 @@ class BeamDecoding(object): ...@@ -134,11 +199,16 @@ class BeamDecoding(object):
if end_pos == -1: break if end_pos == -1: break
self.all_searched_ans.append({ self.all_searched_ans.append({
"score": self.end_scores[seq_offset_in_batch + end_pos], "score":
"sentence_pos": -1, self.end_scores[seq_offset_in_batch + end_pos],
"start_span_pos": -1, "sentence_pos":
"end_span_pos": end_pos, -1,
"parent_ids_in_prev_beam": i "start_span_pos":
-1,
"end_span_pos":
end_pos,
"parent_ids_in_prev_beam":
i
}) })
sub_seq_id += 1 sub_seq_id += 1
...@@ -196,7 +266,8 @@ class BeamDecoding(object): ...@@ -196,7 +266,8 @@ class BeamDecoding(object):
key=lambda x: x["score"], key=lambda x: x["score"],
reverse=True): reverse=True):
self.final_ans[i].append({ self.final_ans[i].append({
"score": ans["score"], "score":
ans["score"],
"label": [ "label": [
ans["sentence_pos"], ans["start_span_pos"], ans["sentence_pos"], ans["start_span_pos"],
ans["end_span_pos"] ans["end_span_pos"]
......
globally_normalized_reader/config.py
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
__all__ = ["ModelConfig"] __all__ = ["ModelConfig", "TrainerConfig"]
class ModelConfig(object): class ModelConfig(object):
...@@ -24,14 +24,15 @@ class ModelConfig(object): ...@@ -24,14 +24,15 @@ class ModelConfig(object):
class TrainerConfig(object): class TrainerConfig(object):
learning_rate = 1e-3 learning_rate = 1e-3
l2_decay_rate = 5e-4
gradient_clipping_threshold = 20
data_dir = "data/featurized" data_dir = "data/featurized"
save_dir = "models" save_dir = "models"
use_gpu = True use_gpu = False
trainer_count = 4 trainer_count = 1
train_batch_size = trainer_count * 10 train_batch_size = trainer_count * 8
test_batch_size = 4
epochs = 20 epochs = 20
......
globally_normalized_reader/data/download.sh 0 → 100755
浏览文件 @ 44a72d3d
#!/bin/bash
wget --no-check-certificate https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json
wget --no-check-certificate https://rajpurkar.github.io/SQuAD-explorer/dataset/dev-v1.1.json
globally_normalized_reader/index.html 0 → 100644
浏览文件 @ 44a72d3d
<html>
<head>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js", "TeX/AMSsymbols.js", "TeX/AMSmath.js"],
jax: ["input/TeX", "output/HTML-CSS"],
tex2jax: {
inlineMath: [ ['$','$'] ],
displayMath: [ ['$$','$$'] ],
processEscapes: true
},
"HTML-CSS": { availableFonts: ["TeX"] }
});
</script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js" async></script>
<script type="text/javascript" src="../.tools/theme/marked.js">
</script>
<link href="http://cdn.bootcss.com/highlight.js/9.9.0/styles/darcula.min.css" rel="stylesheet">
<script src="http://cdn.bootcss.com/highlight.js/9.9.0/highlight.min.js"></script>
<link href="http://cdn.bootcss.com/bootstrap/4.0.0-alpha.6/css/bootstrap.min.css" rel="stylesheet">
<link href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" rel="stylesheet">
<link href="../.tools/theme/github-markdown.css" rel='stylesheet'>
</head>
<style type="text/css" >
.markdown-body {
box-sizing: border-box;
min-width: 200px;
max-width: 980px;
margin: 0 auto;
padding: 45px;
}
</style>
<body>
<div id="context" class="container-fluid markdown-body">
</div>
<!-- This block will be replaced by each markdown file content. Please do not change lines below.-->
<div id="markdown" style='display:none'>
# Globally Normalized Reader
This model implements the work in the following paper:
Jonathan Raiman and John Miller. Globally Normalized Reader. Empirical Methods in Natural Language Processing (EMNLP), 2017.
If you use the dataset/code in your research, please cite the above paper:
```text
@inproceedings{raiman2015gnr,
author={Raiman, Jonathan and Miller, John},
booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
title={Globally Normalized Reader},
year={2017},
}
```
You can also visit https://github.com/baidu-research/GloballyNormalizedReader to get more information.
