layout refine (#115)

README.md

@@ -18,7 +18,7 @@ If **vm.max_map_count** is not larger  than 65535, please run the following comm
 121:/ragflow# sudo sysctl -w vm.max_map_count=262144
 ```
 However, this change is not persistent and will be reset after a system reboot. 
-To make the change permanent, you need to update the **/etc/sysctl.conf file**.
+To make the change permanent, you need to update the **/etc/sysctl.conf**.
 Add or update the following line in the file:
 ```bash
 vm.max_map_count=262144

api/apps/llm_app.py

@@ -82,8 +82,19 @@ def set_api_key():
 @login_required
 def my_llms():
     try:
-        objs = TenantLLMService.get_my_llms(current_user.id)
-        return get_json_result(data=objs)
+        res = {}
+        for o in TenantLLMService.get_my_llms(current_user.id):
+            if o["llm_factory"] not in res:
+                res[o["llm_factory"]] = {
+                    "tags": o["tags"],
+                    "llm": []
+                }
+            res[o["llm_factory"]]["llm"].append({
+                "type": o["model_type"],
+                "name": o["model_name"],
+                "used_token": o["used_tokens"]
+            })
+        return get_json_result(data=res)
     except Exception as e:
         return server_error_response(e)
 

api/db/services/llm_service.py

@@ -49,7 +49,9 @@ def get_my_llms(cls, tenant_id):
             LLMFactories.logo,
             LLMFactories.tags,
             cls.model.model_type,
-            cls.model.llm_name]
+            cls.model.llm_name,
+            cls.model.used_tokens
+        ]
         objs = cls.model.select(*fields).join(LLMFactories, on=(cls.model.llm_factory == LLMFactories.name)).where(
             cls.model.tenant_id == tenant_id).dicts()
 

deepdoc/vision/layout_recognizer.py

@@ -100,7 +100,7 @@ def findLayout(ty):
                     i += 1
 
             for lt in ["footer", "header", "reference", "figure caption",
-                       "table caption", "title", "text", "table", "figure", "equation"]:
+                       "table caption", "title", "table", "text", "figure", "equation"]:
                 findLayout(lt)
 
             # add box to figure layouts which has not text box

rag/app/naive.py

@@ -101,7 +101,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000, lang="Chinese", ca
         d = copy.deepcopy(doc)
         if pdf_parser:
             d["image"], poss = pdf_parser.crop(ck, need_position=True)
-            add_positions(d, poss, from_page)
+            add_positions(d, poss)
             ck = pdf_parser.remove_tag(ck)
         tokenize(d, ck, eng)
         res.append(d)