add the paddlenlp server

applications/text_classification/hierarchical/deploy/SimpleServing/README.md

@@ -0,0 +1,39 @@
+# 基于PaddleNLP SimpleServing 的服务化部署
+
+## 目录
+- [环境准备](#环境准备)
+- [Server启动服务](#Server服务启动)
+- [其他参数设置](#其他参数设置)
+
+## 环境准备
+使用有SimpleServing功能的PaddleNLP版本
+```shell
+pip install paddlenlp --upgrade
+```
+## Server服务启动
+### 分类任务启动
+#### 启动 分类 Server 服务
+```bash
+paddlenlp server server:app --host 0.0.0.0 --port 8189
+```
+
+#### 分类任务发送服务
+```bash
+python client.py
+```
+
+
+## 其他参数设置
+可以在client端设置 `max_seq_len`, `batch_size`, `prob_limit` 参数
+```python
+    data = {
+        'data': {
+            'text': texts,
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size,
+            'prob_limit': args.prob_limit
+        }
+    }
+```

applications/text_classification/hierarchical/deploy/SimpleServing/client.py

@@ -0,0 +1,46 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+import requests
+import json
+
+# yapf: disable
+parser = argparse.ArgumentParser()
+parser.add_argument("--max_seq_len", default=128, type=int, help="The maximum total input sequence length after tokenization.")
+parser.add_argument("--batch_size", default=1, type=int, help="Batch size per GPU/CPU for predicting.")
+parser.add_argument("--prob_limit", default=0.5, type=float, help="The limitation of probability for the label.")
+args = parser.parse_args()
+# yapf: enable
+
+url = "http://0.0.0.0:8189/models/cls_hierarchical"
+headers = {"Content-Type": "application/json"}
+
+if __name__ == "__main__":
+    texts = [
+        '请问木竭胶囊能同高血压药、氨糖同时服吗？', '低压100*高压140*头涨，想吃点降压药。谢谢！', '脑穿通畸形易发人群有哪些',
+        '幼儿乱吃丙硫氧嘧啶片怎么办，我也不知道她吃了几片', '如果是可以降血糖的话,血糖值7点多的大概需要吃几个疗程?'
+    ]
+    data = {
+        'data': {
+            'text': texts,
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size,
+            'prob_limit': args.prob_limit
+        }
+    }
+    r = requests.post(url=url, headers=headers, data=json.dumps(data))
+    print(r.text)

applications/text_classification/hierarchical/deploy/SimpleServing/server.py

@@ -0,0 +1,23 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddlenlp import SimpleServer
+from paddlenlp.server import CustomModelHandler, MultiLabelClassificationPostHandler
+
+app = SimpleServer()
+app.register('cls_hierarchical',
+             model_path="../../export",
+             tokenizer_name='ernie-3.0-medium-zh',
+             model_handler=CustomModelHandler,
+             post_handler=MultiLabelClassificationPostHandler)

applications/text_classification/multi_class/deploy/SimpleServing/README.md

@@ -0,0 +1,39 @@
+# 基于PaddleNLP SimpleServing 的服务化部署
+
+## 目录
+- [环境准备](#环境准备)
+- [Server启动服务](#Server服务启动)
+- [其他参数设置](#其他参数设置)
+
+## 环境准备
+使用有SimpleServing功能的PaddleNLP版本
+```shell
+pip install paddlenlp >= 2.4.4
+```
+## Server服务启动
+### 分类任务启动
+#### 启动分类 Server 服务
+```bash
+paddlenlp server server:app --host 0.0.0.0 --port 8189
+```
+
+#### 启动分类 Client 服务
+```bash
+python client.py
+```
+
+
+## 其他参数设置
+可以在client端设置 `max_seq_len`, `batch_size` 参数
+```python
+    data = {
+        'data': {
+            'text': texts,
+            'text_pair': text_pairs if len(text_pairs) > 0 else None
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size
+        }
+    }
+```

applications/text_classification/multi_class/deploy/SimpleServing/client.py

@@ -0,0 +1,45 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+import requests
+import json
+
+# yapf: disable
+parser = argparse.ArgumentParser()
+parser.add_argument("--max_seq_len", default=128, type=int, help="The maximum total input sequence length after tokenization.")
+parser.add_argument("--batch_size", default=1, type=int, help="Batch size per GPU/CPU for predicting.")
+args = parser.parse_args()
+# yapf: enable
+
+url = "http://0.0.0.0:8189/models/cls_multi_class"
+headers = {"Content-Type": "application/json"}
+
+if __name__ == "__main__":
+    texts = [
+        '黑苦荞茶的功效与作用及食用方法', '交界痣会凸起吗', '检查是否能怀孕挂什么科', '鱼油怎么吃咬破吃还是直接咽下去',
+        '幼儿挑食的生理原因是'
+    ]
+    data = {
+        'data': {
+            'text': texts,
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size
+        }
+    }
+    r = requests.post(url=url, headers=headers, data=json.dumps(data))
+    result_json = json.loads(r.text)
+    print(result_json)