# Installation
1. Please use [docker image](http://doc.paddlepaddle.org/develop/doc/getstarted/build_and_install/docker_install_en.html) to install the latest PaddlePaddle, by running:
```bash
docker pull paddledev/paddle
```
2. Download all necessary data by running:
```bash
cd data && ./download.sh
```
3. Featurize the data by running:
```
python featurize.py --datadir data --outdir featurized
```
# Training a Model
- Configurate the model by modifying `config.py` if needed, and then run:
```bash
python train.py 2>&1 | tee train.log
```
# Inferring by a Trained Model
- Infer by a trained model by running:
```bash
python infer.py \
--model_path models/pass_00000.tar.gz \
--data_dir data/featurized/ \
--batch_size 2 \
--use_gpu 0 \
--trainer_count 1 \
2>&1 | tee infer.log
```
</div>
<!-- You can change the lines below now. -->
<script type="text/javascript">
marked.setOptions({
renderer: new marked.Renderer(),
gfm: true,
breaks: false,
smartypants: true,
highlight: function(code, lang) {
code = code.replace(/&amp;/g, "&")
code = code.replace(/&gt;/g, ">")
code = code.replace(/&lt;/g, "<")
code = code.replace(/&nbsp;/g, " ")
return hljs.highlightAuto(code, [lang]).value;
}
});
document.getElementById("context").innerHTML = marked(
document.getElementById("markdown").innerHTML)
</script>
</body>
globally_normalized_reader/infer.py 100755 → 100644
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
import os import os
import sys import sys
import argparse
import gzip import gzip
import logging import logging
import numpy as np import numpy as np
...@@ -12,14 +14,62 @@ import reader ...@@ -12,14 +14,62 @@ import reader
from model import GNR from model import GNR
from train import choose_samples from train import choose_samples
from config import ModelConfig, TrainerConfig from config import ModelConfig
from beam_decoding import BeamDecoding from beam_decoding import BeamDecoding
logger = logging.getLogger("paddle") logger = logging.getLogger("paddle")
logger.setLevel(logging.INFO) logger.setLevel(logging.INFO)
def parse_cmd():
"""
Build the command line arguments parser for inferring task.
"""
parser = argparse.ArgumentParser(
description="Globally Normalized Reader in PaddlePaddle.")
parser.add_argument(
"--model_path",
required=True,
type=str,
help="Path of the trained model to evaluate.",
default="")
parser.add_argument(
"--data_dir",
type=str,
required=True,
help="Path of the training and testing data.",
default="")
parser.add_argument(
"--batch_size",
type=int,
required=False,
help="The batch size for inferring.",
default=1)
parser.add_argument(
"--use_gpu",
type=int,
required=False,
help="Whether to run the inferring on GPU.",
default=0)
parser.add_argument(
"--trainer_count",
type=int,
required=False,
help=("The thread number used in inferring. When set "
"use_gpu=True, the trainer_count cannot excess "
"the gpu device number in your computer."),
default=1)
return parser.parse_args()
def load_reverse_dict(dict_file): def load_reverse_dict(dict_file):
""" Build the dict which is used to map the word index to word string.
The keys are word index and the values are word strings.
Arguments:
- dict_file: The path of a word dictionary.
"""
word_dict = {} word_dict = {}
with open(dict_file, "r") as fin: with open(dict_file, "r") as fin:
for idx, line in enumerate(fin): for idx, line in enumerate(fin):
...@@ -28,6 +78,34 @@ def load_reverse_dict(dict_file): ...@@ -28,6 +78,34 @@ def load_reverse_dict(dict_file):
def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1): def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1):
""" Print the readable predicted answers.
Format of the output:
query:\tthe input query.
documents:\n
0\tthe first sentence in the document.
1\tthe second sentence in the document.
...
gold:\t[i j k] the answer words.
(i: the sentence index;
j: the start span index;
k: the end span index)
top answers:
score0\t[i j k] the answer with the highest score.
score1\t[i j k] the answer with the second highest score.
(i, j, k has a same meaning as in gold.)
...
By default, top 10 answers will be printed.
Arguments:
- test_batch: A test batch returned by reader.
- predicted_ans: The beam decoding results.
- ids_2_word: The dict whose key is word index and the values are
word strings.
- print_top_k: Indicating how many answers will be printed.
"""
for i, sample in enumerate(test_batch): for i, sample in enumerate(test_batch):
query_words = [ids_2_word[ids] for ids in sample[0]] query_words = [ids_2_word[ids] for ids in sample[0]]
print("query:\t%s" % (" ".join(query_words))) print("query:\t%s" % (" ".join(query_words)))
...@@ -42,7 +120,7 @@ def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1): ...@@ -42,7 +120,7 @@ def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1):
sample[3], sample[4], sample[5], " ".join( sample[3], sample[4], sample[5], " ".join(
[ids_2_word[ids] for ids in sample[1][sample[3]][start:end]]))) [ids_2_word[ids] for ids in sample[1][sample[3]][start:end]])))
print("predicted:") print("top answers:")
for k in range(print_top_k): for k in range(print_top_k):
label = predicted_ans[i][k]["label"] label = predicted_ans[i][k]["label"]
start = label[1] start = label[1]
...@@ -57,14 +135,42 @@ def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1): ...@@ -57,14 +135,42 @@ def print_result(test_batch, predicted_ans, ids_2_word, print_top_k=1):
def infer_a_batch(inferer, test_batch, ids_2_word, out_layer_count): def infer_a_batch(inferer, test_batch, ids_2_word, out_layer_count):
""" Call the PaddlePaddle's infer interface to infer by batch.
Arguments:
- inferer: The PaddlePaddle Inference object.
- test_batch: A test batch returned by reader.
- ids_2_word: The dict whose key is word index and the values are
word strings.
- out_layer_count: The number of output layers in the inferring process.
