-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathREADME
109 lines (86 loc) · 4.85 KB
/
README
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
NOTE
Any issues related to Turbo Parser itself, or issues running samples/tests, please file in TurboParser's main repository: https://github.com/andre-martins/TurboParser
README
TurboTextAnalysis project folder analysis
TurboTextAnalysis\Data
Models for TurboParser in english and spanish.
TurboTextAnalysis\deps
External files/projects that are required. A version of config++ is hereby made available.
TurboTextAnalysis\TurboTextAnalysis
Library that exposes the complete NLP pipeline exclusively in C++
on top of TurboParser (namely, the libturboparser library)
(replaces python code in "python" folder of TurboParser project folder).
TurboTextAnalysis\DemoTextAnalysisTest
Example code that uses the TurboTextAnalysis library.
TurboTextAnalysis\TurboTextAnalysisPython
Python wrapper over the TurboTextAnalysis, made with Cython.
TurboTextAnalysis\CrossPlatfTurboTextAnalysis
Cross-platform run-time dynamic linking library for TurboTextAnalysis.
(Reference: https://github.com/davidalbertonogueira/CrossPlatfDynamicLibraryPlugInArchitecture)
TurboTextAnalysis\TestCrossPlatfTurboTextAnalysis
Example code that uses the TurboTextAnalysis run-time dynamic linking library.
NOTE:
Python 3.X is required.
Cython is required (sudo apt-get install cython3).
ICU library is required (sudo apt-get install libicu-dev icu-devtools).
Build process
To Build TurboParser
1. Clone or download TurboParser from master branch
of https://github.com/Priberam/TurboParser
2. If OS is Linux - Ubuntu distro
2.1. you can use the install-ubuntu.sh script.
If OS is Linux - other distro
2.1. execute the following commands:
./install_deps.sh
automake --add-missing
./configure && make && make install
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:`pwd;`/deps/local/lib:"
If OS is Windows
2.1. In vsprojects folder you will find a Microsoft Visual Studio solution.
2.2. Firstly, in deps folder, you need to extract the dependency libraries
and compile them (they also have Microsoft Visual Studio solutions).
2.3. Use the Microsoft Visual Studio solution to build all the projects.
To Build libturboparser
1. Clone or download TurboParser from master branch
of https://github.com/Priberam/TurboParser
2. If OS is Linux
2.1 Go to the libturboparser folder in the TurboParser project folder.
2.2. Then, execute make using that Makefile.
If OS is Windows
2.1 In the Microsoft Visual Studio solution, one of the projects
is libturboparser. You just need to build that project
(if you already built the solution (i.e., all the projects),
you can ignore this step).
To Build TurboTextAnalysis
1. Obtain this repository.
1.1. In deps/, compile libconfig lib in libconfig-1.4.9/ folder.
2. If OS is Linux
2.1. Go to the TurboTextAnalysis folder in the TurboTextAnalysis
project folder, where you will find a Makefile.
2.2. Edit the Makefile in such a way that the paths are corrected
to the path in which you have TurboParser.
2.3. Then, execute make using that edited Makefile.
If OS is Windows
2.1. In the root folder of TurboTextAnalysis you will find a
Microsoft Visual Studio solution.
2.2. Edit the project files in order for the include directories
to point to the path in which you have TurboParser.
2.3. Build the TurboTextAnalysis project.
To Build TurboTextAnalysisPython
1. Go to the TurboTextAnalysisPython folder in the TurboTextAnalysis project folder.
2. Edit setup.py in order to point to the paths
of the necessary libraries (TurboTextAnalysis and TurboParser).
3. Execute
python setup.py build_ext --inplace
Note: If the default python version is 2.X, in linux you may need to call, for example:
python3.4 setup.py build_ext --inplace
If you are in Windows with an Anaconda environment, you may need to
activate a python 3 environment if you default is python 2.
4. Run
python test.py
to check that everything works.
================================================================================
Priberam’s TurboTextAnalysis is a TurboParser's wrapper for easy text analysis, allowing it to be readily used in production systems.
Copyright 2018 by PRIBERAM INFORMÁTICA, S.A. - www.priberam.com
Usage subject to The terms & Conditions of the "Priberam TurboTextAnalysis OS Software License" available at https://www.priberam.pt/docs/Priberam_TurboTextAnalysis_OS_Software_License.pdf
================================================================================