Ver Fonte

实体连接,上线eas配置

luojiehua há 3 anos atrás
pai
commit
8c8a4419d3

+ 1 - 0
.gitignore

@@ -8,3 +8,4 @@
 /BiddingKG/dl/test/traindata/
 /BiddingKG/dl/product/data/
 /BiddingKG/dl/channel/data/
+node_modules

+ 19 - 1
.idea/sonarlint/issuestore/index.pb

@@ -38,4 +38,22 @@ L
 Q
 !BiddingKG/dl/form/generateData.py,7\e\7e590aa47c1871cc7d75ac844d5769bff50a6e70
 U
-%BiddingKG/maxcompute/attachmentRec.py,b\e\be8f50b8961bc8ae61e105517763f21c707ea3ec
+%BiddingKG/maxcompute/attachmentRec.py,b\e\be8f50b8961bc8ae61e105517763f21c707ea3ec
+P
+ BiddingKG/maxcompute/cycleRec.py,b\d\bdbd92638e7f5983e655c67b07bb464d62021b36
+P
+ BiddingKG/dl/bidway/re_bidway.py,4\b\4bbee1b8e2177ffd4dbcc10e26686b81b38db517
+G
+BiddingKG/dl/test/12.py,5\c\5c99d16b0fcfaac86fa00d720a060d38778939c6
+J
+BiddingKG/dl/form/train.py,9\0\9092cdc516e6529e04ba5c9e245978a9778f1457
+U
+%BiddingKG/maxcompute/extract_check.py,c\b\cb469c6b5dccfb880cb12b739ba36dd30aa17830
+Q
+!BiddingKG/dl/LEGAL_ENTERPRISE.txt,6\8\685bd49ae2f5f0de419c93a217a0e57564d705ab
+U
+%BiddingKG/maxcompute/documentMerge.py,0\2\0281d029b2e1edefc27911179172143779deed49
+Q
+!BiddingKG/maxcompute/evaluates.py,4\b\4bf6acd495095b59143b1620d2f455e5e651a071
+L
+BiddingKG/dl/common/Utils.py,f\4\f4c35e30342829a2fc89108259e28edc0a425cce

+ 23 - 0
BiddingKG/app.json

@@ -0,0 +1,23 @@
+{
+  "generate_token": "true",
+  "metadata": {
+    "cpu": 5,
+    "instance": 5,
+    "memory": 20000,
+    "region": "cn-hangzhou",
+    "resource": "EAS-SrhheASr495",
+    "rpc": {
+      "batching": "true",
+      "keepalive": 60000,
+      "max_batch_size": 40
+    }
+  },
+  "workers":7,
+  "name": "content_extract",
+  "processor_entry": "./app.py",
+  "processor_path": "oss://eas-model-hangzhou/1255640119316927/BiddingKG_eas.zip",
+  "data_image": "registry-vpc.cn-hangzhou.aliyuncs.com/bxkc/eas-service:latest",
+  "processor_type": "python",
+  "token": "NzZmOWZlMmU2MGY3YmQ4MDBjM2E5MDAyZjhjNjQ0MzZlMmE0NTMwZg==",
+  "workers": 7
+}

+ 1 - 1
BiddingKG/dl/entityLink/entityLink.py

@@ -80,7 +80,7 @@ def link_entitys(list_entitys,on_value=0.8):
 
 
 def getEnterprisePath():
-    filename = "../LEGAL_ENTERPRISE.txt"
+    filename = "LEGAL_ENTERPRISE.txt"
     real_path = getFileFromSysPath(filename)
     if real_path is None:
         real_path = filename

+ 14 - 8
BiddingKG/dl/interface/extract.py

@@ -52,6 +52,12 @@ def predict(doc_id,text,title="",page_time="",**kwargs):
     cost_time["preprocess"] = round(time.time()-start_time,2)
     cost_time.update(_cost_time)
 
+    for list_entity in list_entitys:
+        for _entity in list_entity:
+            log("type:%s,text:%s,label:%s,values:%s,sentence:%s,begin_index:%s,end_index:%s"%
+                (str(_entity.entity_type),str(_entity.entity_text),str(_entity.label),str(_entity.values),str(_entity.sentence_index),
+                 str(_entity.begin_index),str(_entity.end_index)))
+
     #依赖句子顺序
     start_time = time.time() # 公告类型/生命周期提取
     list_channel_dic = predictor.getPredictor("channel").predict(title=title, content=list_sentences[0])
@@ -142,14 +148,14 @@ def predict(doc_id,text,title="",page_time="",**kwargs):
     data_res["cost_time"] = cost_time
     data_res["success"] = True
 
-    # for _article in list_articles:
-    #     log(_article.content)
-    #
-    # for list_entity in list_entitys:
-    #     for _entity in list_entity:
-    #         log("type:%s,text:%s,label:%s,values:%s,sentence:%s,begin_index:%s,end_index:%s"%
-    #               (str(_entity.entity_type),str(_entity.entity_text),str(_entity.label),str(_entity.values),str(_entity.sentence_index),
-    #                str(_entity.begin_index),str(_entity.end_index)))
+    for _article in list_articles:
+        log(_article.content)
+
+    for list_entity in list_entitys:
+        for _entity in list_entity:
+            log("type:%s,text:%s,label:%s,values:%s,sentence:%s,begin_index:%s,end_index:%s"%
+                  (str(_entity.entity_type),str(_entity.entity_text),str(_entity.label),str(_entity.values),str(_entity.sentence_index),
+                   str(_entity.begin_index),str(_entity.end_index)))
 
     return json.dumps(data_res,cls=MyEncoder,sort_keys=True,indent=4,ensure_ascii=False)