"""
outs = inferer.infer(input=test_batch, flatten_result=False, field="value") outs = inferer.infer(input=test_batch, flatten_result=False, field="value")
decoder = BeamDecoding([sample[1] for sample in test_batch], *outs) decoder = BeamDecoding([sample[1] for sample in test_batch], *outs)
print_result(test_batch, decoder.decoding(), ids_2_word, print_top_k=10) print_result(test_batch, decoder.decoding(), ids_2_word, print_top_k=10)
def infer(model_path, data_dir, test_batch_size, config): def infer(model_path,
data_dir,
batch_size,
config,
use_gpu=False,
trainer_count=1):
""" The inferring process.
Arguments:
- model_path: The path of trained model.
- data_dir: The directory path of test data.
- batch_size: The batch_size.
- config: The model configuration.
- use_gpu: Whether to run the inferring on GPU.
- trainer_count: The thread number used in inferring. When set
use_gpu=True, the trainer_count cannot excess
the gpu device number in your computer.
"""
assert os.path.exists(model_path), "The model does not exist." assert os.path.exists(model_path), "The model does not exist."
paddle.init(use_gpu=True, trainer_count=1) paddle.init(use_gpu=use_gpu, trainer_count=trainer_count)
ids_2_word = load_reverse_dict(config.dict_path) ids_2_word = load_reverse_dict(config.dict_path)
...@@ -84,7 +190,7 @@ def infer(model_path, data_dir, test_batch_size, config): ...@@ -84,7 +190,7 @@ def infer(model_path, data_dir, test_batch_size, config):
test_batch = [] test_batch = []
for i, item in enumerate(test_reader()): for i, item in enumerate(test_reader()):
test_batch.append(item) test_batch.append(item)
if len(test_batch) == test_batch_size: if len(test_batch) == batch_size:
infer_a_batch(inferer, test_batch, ids_2_word, len(outputs)) infer_a_batch(inferer, test_batch, ids_2_word, len(outputs))
test_batch = [] test_batch = []
...@@ -93,7 +199,16 @@ def infer(model_path, data_dir, test_batch_size, config): ...@@ -93,7 +199,16 @@ def infer(model_path, data_dir, test_batch_size, config):
test_batch = [] test_batch = []
def main(args):
infer(
model_path=args.model_path,
data_dir=args.data_dir,
batch_size=args.batch_size,
config=ModelConfig,
use_gpu=args.use_gpu,
trainer_count=args.trainer_count)
if __name__ == "__main__": if __name__ == "__main__":
# infer("models/round1/pass_00000.tar.gz", TrainerConfig.data_dir, args = parse_cmd()
infer("models/round2_on_cpu/pass_00000.tar.gz", TrainerConfig.data_dir, main(args)
TrainerConfig.test_batch_size, ModelConfig)
globally_normalized_reader/model.py 100755 → 100644
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
import paddle.v2 as paddle import paddle.v2 as paddle
from paddle.v2.layer import parse_network from paddle.v2.layer import parse_network
import basic_modules import basic_modules
...@@ -9,52 +10,115 @@ __all__ = ["GNR"] ...@@ -9,52 +10,115 @@ __all__ = ["GNR"]
def build_pretrained_embedding(name, data_type, emb_dim, emb_drop=0.): def build_pretrained_embedding(name, data_type, emb_dim, emb_drop=0.):
"""create word a embedding layer which loads pre-trained embeddings.
Arguments:
- name: The name of the data layer which accepts one-hot input.
- data_type: PaddlePaddle's data type for data layer.
- emb_dim: The path to the data files.
"""
return paddle.layer.embedding( return paddle.layer.embedding(
input=paddle.layer.data( input=paddle.layer.data(name=name, type=data_type),
name=name, type=data_type),
size=emb_dim, size=emb_dim,
param_attr=paddle.attr.Param( param_attr=paddle.attr.Param(name="GloveVectors", is_static=True),
name="GloveVectors", is_static=True),
layer_attr=paddle.attr.ExtraLayerAttribute(drop_rate=emb_drop), ) layer_attr=paddle.attr.ExtraLayerAttribute(drop_rate=emb_drop), )
def encode_question(input_embedding, config, prefix): def encode_question(input_embedding,
lstm_hidden_dim,
depth,
passage_indep_embedding_dim,
prefix=""):
"""build question encoding by using bidirectional LSTM.
Each question word is encoded by runing a stack of bidirectional LSTM over
word embedding in question, producing hidden states. The hidden states are
used to compute a passage-independent question embedding.
The final question encoding is constructed by concatenating the final
hidden states of the forward and backward LSTMs and the passage-independent
embedding.
Arguments:
- input_embedding: The question word embeddings.
- lstm_hidden_dim: The dimension of bi-directional LSTM.
- depth: The depth of stacked bi-directional LSTM.
- passage_indep_embedding_dim: The dimension of passage-independent
embedding.
- prefix: A string which will be appended to name of each layer
created in this function. Each layer in a network should
has a unique name. The prefix makes this fucntion can be
called multiple times.
