pycityagent 1.1.0__tar.gz → 1.1.2__tar.gz

{pycityagent-1.1.0 → pycityagent-1.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycityagent
-Version: 1.1.0
+Version: 1.1.2
 Summary: LLM-based城市模拟器agent构建库
 Author-email: Yuwei Yan <pinkgranite86@gmail.com>
 License: MIT License
@@ -60,7 +60,7 @@ Requires-Dist: transitions>=0.9.0
 - ![framwork](./static/framework.png)
 
 ### Workflow of CityAgent
-- ![workflow](./static/workflow.png)
+- ![workflow](./static/workflow_1.png)
 
 ## Hands On - By An Easy Demo
 ### Apply for your App
@@ -74,9 +74,10 @@ Requires-Dist: transitions>=0.9.0
 ```yaml
 llm_request:
   text_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
     model: xxx
+    http_client: xxx (if you use opanai and want to use your own backend LLM model)
   img_understand_request:
     request_type: qwen
     api_key: xxx
@@ -111,7 +112,8 @@ apphub_request:
 - As you can see, the whole CityAgent is based on the LLM, by now, there are three different parts of config items: **text_request**, **img_understand_request** and **img_generate_request**
 - By now, we support [**qwen**](https://tongyi.aliyun.com/) and [**openai**](https://openai.com/)
     - `Notice: Our environments are basically conducted with qwen. If you prefer to use openai, then you may encounter hardships. AND fell free to issue us.`
-- Get your **api_key** and chooce your **model**s
+- Get your **api_key** and chooce your **model**
+- If you want to use your backend models, set the **http_client** (only available when using **openai**)
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**

{pycityagent-1.1.0 → pycityagent-1.1.2}/README.md

@@ -13,7 +13,7 @@
 - ![framwork](./static/framework.png)
 
 ### Workflow of CityAgent
-- ![workflow](./static/workflow.png)
+- ![workflow](./static/workflow_1.png)
 
 ## Hands On - By An Easy Demo
 ### Apply for your App
@@ -27,9 +27,10 @@
 ```yaml
 llm_request:
   text_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
     model: xxx
+    http_client: xxx (if you use opanai and want to use your own backend LLM model)
   img_understand_request:
     request_type: qwen
     api_key: xxx
@@ -64,7 +65,8 @@ apphub_request:
 - As you can see, the whole CityAgent is based on the LLM, by now, there are three different parts of config items: **text_request**, **img_understand_request** and **img_generate_request**
 - By now, we support [**qwen**](https://tongyi.aliyun.com/) and [**openai**](https://openai.com/)
     - `Notice: Our environments are basically conducted with qwen. If you prefer to use openai, then you may encounter hardships. AND fell free to issue us.`
-- Get your **api_key** and chooce your **model**s
+- Get your **api_key** and chooce your **model**
+- If you want to use your backend models, set the **http_client** (only available when using **openai**)
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**

{pycityagent-1.1.0 → pycityagent-1.1.2}/pycityagent/brain/sence.py

@@ -297,7 +297,8 @@ class Sence(BrainFunction):
         - List[dict]: 可达位置列表
             - lane_id (int)
             - s (float)
-            - longlat (Tuple[float, float]): [longitude, latitude]
+            - xy (Tuple[float, float]): (x, y)
+            - longlat (Tuple[float, float]): (longitude, latitude)
             - type (str): 'driving' / 'walking' / 'unspecified'
         '''
         radius_ = self._sence_radius
@@ -318,6 +319,7 @@ class Sence(BrainFunction):
                 positions.append({
                     'lane_id': driving_positions[i]['lane_id'], 
                     's': driving_positions[i]['s'],
+                    'xy': (driving_gates[i]['x'], driving_gates[i]['y']),
                     'longlat': longlat,
                     'type': 'driving'
                 })
@@ -326,6 +328,7 @@ class Sence(BrainFunction):
                 positions.append({
                     'lane_id': walking_positions[i]['lane_id'], 
                     's': walking_positions[i]['s'],
+                    'xy': (walking_gates[i]['x'], walking_gates[i]['y']),
                     'longlat': longlat,
                     'type': 'walking'
                 })
@@ -343,7 +346,11 @@ class Sence(BrainFunction):
                 x, y = get_xy_in_lane(nodes, tmp_s)
                 longlat = self._agent._simulator.map.xy2lnglat(x=x, y=y)
                 type = copy.deepcopy(self._lane_type_mapping.get(lane['type'], 'unspecified'))
-                positions += [{'lane_id': lane_id, 's': tmp_s, 'longlat': longlat, 'type': type}]
+                positions += [{'lane_id': lane_id, 
+                               's': tmp_s, 
+                               'xy': (x, y),
+                               'longlat': longlat, 
+                               'type': type}]
 
                 # 2. 前驱道路
                 pre_lanes = lane['predecessors']
@@ -356,7 +363,11 @@ class Sence(BrainFunction):
                     x, y = get_xy_in_lane(pre_lane_nodes, tmp_s, 'back')
                     longlat = self._agent._simulator.map.xy2lnglat(x=x, y=y)
                     type = self._lane_type_mapping.get(pre_lane_['type'], 'unspecified')
-                    positions += [{'lane_id': pre_lane_id, 's': tmp_s, 'longlat': longlat, 'type': type}]
+                    positions += [{'lane_id': pre_lane_id, 
+                                   's': tmp_s, 
+                                   'xy': (x, y),
+                                   'longlat': longlat, 
+                                   'type': type}]
             elif agent_s == lane['length']:
                 # 处于当前道路的尾部端点位置
                 # 1. 当前道路
@@ -365,7 +376,11 @@ class Sence(BrainFunction):
                 x, y = get_xy_in_lane(nodes, tmp_s, 'back')
                 longlat = self._agent._simulator.map.xy2loglat(x=x, y=y)
                 type = self._lane_type_mapping.get(lane['type'], 'unspecified')
-                positions += [{'lane_id': lane_id, 's': tmp_s, 'longlat': longlat, 'type': type}]
+                positions += [{'lane_id': lane_id, 
+                               's': tmp_s, 
+                               'xy': (x, y),
+                               'longlat': longlat, 
+                               'type': type}]
 
