-
Notifications
You must be signed in to change notification settings - Fork 3
/
get_data_mlm_pretraining.sh
executable file
·161 lines (127 loc) · 4.14 KB
/
get_data_mlm_pretraining.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
#!/usr/bin/env bash
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
#
set -e
# Data preprocessing configuration
CODES=32000 # number of BPE codes
N_THREADS=16 # number of threads in data preprocessing
# Read arguments
POSITIONAL=()
while [[ $# -gt 0 ]]
do
key="$1"
case $key in
--src)
SRC="$2"; shift 2;;
*)
POSITIONAL+=("$1")
shift
;;
esac
done
set -- "${POSITIONAL[@]}"
# Check parameters
if [ "$SRC" == "" ]; then echo "--src not provided"; exit; fi
# Initialize tools and data paths
# main paths
MAIN_PATH=$PWD
TOOLS_PATH=$PWD/tools
DATA_PATH=./data/$SRC
# create paths
mkdir -p $TOOLS_PATH
# moses
MOSES=$TOOLS_PATH/mosesdecoder
REPLACE_UNICODE_PUNCT=$MOSES/scripts/tokenizer/replace-unicode-punctuation.perl
NORM_PUNC=$MOSES/scripts/tokenizer/normalize-punctuation.perl
REM_NON_PRINT_CHAR=$MOSES/scripts/tokenizer/remove-non-printing-char.perl
TOKENIZER=$MOSES/scripts/tokenizer/tokenizer.perl
INPUT_FROM_SGM=$MOSES/scripts/ems/support/input-from-sgm.perl
# fastBPE
FASTBPE_DIR=$TOOLS_PATH/fastBPE
FASTBPE=$TOOLS_PATH/fastBPE/fast
# Sennrich's WMT16 scripts for Romanian pre-processing
WMT16_SCRIPTS=$TOOLS_PATH/wmt16-scripts
NORMALIZE_ROMANIAN=$WMT16_SCRIPTS/preprocess/normalise-romanian.py
REMOVE_DIACRITICS=$WMT16_SCRIPTS/preprocess/remove-diacritics.py
# train/ valid / test file raw data
SRC_TRAIN=$DATA_PATH/train_raw.$SRC
SRC_VALID=$DATA_PATH/valid_raw.$SRC
SRC_TEST=$DATA_PATH/test_raw.$SRC
# tokenized files
SRC_TRAIN_TOK=$SRC_TRAIN.tok
SRC_VALID_TOK=$SRC_VALID.tok
SRC_TEST_TOK=$SRC_TEST.tok
# train / valid / test monolingual BPE data
SRC_TRAIN_BPE=$DATA_PATH/train.$SRC
SRC_VALID_BPE=$DATA_PATH/valid.$SRC
SRC_TEST_BPE=$DATA_PATH/test.$SRC
# BPE / vocab files
BPE_CODES=$DATA_PATH/codes.$SRC
SRC_VOCAB=$DATA_PATH/vocab.$SRC
# install tools
./install-tools.sh
#
# Download monolingual data
#
# preprocessing commands - special case for Romanian
if [ "$SRC" == "ro" ]; then
SRC_PREPROCESSING="$REPLACE_UNICODE_PUNCT | $NORM_PUNC -l $SRC | $REM_NON_PRINT_CHAR | $NORMALIZE_ROMANIAN | $REMOVE_DIACRITICS | $TOKENIZER -l $SRC -no-escape -threads $N_THREADS"
else
SRC_PREPROCESSING="$REPLACE_UNICODE_PUNCT | $NORM_PUNC -l $SRC | $REM_NON_PRINT_CHAR | $TOKENIZER -l $SRC -no-escape -threads $N_THREADS"
fi
# tokenize data
if ! [[ -f "$SRC_TRAIN_TOK" ]]; then
echo "Tokenize $SRC monolingual data..."
eval "cat $SRC_TRAIN | $SRC_PREPROCESSING > $SRC_TRAIN_TOK"
fi
echo "$SRC monolingual data tokenized in: $SRC_TRAIN_TOK"
# learn BPE codes
if [ ! -f "$BPE_CODES" ]; then
echo "Learning BPE codes..."
$FASTBPE learnbpe $CODES $SRC_TRAIN_TOK > $BPE_CODES
fi
echo "BPE learned in $BPE_CODES"
# apply BPE codes
if ! [[ -f "$SRC_TRAIN_BPE" ]]; then
echo "Applying $SRC BPE codes..."
$FASTBPE applybpe $SRC_TRAIN_BPE $SRC_TRAIN_TOK $BPE_CODES
fi
echo "BPE codes applied to $SRC in: $SRC_TRAIN_BPE"
# extract source vocabulary
if ! [[ -f "$SRC_VOCAB" ]]; then
echo "Extracting vocabulary..."
$FASTBPE getvocab $SRC_TRAIN_BPE > $SRC_VOCAB
fi
echo "$SRC vocab in: $SRC_VOCAB"
# binarize data
if ! [[ -f "$SRC_TRAIN_BPE.pth" ]]; then
echo "Binarizing $SRC data..."
$MAIN_PATH/preprocess.py $SRC_VOCAB $SRC_TRAIN_BPE
fi
echo "$SRC binarized data in: $SRC_TRAIN_BPE.pth"
echo "Tokenizing valid and test data..."
# tokenize data
if ! [[ -f "$SRC_VALID_TOK" ]]; then
eval "cat $SRC_VALID | $SRC_PREPROCESSING > $SRC_VALID_TOK"
fi
if ! [[ -f "$SRC_TEST_TOK" ]]; then
eval "cat $SRC_TEST | $SRC_PREPROCESSING > $SRC_TEST_TOK"
fi
echo "Applying BPE to valid and test files..."
$FASTBPE applybpe $SRC_VALID_BPE "$SRC_VALID_TOK" $BPE_CODES $SRC_VOCAB
$FASTBPE applybpe $SRC_TEST_BPE "$SRC_TEST_TOK" $BPE_CODES $SRC_VOCAB
echo "Binarizing data..."
rm -f $SRC_VALID_BPE.pth $SRC_TEST_BPE.pth
$MAIN_PATH/preprocess.py $SRC_VOCAB $SRC_VALID_BPE
$MAIN_PATH/preprocess.py $SRC_VOCAB $SRC_TEST_BPE
# Summary
echo ""
echo "===== Data summary"
echo "Monolingual training data:"
echo " $SRC: $SRC_TRAIN_BPE.pth"
echo "Monolingual validation data:"
echo " $SRC: $SRC_VALID_BPE.pth"
echo "Monolingual test data:"
echo " $SRC: $SRC_TEST_BPE.pth"