"""
# stacked bi-directional LSTM to process question embeddings.
lstm_final, lstm_outs = basic_modules.stacked_bidirectional_lstm( lstm_final, lstm_outs = basic_modules.stacked_bidirectional_lstm(
inputs=input_embedding, input_embedding, lstm_hidden_dim, depth, 0., prefix)
size=config.lstm_hidden_dim,
depth=config.lstm_depth, # compute passage-independent embeddings.
drop_rate=config.lstm_hidden_droprate, candidates = paddle.layer.fc(
prefix=prefix) input=lstm_outs,
# passage-independent embeddings
candidates = paddle.layer.fc(input=lstm_outs,
bias_attr=False, bias_attr=False,
size=config.passage_indep_embedding_dim, size=passage_indep_embedding_dim,
act=paddle.activation.Linear()) act=paddle.activation.Linear())
weights = paddle.layer.fc(input=lstm_outs, weights = paddle.layer.fc(
input=lstm_outs,
size=1, size=1,
bias_attr=False, bias_attr=False,
act=paddle.activation.SequenceSoftmax()) act=paddle.activation.SequenceSoftmax())
weighted_candidates = paddle.layer.scaling(input=candidates, weight=weights) weighted_candidates = paddle.layer.scaling(input=candidates, weight=weights)
passage_indep_embedding = paddle.layer.pooling( passage_indep_embedding = paddle.layer.pooling(
input=weighted_candidates, pooling_type=paddle.pooling.Sum()) input=weighted_candidates, pooling_type=paddle.pooling.Sum())
return paddle.layer.concat( return paddle.layer.concat(
input=[lstm_final, passage_indep_embedding]), lstm_outs input=[lstm_final, passage_indep_embedding]), lstm_outs
def question_aligned_passage_embedding(question_lstm_outs, document_embeddings, def question_aligned_passage_embedding(question_lstm_outs, document_embeddings,
config): passage_aligned_embedding_dim):
def outer_sentence_step(document_embeddings, question_lstm_outs, config): """create question aligned passage embedding.
'''
in this recurrent_group, document_embeddings has scattered into sequence, Arguments:
''' - question_lstm_outs: The dimension of output of LSTM that process
question word embedding.
- document_embeddings: The document embeddings.
- passage_aligned_embedding_dim: The dimension of passage aligned
embedding.
"""
def outer_sentence_step(document_embeddings, question_lstm_outs,
passage_aligned_embedding_dim):
"""step function for PaddlePaddle's recurrent_group.
In this function, the original input document_embeddings are scattered
from nested sequence into sequence by recurrent_group in PaddlePaddle.
The step function iterates over each sentence in the document.
Arguments:
- document_embeddings: The word embeddings of the document.
- question_lstm_outs: The dimension of output of LSTM that
process question word embedding.
- passage_aligned_embedding_dim: The dimension of passage aligned
embedding.
"""
def inner_word_step(word_embedding, question_lstm_outs, def inner_word_step(word_embedding, question_lstm_outs,
question_outs_proj, config): question_outs_proj, passage_aligned_embedding_dim):
''' """
in this recurrent_group, sentence embedding has scattered into word In this recurrent_group, sentence embedding has been scattered into
embeddings. word embeddings. The step function iterates over each word in one
''' sentence in the document.
Arguments:
- word_embedding: The word embeddings of documents.
- question_lstm_outs: The dimension of output of LSTM that
process question word embedding.
- question_outs_proj: The projection of question_lstm_outs
into a new hidden space.
- passage_aligned_embedding_dim: The dimension of passage
aligned embedding.
"""
doc_word_expand = paddle.layer.expand( doc_word_expand = paddle.layer.expand(
input=word_embedding, input=word_embedding,
expand_as=question_lstm_outs, expand_as=question_lstm_outs,
...@@ -62,10 +126,6 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings, ...@@ -62,10 +126,6 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings,
weights = paddle.layer.fc( weights = paddle.layer.fc(
input=[question_lstm_outs, doc_word_expand], input=[question_lstm_outs, doc_word_expand],
param_attr=[
paddle.attr.Param(initial_std=1e-3),
paddle.attr.Param(initial_std=1e-3)
],
size=1, size=1,
bias_attr=False, bias_attr=False,
act=paddle.activation.SequenceSoftmax()) act=paddle.activation.SequenceSoftmax())
...@@ -77,13 +137,13 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings, ...@@ -77,13 +137,13 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings,
question_outs_proj = paddle.layer.fc( question_outs_proj = paddle.layer.fc(
input=question_lstm_outs, input=question_lstm_outs,
bias_attr=False, bias_attr=False,
size=config.passage_aligned_embedding_dim) size=passage_aligned_embedding_dim)
return paddle.layer.recurrent_group( return paddle.layer.recurrent_group(
input=[ input=[
paddle.layer.SubsequenceInput(document_embeddings), paddle.layer.SubsequenceInput(document_embeddings),
paddle.layer.StaticInput(question_lstm_outs), paddle.layer.StaticInput(question_lstm_outs),
paddle.layer.StaticInput(question_outs_proj), paddle.layer.StaticInput(question_outs_proj),
config, passage_aligned_embedding_dim,
], ],
step=inner_word_step, step=inner_word_step,
name="iter_over_word") name="iter_over_word")
...@@ -91,20 +151,39 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings, ...@@ -91,20 +151,39 @@ def question_aligned_passage_embedding(question_lstm_outs, document_embeddings,
return paddle.layer.recurrent_group( return paddle.layer.recurrent_group(
input=[ input=[
paddle.layer.SubsequenceInput(document_embeddings), paddle.layer.SubsequenceInput(document_embeddings),
paddle.layer.StaticInput(question_lstm_outs), config paddle.layer.StaticInput(question_lstm_outs),
passage_aligned_embedding_dim
], ],
step=outer_sentence_step, step=outer_sentence_step,
name="iter_over_sen") name="iter_over_sen")
def encode_documents(input_embedding, same_as_question, question_vector, def encode_documents(input_embedding, same_as_question, question_vector,
question_lstm_outs, config, prefix): question_lstm_outs, passage_indep_embedding_dim, prefix):
"""Build the final question-aware document embeddings.
