ltp_server
基于Python的用FastAPI简单封装的LTP服务
安装
pip install ltp_server
服务端
使用方式
方式一：Python库引用
示例：
from ltp_server import Server
if __name__ == '__main__':
model_path = r"/root/Data/NLP/Model/LTP"
# server = Server(model_path=model_path)
# server.run()
Server(model_path).run()
方式二：shell命令
示例：
ltp_server --model_path=/root/Data/NLP/Model/LTP
可用选项
| 参数名 | 是否可选 | 默认值 | 说明 |
|---|---|---|---|
| model_path | 否 | LTP模型路径（绝对路径） | |
| dict_path | 是 | None | 用户词表路径（绝对路径） |
| max_window | 是 | 4 | 前向分词最大窗口 |
| host | 是 | 127.0.0.1 | 服务主机名 |
| port | 是 | 8000 | 服务监听端口 |
服务概览
| 服务功能 | 服务路由 | 请求方式 |
|---|---|---|
| 分句 | /sent_split | POST |
| 增加自定义词语 | /add_words | POST |
| 分词 | /seg | POST |
| 词性标注 | /pos | POST |
| 命名实体识别 | /ner | POST |
| 语义角色标注 | /srl | POST |
| 依存句法分析 | /dep | POST |
| 语义依存分析（树） | /sdp | POST |
| 语义依存分析（图） | /sdpg | POST |
请求示例
分句
### sent_split
POST http://localhost:8000/sent_split
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值：
{
"texts": [
"曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"
],
"sents": [
"曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"
],
"status": 0
}
增加自定义词语
### add_words
POST http://localhost:8000/add_words
Content-Type: application/json
{
"words": ["江大桥"]
}
返回值
{
"status": 0
}
分词
### seg
POST http://localhost:8000/seg
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"
],
"res": [
[
"曹操",
"和",
"司马懿",
"去",
"赶集",
"，",
"中途",
"遇",
"上",
"关羽",
"，",
"一起",
"吃",
"了",
"个",
"饭",
"。"
]
]
}
词性标注
### pos
POST http://localhost:8000/pos
Content-Type: application/json
{
"texts": ["南京市长江大桥"]
}
返回值
{
"status": 0,
"texts": [
"南京市长江大桥"
],
"res": [
[
[
"南京市",
"ns"
],
[
"长江",
"ns"
],
[
"大桥",
"n"
]
]
],
"seg": [
[
"南京市",
"长江",
"大桥"
]
]
}
命名实体识别
### ner
POST http://localhost:8000/ner
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"乔丹是一位出生在纽约的美国职业篮球运动员。"
],
"res": [
[
[
"乔丹",
"Nh",
0,
0
],
[
"纽约",
"Ns",
6,
6
],
[
"美国",
"Ns",
8,
8
]
]
],
"seg": [
[
"乔丹",
"是",
"一",
"位",
"出生",
"在",
"纽约",
"的",
"美国",
"职业",
"篮球",
"运动员",
"。"
]
]
}
语义角色标注
### srl
POST http://localhost:8000/srl
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"乔丹是一位出生在纽约的美国职业篮球运动员。"
],
"res": [
[
[
"是",
1,
[
[
"A0",
[
"乔丹"
],
0,
0
],
[
"A1",
[
"一",
"位",
"出生",
"在",
"纽约",
"的",
"美国",
"职业",
"篮球",
"运动员"
],
2,
11
]
]
],
[
"出生",
4,
[
[
"A1",
[
"在",
"纽约"
],
5,
6
],
[
"A0",
[
"职业",
"篮球",
"运动员"
],
9,
11
]
]
]
]
],
"seg": [
[
"乔丹",
"是",
"一",
"位",
"出生",
"在",
"纽约",
"的",
"美国",
"职业",
"篮球",
"运动员",
"。"
]
]
}
依存句法分析
### dep
POST http://localhost:8000/dep
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"乔丹是一位出生在纽约的美国职业篮球运动员。"
],
"res": [
[
[
1,
"乔丹",
2,
"是",
"SBV"
],
[
2,
"是",
0,
"ROOT",
"HED"
],
[
3,
"一",
4,
"位",
"ATT"
],
[
4,
"位",
12,
"运动员",
"ATT"
],
[
5,
"出生",
12,
"运动员",
"ATT"
],
[
6,
"在",
5,
"出生",
"CMP"
],
[
7,
"纽约",
6,
"在",
"POB"
],
[
8,
"的",
5,
"出生",
"RAD"
],
[
9,
"美国",
12,
"运动员",
"ATT"
],
