Merge pull request #119 from HIT-SCIR/develop

Update SRL API and models to support LTP 3.4.0
HIT-SCIR · Dec 10, 2017 · cd04600 · cd04600
2 parents d931476 + 67ccdee
commit cd04600
Show file tree

Hide file tree

Showing 2,587 changed files with 390,223 additions and 109,474 deletions.
diff --git a/.gitignore b/.gitignore
@@ -16,9 +16,18 @@ build
 include/
 lib/
 bin/
+dist
+!patch/include
+pyltp.egg-info
 
 *.swp
 doc/_build
 doc/_static
 doc/_templates
 !doc/Makefile
+
+###############
+# data        #
+###############
+ltp_data
+
diff --git a/.travis.yml b/.travis.yml
@@ -11,10 +11,6 @@ matrix:
           sudo: required
           python: 2.7
           env: TOXENV=py27
-        - os: linux
-          sudo: required
-          python: 3.2
-          env: TOXENV=py32
         - os: linux
           sudo: required
           python: 3.3
@@ -27,6 +23,10 @@ matrix:
           sudo: required
           python: 3.5
           env: TOXENV=py35
+        - os: linux
+          sudo: required
+          python: 3.6
+          env: TOXENV=py36
         - os: osx
           language: generic
           env: TOXENV=py2

diff --git a/.travis/install.sh b/.travis/install.sh
@@ -26,6 +26,6 @@ export PYLTPVER=$(${PY} setup.py --version)
 $PY setup.py build
 $PY setup.py sdist
 cd dist/
-tar zxvf pyltp-$PYLTPVER.tar.gz > /dev/null
+tar zxvf pyltp-$PYLTPVER.tar.gz
 cd pyltp-$PYLTPVER
-$PY setup.py build >& /dev/null
+$PY setup.py build
diff --git a/MANIFEST.in b/MANIFEST.in
@@ -9,8 +9,9 @@ recursive-include ltp/src/segmentor *.cpp *.h *.hpp
 recursive-include ltp/src/postagger *.cpp *.h *.hpp
 recursive-include ltp/src/ner *.cpp *.h *.hpp
 recursive-include ltp/src/parser.n *.cpp *.h *.hpp
-recursive-include ltp/src/srl *.cpp *.h *.hpp *.hh
-recursive-include ltp/thirdparty/boost *.h *.hpp *.cpp
-recursive-include ltp/thirdparty/eigen-3.2.4 *
+recursive-include ltp/src/srl *.cpp *.h
+recursive-include ltp/thirdparty/boost *.h *.hpp *.cpp *.ipp
+recursive-include ltp/thirdparty/eigen *
+recursive-include ltp/thirdparty/dynet *
 recursive-include ltp/thirdparty/maxent *.h *.cpp
 recursive-include patch *.h *.hpp *.cpp
diff --git a/README.md b/README.md
@@ -1,4 +1,4 @@
-# pyltp
+# pyltp
 