Each word in the document is represented as concatenation of its word
vector, the question vector, boolean features indicating if a word appers
in the question or is repeated, and a question aligned embedding.
Arguments:
- input_embedding: The word embeddings of the document.
- same_as_question: The boolean features indicating if a word appears
in the question or is repeated.
- question_lstm_outs: The final question encoding.
- passage_indep_embedding_dim: The dimension of passage independent
embedding.
- prefix: The prefix which will be appended to name of each layer in
This function.
"""
question_expanded = paddle.layer.expand( question_expanded = paddle.layer.expand(
input=question_vector, input=question_vector,
expand_as=input_embedding, expand_as=input_embedding,
expand_level=paddle.layer.ExpandLevel.FROM_NO_SEQUENCE) expand_level=paddle.layer.ExpandLevel.FROM_NO_SEQUENCE)
question_aligned_embedding = question_aligned_passage_embedding( question_aligned_embedding = question_aligned_passage_embedding(
question_lstm_outs, input_embedding, config) question_lstm_outs, input_embedding, passage_indep_embedding_dim)
return paddle.layer.concat(input=[ return paddle.layer.concat(input=[
input_embedding, question_expanded, same_as_question, input_embedding, question_expanded, same_as_question,
question_aligned_embedding question_aligned_embedding
...@@ -113,15 +192,48 @@ def encode_documents(input_embedding, same_as_question, question_vector, ...@@ -113,15 +192,48 @@ def encode_documents(input_embedding, same_as_question, question_vector,
def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config, def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config,
is_infer): is_infer):
"""Search the answer from the document.
The search process for this layer begins with searching a target sequence
from a nested sequence by using paddle.lauer.kmax_seq_score and
paddle.layer.sub_nested_seq_layer. In the first search step, top beam size
sequences with highest scores, indices of these top k sequences in the
original nested sequence, and the ground truth (also called gold)
altogether (a triple) make up of the first beam.
Then, start and end positions are searched. In these searches, top k
positions with highest scores are selected, and then sequence, starting
from the selected starts till ends of the sequences are taken to search
next by using paddle.layer.seq_slice.
Finally, the layer paddle.layer.cross_entropy_over_beam takes all the beam
expansions which contain several candidate targets found along the
three-step search. cross_entropy_over_beam calculates cross entropy over
the expanded beams which all the candidates in the beam as the normalized
factor.
Note that, if gold falls off the beam at search step t, then the cost is
calculated over the beam at step t.
Arguments:
- doc_lstm_outs: The output of LSTM that process each document words.
- sentence_idx: Ground-truth indicating sentence index of the answer
in the document.
- start_idx: Ground-truth indicating start span index of the answer
in the sentence.
- end_idx: Ground-truth indicating end span index of the answer
in the sentence.
- is_infer: The boolean parameter indicating inferring or training.