[
10,
"职业",
12,
"运动员",
"ATT"
],
[
11,
"篮球",
12,
"运动员",
"ATT"
],
[
12,
"运动员",
2,
"是",
"VOB"
],
[
13,
"。",
2,
"是",
"WP"
]
]
],
"seg": [
[
"乔丹",
"是",
"一",
"位",
"出生",
"在",
"纽约",
"的",
"美国",
"职业",
"篮球",
"运动员",
"。"
]
]
}
语义依存分析（树）
### sdp
POST http://localhost:8000/sdp
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"
],
"res": [
[
[
1,
"曹操",
4,
"去",
"AGT"
],
[
1,
"曹操",
5,
"赶集",
"AGT"
],
[
2,
"和",
3,
"司马懿",
"mRELA"
],
[
3,
"司马懿",
4,
"去",
"AGT"
],
[
4,
"去",
0,
"ROOT",
"Root"
],
[
5,
"赶集",
4,
"去",
"eSUCC"
],
[
6,
"，",
5,
"赶集",
"mPUNC"
],
[
7,
"中途",
8,
"遇",
"MANN"
],
[
8,
"遇",
5,
"赶集",
"eSUCC"
],
[
9,
"上",
8,
"遇",
"mDEPD"
],
[
10,
"关羽",
8,
"遇",
"DATV"
],
[
11,
"，",
8,
"遇",
"mPUNC"
],
[
12,
"一起",
13,
"吃",
"MANN"
],
[
13,
"吃",
8,
"遇",
"eSUCC"
],
[
14,
"了",
13,
"吃",
"mDEPD"
],
[
15,
"个",
16,
"饭",
"MEAS"
],
[
16,
"饭",
13,
"吃",
"PAT"
],
[
17,
"。",
13,
"吃",
"mPUNC"
]
]
],
"seg": [
[
"曹操",
"和",
"司马懿",
"去",
"赶集",
"，",
"中途",
"遇",
"上",
"关羽",
"，",
"一起",
"吃",
"了",
"个",
"饭",
"。"
]
]
}
语义依存分析（图）
### sdpg
POST http://localhost:8000/sdpg
Content-Type: application/json
{
"texts": ["曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"]
}
返回值
{
"status": 0,
"texts": [
"曹操和司马懿去赶集，中途遇上关羽，一起吃了个饭。"
],
"res": [
[
[
1,
"曹操",
4,
"去",
"AGT"
],
[
1,
"曹操",
5,
"赶集",
"AGT"
],
[
2,
"和",
3,
"司马懿",
"mRELA"
],
[
3,
"司马懿",
4,
"去",
"AGT"
],
[
4,
"去",
0,
"ROOT",
"Root"
],
[
5,
"赶集",
4,
"去",
"eSUCC"
],
[
6,
"，",
5,
"赶集",
"mPUNC"
],
[
7,
"中途",
8,
"遇",
"MANN"
],
[
8,
"遇",
5,
"赶集",
"eSUCC"
],
[
9,
"上",
8,
"遇",
"mDEPD"
],
[
10,
"关羽",
8,
"遇",
"DATV"
],
[
11,
"，",
8,
"遇",
"mPUNC"
],
[
12,
"一起",
13,
"吃",
"MANN"
],
[
13,
"吃",
8,
"遇",
"eSUCC"
],
[
14,
"了",
13,
"吃",
"mDEPD"
],
[
15,
"个",
16,
"饭",
"MEAS"
],
[
16,
"饭",
13,
"吃",
"PAT"
],
[
17,
"。",
13,
"吃",
"mPUNC"
]
]
],
"seg": [
[
"曹操",
"和",
"司马懿",
"去",
"赶集",
"，",
"中途",
"遇",
"上",
"关羽",
"，",
"一起",
"吃",
"了",
"个",
"饭",
"。"
]
]
}
客户端
使用方式
方式一：Python库使用
示例如下：
from ltp_server import Client
if __name__ == '__main__':
client = Client()
texts = ["乔丹是一位出生在纽约的美国职业篮球运动员。"]
print(client.sent_split(texts))
print(client.seg(texts))
print(client.pos(texts))
print(client.ner(texts))
print(client.srl(texts))
print(client.dep(texts))
print(client.sdp(texts))
print(client.sdpg(texts))
请求结果：
{'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'status': 0}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[['乔丹', 'nh'], ['是', 'v'], ['一', 'm'], ['位', 'q'], ['出生', 'v'], ['在', 'p'], ['纽约', 'ns'], ['的', 'u'], ['美国', 'ns'], ['职业', 'n'], ['篮球', 'n'], ['运动员', 'n'], ['。', 'wp']]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[['乔丹', 'Nh', 0, 0], ['纽约', 'Ns', 6, 6], ['美国', 'Ns', 8, 8]]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[['是', 1, [['A0', ['乔丹'], 0, 0], ['A1', ['一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员'], 2, 11]]], ['出生', 4, [['A1', ['在', '纽约'], 5, 