 [![PyPI Status](https://badge.fury.io/py/pyltp.svg)](https://badge.fury.io/py/pyltp)
 [![Readthedocs](https://readthedocs.org/projects/pyltp/badge/?version=latest)](http://pyltp.readthedocs.io/)
@@ -37,23 +37,23 @@ segmentor.release()
 	$ pip install pyltp
 	```
 	或从源代码安装
-	
+
 	```
 	$ git clone https://github.com/HIT-SCIR/pyltp
 	$ git submodule init
 	$ git submodule update
-	$ python setup.py install
+	$ python setup.py install # Mac系统出现版本问题使用 MACOSX_DEPLOYMENT_TARGET=10.7 python setup.py install
 	```
 
 * 第二步，下载模型文件
 
-	[百度云](http://pan.baidu.com/share/link?shareid=1988562907&uk=2738088569)，当前模型版本 3.3.1
+	[七牛云](http://ltp.ai/download.html)，当前模型版本 3.4.0
 
 ## 版本对应
 
-* pyltp 版本：0.1.9
-* LTP 版本：3.3.2
-* 模型版本：3.3.1
+* pyltp 版本：0.2.0
+* LTP 版本：3.4.0
+* 模型版本：3.4.0
 
 ## 作者
 

diff --git a/appveyor.yml b/appveyor.yml
@@ -1,6 +1,15 @@
 before_build:
   - git submodule init
   - git submodule update
-
+
+image:
+  - Visual Studio 2015
+  - Visual Studio 2017
+
+environment:
+  matrix:
+    - PY: C:\Python36-x64
+    - PY: C:\Python35-x64
+
 build_script:
-  - python setup.py build
+  - "%PY%\\python.exe setup.py build"
diff --git a/doc/api.rst b/doc/api.rst
@@ -11,13 +11,10 @@ pyltp 是 `LTP <https://github.com/HIT-SCIR/ltp>`_ 的 Python 封装，提供了
 请先下载完整的 LTP 模型文件
 
 * 下载地址 - `百度云 <http://pan.baidu.com/share/link?shareid=1988562907&uk=2738088569>`_
-* 当前模型版本 - 3.3.1
+* 当前模型版本 - 3.4.0
 
 请确保下载的模型版本与当前版本的 pyltp 对应，否则会导致程序无法正确加载模型。
 
-如果您是从 Github 获取的源代码，其中包含的 :file:`ltp_data` 目录仅为测试用的模型文件，不能产生正确的分析结果。
-
-
 请注意编码
 ----------
 
@@ -262,9 +259,8 @@ B、I、E、S位置标签和实体类型标签之间用一个横线 :code:`-`
 
     words = ['元芳', '你', '怎么', '看']
     postags = ['nh', 'r', 'r', 'v']
-    netags = ['S-Nh', 'O', 'O', 'O']
     # arcs 使用依存句法分析的结果
-    roles = labeller.label(words, postags, netags, arcs)  # 语义角色标注
+    roles = labeller.label(words, postags, arcs)  # 语义角色标注
 
     # 打印结果
     for role in roles:

diff --git a/doc/changelog.rst b/doc/changelog.rst
@@ -0,0 +1 @@
+* 2017年12月05日 升级更新兼容 LTP 3.4.0
diff --git a/doc/conf.py b/doc/conf.py
@@ -46,17 +46,17 @@
 
 # General information about the project.
 project = u'pyltp'
-copyright = u'2016, HIT-SCIR'
+copyright = u'2017, HIT-SCIR'
 author = u'HIT-SCIR'
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
 # built documents.
 #
 # The short X.Y version.
-version = u'0.1.9'
+version = u'0.2.0'
 # The full version, including alpha/beta/rc tags.
-release = u'0.1.9'
+release = u'0.2.0'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

diff --git a/doc/install.rst b/doc/install.rst
@@ -1,6 +1,8 @@
 安装 pyltp
 ===========
 
+* 注：由于新版本增加了新的第三方依赖如dynet等，不再支持 windows 下 python2 环境。
+
 使用 pip 安装
 -------------
 
@@ -10,8 +12,9 @@
 
 接下来，需要下载 LTP 模型文件。
 
-* 下载地址 - `百度云 <http://pan.baidu.com/share/link?shareid=1988562907&uk=2738088569>`_
-* 当前模型版本 - 3.3.1
+* 下载地址 - `模型下载 http://ltp.ai/download.html`_
+* 当前模型版本 - 3.4.0
+* 注意在windows下 3.4.0 版本的 语义角色标注模块 模型需要单独下载，具体查看下载地址链接中的说明。
 
 请确保下载的模型版本与当前版本的 pyltp 对应，否则会导致程序无法正确加载模型。
 

diff --git a/example/example.py b/example/example.py
@@ -6,7 +6,7 @@
 sys.path = [os.path.join(ROOTDIR, "lib")] + sys.path
 
 # Set your own model path
-MODELDIR=os.path.join(ROOTDIR, "ltp_data")
+MODELDIR=os.path.join(ROOTDIR, "./ltp_data")
 
 from pyltp import SentenceSplitter, Segmentor, Postagger, Parser, NamedEntityRecognizer, SementicRoleLabeller
 
@@ -17,33 +17,33 @@
 segmentor = Segmentor()
 segmentor.load(os.path.join(MODELDIR, "cws.model"))
 words = segmentor.segment(sentence)
-print "\t".join(words)
+print("\t".join(words))
 
 postagger = Postagger()
 postagger.load(os.path.join(MODELDIR, "pos.model"))
 postags = postagger.postag(words)
 # list-of-string parameter is support in 0.1.5
 # postags = postagger.postag(["中国","进出口","银行","与","中国银行","加强","合作"])
-print "\t".join(postags)
+print("\t".join(postags))
 
 parser = Parser()
 parser.load(os.path.join(MODELDIR, "parser.model"))
 arcs = parser.parse(words, postags)
 
-print "\t".join("%d:%s" % (arc.head, arc.relation) for arc in arcs)
+print("\t".join("%d:%s" % (arc.head, arc.relation) for arc in arcs))
 
 recognizer = NamedEntityRecognizer()
 recognizer.load(os.path.join(MODELDIR, "ner.model"))
 netags = recognizer.recognize(words, postags)
-print "\t".join(netags)
+print("\t".join(netags))
 
 labeller = SementicRoleLabeller()
-labeller.load(os.path.join(MODELDIR, "srl/"))
-roles = labeller.label(words, postags, netags, arcs)
+labeller.load(os.path.join(MODELDIR, "pisrl.model"))
+roles = labeller.label(words, postags, arcs)
 
 for role in roles:
-    print role.index, "".join(
-            ["%s:(%d,%d)" % (arg.name, arg.range.start, arg.range.end) for arg in role.arguments])
+    print(role.index, "".join(
+            ["%s:(%d,%d)" % (arg.name, arg.range.start, arg.range.end) for arg in role.arguments]))
 
 segmentor.release()
 postagger.release()

diff --git a/ltp b/ltp
diff --git a/ltp_data/cws.model b/ltp_data/cws.model
diff --git a/ltp_data/ner.model b/ltp_data/ner.model
diff --git a/ltp_data/parser.model b/ltp_data/parser.model
diff --git a/ltp_data/pos.model b/ltp_data/pos.model
diff --git a/ltp_data/srl/Chinese.xml b/ltp_data/srl/Chinese.xml
diff --git a/ltp_data/srl/prg.model b/ltp_data/srl/prg.model
diff --git a/ltp_data/srl/srl.cfg b/ltp_data/srl/srl.cfg
diff --git a/ltp_data/srl/srl.model b/ltp_data/srl/srl.model
diff --git a/patch/__init__.py b/patch/__init__.py