"""
last_state_of_sentence = paddle.layer.last_seq( last_state_of_sentence = paddle.layer.last_seq(
input=doc_lstm_outs, agg_level=paddle.layer.AggregateLevel.TO_SEQUENCE) input=doc_lstm_outs, agg_level=paddle.layer.AggregateLevel.TO_SEQUENCE)
sentence_scores = paddle.layer.fc( sentence_scores = paddle.layer.fc(
input=last_state_of_sentence, input=last_state_of_sentence,
size=1, size=1,
bias_attr=False, bias_attr=False,
param_attr=paddle.attr.Param(initial_std=1e-3),
act=paddle.activation.Linear()) act=paddle.activation.Linear())
topk_sentence_ids = paddle.layer.kmax_sequence_score( topk_sentence_ids = paddle.layer.kmax_seq_score(
input=sentence_scores, beam_size=config.beam_size) input=sentence_scores, beam_size=config.beam_size)
topk_sen = paddle.layer.sub_nested_seq( topk_sen = paddle.layer.sub_nested_seq(
input=doc_lstm_outs, selected_indices=topk_sentence_ids) input=doc_lstm_outs, selected_indices=topk_sentence_ids)
...@@ -131,29 +243,24 @@ def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config, ...@@ -131,29 +243,24 @@ def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config,
input=topk_sen, input=topk_sen,
size=1, size=1,
layer_attr=paddle.attr.ExtraLayerAttribute( layer_attr=paddle.attr.ExtraLayerAttribute(
error_clipping_threshold=10.0), error_clipping_threshold=5.0),
bias_attr=False, bias_attr=False,
param_attr=paddle.attr.Param(initial_std=1e-3),
act=paddle.activation.Linear()) act=paddle.activation.Linear())
topk_start_pos_ids = paddle.layer.kmax_sequence_score( topk_start_pos_ids = paddle.layer.kmax_seq_score(
input=start_pos_scores, beam_size=config.beam_size) input=start_pos_scores, beam_size=config.beam_size)
topk_start_spans = paddle.layer.seq_slice( topk_start_spans = paddle.layer.seq_slice(
input=topk_sen, starts=topk_start_pos_ids, ends=None) input=topk_sen, starts=topk_start_pos_ids, ends=None)
# expand beam to search end positions on selected start spans # expand beam to search end positions on selected start spans
_, end_span_embedding = basic_modules.stacked_bidirectional_lstm( _, end_span_embedding = basic_modules.stacked_bidirectional_lstm(
inputs=topk_start_spans, topk_start_spans, config.lstm_hidden_dim, config.lstm_depth,
size=config.lstm_hidden_dim, config.lstm_hidden_droprate, "__end_span_embeddings__")
depth=config.lstm_depth,
drop_rate=config.lstm_hidden_droprate,
prefix="__end_span_embeddings__")
end_pos_scores = paddle.layer.fc( end_pos_scores = paddle.layer.fc(
input=end_span_embedding, input=end_span_embedding,
size=1, size=1,
bias_attr=False, bias_attr=False,
param_attr=paddle.attr.Param(initial_std=1e-3),
act=paddle.activation.Linear()) act=paddle.activation.Linear())
topk_end_pos_ids = paddle.layer.kmax_sequence_score( topk_end_pos_ids = paddle.layer.kmax_seq_score(
input=end_pos_scores, beam_size=config.beam_size) input=end_pos_scores, beam_size=config.beam_size)
if is_infer: if is_infer:
...@@ -172,15 +279,23 @@ def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config, ...@@ -172,15 +279,23 @@ def search_answer(doc_lstm_outs, sentence_idx, start_idx, end_idx, config,
def GNR(config, is_infer=False): def GNR(config, is_infer=False):
# encoding question words """Build the globally normalized reader model.
Arguments:
- config: The model configuration.
- is_infer: The boolean parameter indicating inferring or training.
"""
# encode question words
question_embeddings = build_pretrained_embedding( question_embeddings = build_pretrained_embedding(
"question", "question",
paddle.data_type.integer_value_sequence(config.vocab_size), paddle.data_type.integer_value_sequence(config.vocab_size),
config.embedding_dim, config.embedding_droprate) config.embedding_dim, config.embedding_droprate)
question_vector, question_lstm_outs = encode_question( question_vector, question_lstm_outs = encode_question(
input_embedding=question_embeddings, config=config, prefix="__ques") question_embeddings, config.lstm_hidden_dim, config.lstm_depth,
config.passage_indep_embedding_dim, "__ques")
# encoding document words # encode document words
document_embeddings = build_pretrained_embedding( document_embeddings = build_pretrained_embedding(
"documents", "documents",
paddle.data_type.integer_value_sub_sequence(config.vocab_size), paddle.data_type.integer_value_sub_sequence(config.vocab_size),
...@@ -190,20 +305,14 @@ def GNR(config, is_infer=False): ...@@ -190,20 +305,14 @@ def GNR(config, is_infer=False):
type=paddle.data_type.dense_vector_sub_sequence(1)) type=paddle.data_type.dense_vector_sub_sequence(1))
document_words_ecoding = encode_documents( document_words_ecoding = encode_documents(
input_embedding=document_embeddings, document_embeddings, same_as_question, question_vector,
question_vector=question_vector, question_lstm_outs, config.passage_indep_embedding_dim, "__doc")
question_lstm_outs=question_lstm_outs,
same_as_question=same_as_question, doc_lstm_outs = basic_modules.stacked_bidirectional_lstm_by_nested_seq(
config=config, document_words_ecoding, config.lstm_depth, config.lstm_hidden_dim,
prefix="__doc") "__doc_lstm")
doc_lstm_outs = basic_modules.stacked_bi_lstm_by_nested_seq( # search the answer.
input_layer=document_words_ecoding,
hidden_dim=config.lstm_hidden_dim,
depth=config.lstm_depth,
prefix="__doc_lstm")
# define labels
sentence_idx = paddle.layer.data( sentence_idx = paddle.layer.data(
name="sen_idx", type=paddle.data_type.integer_value(1)) name="sen_idx", type=paddle.data_type.integer_value(1))
start_idx = paddle.layer.data( start_idx = paddle.layer.data(
......
globally_normalized_reader/reader.py 100755 → 100644
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
import os import os
import random import random
import json import json
...@@ -10,8 +11,16 @@ logger.setLevel(logging.INFO) ...@@ -10,8 +11,16 @@ logger.setLevel(logging.INFO)
def data_reader(data_list, is_train=True): def data_reader(data_list, is_train=True):
""" Data reader.