6], ['A0', ['职业', '篮球', '运动员'], 9, 11]]]]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[[1, '乔丹', 2, '是', 'SBV'], [2, '是', 0, 'ROOT', 'HED'], [3, '一', 4, '位', 'ATT'], [4, '位', 12, '运动员', 'ATT'], [5, '出生', 12, '运动员', 'ATT'], [6, '在', 5, '出生', 'CMP'], [7, '纽约', 6, '在', 'POB'], [8, '的', 5, '出生', 'RAD'], [9, '美国', 12, '运动员', 'ATT'], [10, '职业', 12, '运动员', 'ATT'], [11, '篮球', 12, '运动员', 'ATT'], [12, '运动员', 2, '是', 'VOB'], [13, '。', 2, '是', 'WP']]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[[1, '乔丹', 2, '是', 'EXP'], [2, '是', 0, 'ROOT', 'Root'], [3, '一', 4, '位', 'MEAS'], [4, '位', 12, '运动员', 'MEAS'], [5, '出生', 12, '运动员', 'rEXP'], [6, '在', 7, '纽约', 'mRELA'], [7, '纽约', 5, '出生', 'LOC'], [8, '的', 5, '出生', 'mDEPD'], [9, '美国', 12, '运动员', 'FEAT'], [10, '职业', 11, '篮球', 'FEAT'], [10, '职业', 12, '运动员', 'FEAT'], [11, '篮球', 12, '运动员', 'FEAT'], [12, '运动员', 2, '是', 'LINK'], [13, '。', 2, '是', 'mPUNC']]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
{'status': 0, 'texts': ['乔丹是一位出生在纽约的美国职业篮球运动员。'], 'res': [[[1, '乔丹', 2, '是', 'EXP'], [2, '是', 0, 'ROOT', 'Root'], [3, '一', 4, '位', 'MEAS'], [4, '位', 12, '运动员', 'MEAS'], [5, '出生', 12, '运动员', 'rEXP'], [6, '在', 7, '纽约', 'mRELA'], [7, '纽约', 5, '出生', 'LOC'], [8, '的', 5, '出生', 'mDEPD'], [9, '美国', 12, '运动员', 'FEAT'], [10, '职业', 11, '篮球', 'FEAT'], [10, '职业', 12, '运动员', 'FEAT'], [11, '篮球', 12, '运动员', 'FEAT'], [12, '运动员', 2, '是', 'LINK'], [13, '。', 2, '是', 'mPUNC']]], 'seg': [['乔丹', '是', '一', '位', '出生', '在', '纽约', '的', '美国', '职业', '篮球', '运动员', '。']]}
方式二：自己通过http请求调用
略
参考
1 maintainer
Release files for ltp-server 0.2.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ltp_server-0.2.1.tar.gz | 11.1 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ltp_server-0.2.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 19.2 kB
Release files / ltp_server-0.2.1.tar.gz
| Download URL | ltp_server-0.2.1.tar.gz |
|---|---|
| Size | 11.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a0aa2973530aeb446afae52c6e3d668c168b38800b07999ecd0d8b94bb4fbba5
|
|
BLAKE2b-256 checksum How to use checksums |
e5d399726f699336d665f1a02f71b56247eb06ad2a0d02445a2620381a35ea65
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.2.0 pkginfo/1.6.1 requests/2.23.0 setuptools/46.4.0.post20200518 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.8.3
|
Release files / ltp_server-0.2.1-py3-none-any.whl
| Download URL | ltp_server-0.2.1-py3-none-any.whl |
|---|---|
| Size | 8.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
78437acf2462b66487b574235bd1aed1522a3eadf62c94bae0e859bb60603755
|
|
BLAKE2b-256 checksum How to use checksums |
3bca044ffe38cd7b952c900281919edc95c6c3d6beaddc3a47440fdd86850a01
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.2.0 pkginfo/1.6.1 requests/2.23.0 setuptools/46.4.0.post20200518 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.8.3
|