applications/text_classification/multi_class/deploy/SimpleServing/server.py

@@ -0,0 +1,23 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddlenlp import SimpleServer
+from paddlenlp.server import CustomModelHandler, MultiClassificationPostHandler
+
+app = SimpleServer()
+app.register('cls_multi_class',
+             model_path="../../export",
+             tokenizer_name='ernie-3.0-medium-zh',
+             model_handler=CustomModelHandler,
+             post_handler=MultiClassificationPostHandler)

applications/text_classification/multi_label/deploy/SimpleServing/README.md

@@ -0,0 +1,38 @@
+# 基于PaddleNLP SimpleServing 的服务化部署
+
+## 目录
+- [环境准备](#环境准备)
+- [Server启动服务](#Server服务启动)
+- [其他参数设置](#其他参数设置)
+
+## 环境准备
+使用有SimpleServing功能的PaddleNLP版本
+```shell
+pip install paddlenlp --upgrade
+```
+## Server服务启动
+### 分类任务启动
+#### 启动 分类 Server 服务
+```bash
+paddlenlp server server:app --host 0.0.0.0 --port 8189
+```
+
+#### 分类任务发送服务
+```bash
+python client.py
+```
+
+## 其他参数设置
+可以在client端设置 `max_seq_len`, `batch_size`, `prob_limit` 参数
+```python
+    data = {
+        'data': {
+            'text': texts,
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size,
+            'prob_limit': args.prob_limit
+        }
+    }
+```

applications/text_classification/multi_label/deploy/SimpleServing/client.py

@@ -0,0 +1,47 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+import requests
+import json
+
+# yapf: disable
+parser = argparse.ArgumentParser()
+parser.add_argument("--max_seq_len", default=128, type=int, help="The maximum total input sequence length after tokenization.")
+parser.add_argument("--batch_size", default=1, type=int, help="Batch size per GPU/CPU for predicting.")
+parser.add_argument("--prob_limit", default=0.5, type=float, help="The limitation of probability for the label.")
+args = parser.parse_args()
+# yapf: enable
+
+url = "http://0.0.0.0:8189/models/cls_multi_label"
+headers = {"Content-Type": "application/json"}
+
+if __name__ == "__main__":
+    texts = [
+        '原、被告另购置橱柜、碗架、电磁炉、电饭锅各一个归原告王某某所有。',
+        '于是原告到儿子就读的幼儿园进行探望，被告碰见后对原告破口大骂，还不让儿子叫原告妈妈，而叫被告现在的妻子做妈妈。',
+        '由我全额出资购买的联想台式电脑，我均依次放弃。'
+    ]
+    data = {
+        'data': {
+            'text': texts,
+        },
+        'parameters': {
+            'max_seq_len': args.max_seq_len,
+            'batch_size': args.batch_size,
+            'prob_limit': args.prob_limit
+        }
+    }
+    r = requests.post(url=url, headers=headers, data=json.dumps(data))
+    print(json.loads(r.text))

applications/text_classification/multi_label/deploy/SimpleServing/server.py

@@ -0,0 +1,23 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddlenlp import SimpleServer
+from paddlenlp.server import CustomModelHandler, MultiLabelClassificationPostHandler
+
+app = SimpleServer()
+app.register('cls_multi_label',
+             model_path="../../export",
+             tokenizer_name='ernie-3.0-medium-zh',
+             model_handler=CustomModelHandler,
+             post_handler=MultiLabelClassificationPostHandler)

model_zoo/ernie-health/cblue/README.md

@@ -108,11 +108,22 @@ $ python train_spo.py --batch_size 12 --max_seq_length 300 --learning_rate 6e-5
 使用动态图训练结束之后，还可以将动态图参数导出成静态图参数，用于部署推理等，具体代码见export_model.py。静态图参数保存在`output_path`指定路径中。
 
 运行方式：
+1. 分类任务静态图模型导出
+```shell
+python export_model.py --train_dataset CHIP-CDN-2C --params_path=./checkpoint/model_900/ --output_path=./export
+```
 
+2. SPO任务静态图模型导出
 ```shell
 python export_model.py --train_dataset CMeIE --params_path=./checkpoint/model_900/ --output_path=./export
 ```
 
+3. NER任务静态图模型导出
+```shell
+python export_model.py --train_dataset CMeEE --params_path=./checkpoint/model_1500/ --output_path=./export
+```
+
+**NOTICE**: train_dataset分类任务选择填上训练数据集名称，params_path选择最好参数的模型的路径。
 
 [1] CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark [pdf](https://arxiv.org/abs/2106.08087) [git](https://github.com/CBLUEbenchmark/CBLUE) [web](https://tianchi.aliyun.com/specials/promotion/2021chinesemedicalnlpleaderboardchallenge)