Arguments:
- data_list: A python list which contains path of training samples.
- is_train: A boolean parameter indicating this function is called
in training or in inferring.
"""
def reader(): def reader():
# every pass shuffle the data list again """shuffle the data list again at the begining of every pass"""
if is_train: if is_train:
random.shuffle(data_list) random.shuffle(data_list)
...@@ -33,12 +42,3 @@ def data_reader(data_list, is_train=True): ...@@ -33,12 +42,3 @@ def data_reader(data_list, is_train=True):
data['ans_end'] - data['ans_start']) data['ans_end'] - data['ans_start'])
return reader return reader
if __name__ == "__main__":
from train import choose_samples
train_list, dev_list = choose_samples("data/featurized")
for i, item in enumerate(data_reader(train_list)()):
print(item)
if i > 5: break
globally_normalized_reader/train.py 100755 → 100644
浏览文件 @ 44a72d3d
#!/usr/bin/env python #!/usr/bin/env python
#coding=utf-8 #coding=utf-8
from __future__ import print_function from __future__ import print_function
import os import os
...@@ -21,44 +22,59 @@ logger.setLevel(logging.INFO) ...@@ -21,44 +22,59 @@ logger.setLevel(logging.INFO)
def load_initial_model(model_path, parameters): def load_initial_model(model_path, parameters):
""" """ Initalize parameters in the network from a trained model.
initalize parameters in the network from a trained model.
This is useful in resuming the training from previously saved models.
Arguments:
- model_path: The path of a trained model.
- parameters: The parameters in a network which will be initialized
from the specified model.
""" """
with gzip.open(model_path, "rb") as f: with gzip.open(model_path, "rb") as f:
parameters.init_from_tar(f) parameters.init_from_tar(f)
def load_pretrained_parameters(path, height, width): def load_pretrained_parameters(path):
""" Load one pre-trained parameter.
Arguments:
- path: The path of the pre-trained parameter.
"""
return np.load(path) return np.load(path)
def save_model(save_path, parameters): def save_model(save_path, parameters):
""" Save the trained parameters.
Arguments:
- save_path: The path to save the trained parameters.
- parameters: The trained model parameters.
"""
with gzip.open(save_path, "w") as f: with gzip.open(save_path, "w") as f:
parameters.to_tar(f) parameters.to_tar(f)
def load_initial_model(model_path, parameters):
with gzip.open(model_path, "rb") as f:
parameters.init_from_tar(f)
def show_parameter_init_info(parameters): def show_parameter_init_info(parameters):
""" Print the information of initialization mean and std of parameters.
Arguments:
- parameters: The parameters created in a model.
"""
for p in parameters: for p in parameters:
logger.info("%s : initial_mean %.4f initial_std %.4f" % logger.info("%s : initial_mean %.4f initial_std %.4f" %
(p, parameters.__param_conf__[p].initial_mean, (p, parameters.__param_conf__[p].initial_mean,
parameters.__param_conf__[p].initial_std)) parameters.__param_conf__[p].initial_std))
def dump_value_matrix(param_name, dims, value): def show_parameter_status(parameters):
np.savetxt( """ Print some statistical information of parameters in a network.
param_name + ".txt",
value.reshape(dims[0], dims[1]),
fmt="%.4f",
delimiter=",")
This is used for debugging the model.
def show_parameter_status(parameters): Arguments:
# for debug print - parameters: The parameters created in a model.
"""
for p in parameters: for p in parameters:
value = parameters.get(p) value = parameters.get(p)
...@@ -75,8 +91,10 @@ def show_parameter_status(parameters): ...@@ -75,8 +91,10 @@ def show_parameter_status(parameters):
def choose_samples(path): def choose_samples(path):
""" """Load filenames for train, dev, and augmented samples.
Load filenames for train, dev, and augmented samples.
Arguments:
- path: The path of training data.
""" """
if not os.path.exists(os.path.join(path, "train")): if not os.path.exists(os.path.join(path, "train")):
print( print(
...@@ -97,8 +115,11 @@ def choose_samples(path): ...@@ -97,8 +115,11 @@ def choose_samples(path):
def build_reader(data_dir, batch_size): def build_reader(data_dir, batch_size):
""" """Build the data reader for this model.
Build the data reader for this model.
Arguments:
- data_dir: The path of training data.
- batch_size: batch size for the training task.
""" """
train_samples, valid_samples = choose_samples(data_dir) train_samples, valid_samples = choose_samples(data_dir)
...@@ -109,15 +130,18 @@ def build_reader(data_dir, batch_size): ...@@ -109,15 +130,18 @@ def build_reader(data_dir, batch_size):
# testing data is not shuffled # testing data is not shuffled
test_reader = paddle.batch( test_reader = paddle.batch(
reader.data_reader( reader.data_reader(valid_samples, is_train=False),
valid_samples, is_train=False),
batch_size=batch_size) batch_size=batch_size)
return train_reader, test_reader return train_reader, test_reader, len(train_samples)
def build_event_handler(config, parameters, trainer, test_reader): def build_event_handler(config, parameters, trainer):
""" """Build the event handler for this model.