                 # 2. 后继道路
                 suc_lanes = lane['successors']
@@ -378,7 +393,11 @@ class Sence(BrainFunction):
                     x, y = get_xy_in_lane(suc_lane_nodes, tmp_s)
                     longlat = self._agent._simulator.map.xy2loglat(x=x, y=y)
                     type = self._lane_type_mapping.get(lane['type'], 'unspecified')
-                    positions += [{'lane_id': suc_lane_id, 's': tmp_s, 'longlat': longlat, 'type': type}]
+                    positions += [{'lane_id': suc_lane_id, 
+                                   's': tmp_s, 
+                                   'xy': (x, y),
+                                   'longlat': longlat, 
+                                   'type': type}]
             else:
                 # 非端点位置
                 neg_s = agent_s - radius_
@@ -386,14 +405,22 @@ class Sence(BrainFunction):
                 x, y = get_xy_in_lane(nodes, neg_s, 'back')
                 longlat = self._agent._simulator.map.xy2loglat(x=x, y=y)
                 type = self._lane_type_mapping.get(lane['type'], 'unspecified')
-                positions += [{'lans_id': lane_id, 's': neg_s, 'longlat': longlat, 'type': type}]
+                positions += [{'lans_id': lane_id, 
+                               's': neg_s, 
+                               'xy': (x, y),
+                               'longlat': longlat, 
+                               'type': type}]
 
                 pos_s = agent_s + radius_
                 pos_s = pos_s if pos_s <= lane['length'] else lane['length']
                 x, y = get_xy_in_lane(nodes, pos_s)
                 longlat = self._agent._simulator.map.xy2loglat(x=x, y=y)
                 type = self._lane_type_mapping.get(lane['type'], 'unspecified')
-                positions += [{'lans_id': lane_id, 's': neg_s, 'longlat': longlat, 'type': type}]
+                positions += [{'lans_id': lane_id, 
+                               's': neg_s, 
+                               'xy': (x, y),
+                               'longlat': longlat, 
+                               'type': type}]
         return positions
 
     async def PerceivePoi(self, radius:int=None, category:str=None):

{pycityagent-1.1.0 → pycityagent-1.1.2}/pycityagent/urbanllm/urbanllm.py

@@ -41,10 +41,12 @@ class UrbanLLM:
         Returns:
         - (str): the response content
         """
+        client = None if self.config.text['http_client'] == None else self.config.text['http_client']
         if self.config.text['request_type'] == 'openai':
             client = OpenAI(
                 api_key=self.config.text['api_key'], 
-                base_url=self.config.text['api_base']
+                base_url=self.config.text['api_base'],
+                http_client=client
             )
             response = client.chat.completions.create(
                 model=self.config.text['model'],
@@ -53,7 +55,6 @@ class UrbanLLM:
             )
             return response.choices[0].message.content
         elif self.config.text['request_type'] == 'qwen':
-
             response = dashscope.Generation.call(
                 model=self.config.text['model'],
                 api_key=self.config.text['api_key'],
@@ -82,7 +83,7 @@ class UrbanLLM:
         """
         ppt = "如何理解这幅图像？"
         if prompt != None:
-            ppt += prompt
+            ppt = prompt
         dialog = [{
             'role': 'user',
             'content': [
@@ -101,7 +102,7 @@ class UrbanLLM:
             print(response.code)  # The error code.
             return "Error"
 
-    def img_generate(self, prompt, size:str='512*512', quantity:int = 1):
+    def img_generate(self, prompt:str, size:str='512*512', quantity:int = 1):
         """
         图像生成
         Image generation

{pycityagent-1.1.0 → pycityagent-1.1.2}/pycityagent.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycityagent
-Version: 1.1.0
+Version: 1.1.2
 Summary: LLM-based城市模拟器agent构建库
 Author-email: Yuwei Yan <pinkgranite86@gmail.com>
 License: MIT License
@@ -60,7 +60,7 @@ Requires-Dist: transitions>=0.9.0
 - ![framwork](./static/framework.png)
 
 ### Workflow of CityAgent
-- ![workflow](./static/workflow.png)
+- ![workflow](./static/workflow_1.png)
 
 ## Hands On - By An Easy Demo
 ### Apply for your App
@@ -74,9 +74,10 @@ Requires-Dist: transitions>=0.9.0
 ```yaml
 llm_request:
   text_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
     model: xxx
+    http_client: xxx (if you use opanai and want to use your own backend LLM model)
   img_understand_request:
     request_type: qwen
     api_key: xxx
@@ -111,7 +112,8 @@ apphub_request:
 - As you can see, the whole CityAgent is based on the LLM, by now, there are three different parts of config items: **text_request**, **img_understand_request** and **img_generate_request**
 - By now, we support [**qwen**](https://tongyi.aliyun.com/) and [**openai**](https://openai.com/)
     - `Notice: Our environments are basically conducted with qwen. If you prefer to use openai, then you may encounter hardships. AND fell free to issue us.`
-- Get your **api_key** and chooce your **model**s
+- Get your **api_key** and chooce your **model**
+- If you want to use your backend models, set the **http_client** (only available when using **openai**)
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**