Build the event handler for this model.
Arguments:
- config: The training task configuration for this model.
- parameters: The parameters in the network.
- trainer: The trainer object.
""" """
# End batch and end pass event handler # End batch and end pass event handler
...@@ -127,12 +151,8 @@ def build_event_handler(config, parameters, trainer, test_reader): ...@@ -127,12 +151,8 @@ def build_event_handler(config, parameters, trainer, test_reader):
if isinstance(event, paddle.event.EndIteration): if isinstance(event, paddle.event.EndIteration):
if event.batch_id and \ if event.batch_id and \
(not event.batch_id % config.checkpoint_period): (not event.batch_id % config.checkpoint_period):
# save_path = os.path.join(config.save_dir,
# "checkpoint_param.latest.tar.gz")
save_path = os.path.join(config.save_dir, save_path = os.path.join(config.save_dir,
"pass_%05d_%03d.tar.gz" % "checkpoint_param.latest.tar.gz")
(event.pass_id, event.batch_id))
save_model(save_path, parameters) save_model(save_path, parameters)
if event.batch_id and not event.batch_id % config.log_period: if event.batch_id and not event.batch_id % config.log_period:
...@@ -148,14 +168,17 @@ def build_event_handler(config, parameters, trainer, test_reader): ...@@ -148,14 +168,17 @@ def build_event_handler(config, parameters, trainer, test_reader):
"pass_%05d.tar.gz" % event.pass_id) "pass_%05d.tar.gz" % event.pass_id)
save_model(save_path, parameters) save_model(save_path, parameters)
# result = trainer.test(reader=test_reader)
# logger.info("Test with Pass %d, %s" %
# (event.pass_id, result.metrics))
return event_handler return event_handler
def train(model_config, trainer_config): def train(model_config, trainer_config):
"""Training the GNR model.
Arguments:
- modle_config: The model configuration for this model.
- trainer_config: The training task configuration for this model.
"""
if not os.path.exists(trainer_config.save_dir): if not os.path.exists(trainer_config.save_dir):
os.mkdir(trainer_config.save_dir) os.mkdir(trainer_config.save_dir)
...@@ -163,13 +186,24 @@ def train(model_config, trainer_config): ...@@ -163,13 +186,24 @@ def train(model_config, trainer_config):
use_gpu=trainer_config.use_gpu, use_gpu=trainer_config.use_gpu,
trainer_count=trainer_config.trainer_count) trainer_count=trainer_config.trainer_count)
# define the optimizer train_reader, test_reader, train_sample_count = build_reader(
trainer_config.data_dir, trainer_config.train_batch_size)
"""
Define the optimizer. The learning rate will decrease according to
the following formula:
lr = learning_rate * pow(learning_rate_decay_a,
floor(num_samples_processed /
learning_rate_decay_b))
"""
optimizer = paddle.optimizer.Adam( optimizer = paddle.optimizer.Adam(
learning_rate=trainer_config.learning_rate, learning_rate=trainer_config.learning_rate,
gradient_clipping_threshold=50, gradient_clipping_threshold=trainer_config.gradient_clipping_threshold,
regularization=paddle.optimizer.L2Regularization(rate=5e-4), regularization=paddle.optimizer.L2Regularization(
model_average=paddle.optimizer.ModelAverage( trainer_config.l2_decay_rate),
average_window=0.5, max_average_window=1000)) learning_rate_decay_a=0.5,
learning_rate_decay_b=train_sample_count,
learning_rate_schedule="discexp")
# define network topology # define network topology
loss = GNR(model_config) loss = GNR(model_config)
...@@ -180,22 +214,14 @@ def train(model_config, trainer_config): ...@@ -180,22 +214,14 @@ def train(model_config, trainer_config):
load_initial_model(trainer_config.init_model_path, parameters) load_initial_model(trainer_config.init_model_path, parameters)
else: else:
show_parameter_init_info(parameters) show_parameter_init_info(parameters)
# load the pre-trained embeddings parameters.set(
parameters.set("GloveVectors", "GloveVectors",
load_pretrained_parameters( load_pretrained_parameters(ModelConfig.pretrained_emb_path))
ModelConfig.pretrained_emb_path,
height=ModelConfig.vocab_size, trainer = paddle.trainer.SGD(
width=ModelConfig.embedding_dim)) cost=loss, parameters=parameters, update_equation=optimizer)
trainer = paddle.trainer.SGD(cost=loss, event_handler = build_event_handler(trainer_config, parameters, trainer)
parameters=parameters,
update_equation=optimizer)
train_reader, test_reader = build_reader(trainer_config.data_dir,
trainer_config.train_batch_size)
event_handler = build_event_handler(trainer_config, parameters, trainer,
test_reader)
trainer.train( trainer.train(
reader=train_reader, reader=train_reader,
num_passes=trainer_config.epochs, num_passes=trainer_config.epochs,
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或