pycityagent 1.1.8__tar.gz → 1.1.10__tar.gz

{pycityagent-1.1.8 → pycityagent-1.1.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycityagent
-Version: 1.1.8
+Version: 1.1.10
 Summary: LLM-based城市模拟器agent构建库
 Author-email: Yuwei Yan <pinkgranite86@gmail.com>
 License: MIT License
@@ -79,9 +79,10 @@ llm_request:
     model: xxx
     (api_base): xxx (this is an optional config, if you use opanai and want to use your own backend LLM model, default to "https://api.openai.com/v1")
   img_understand_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
-    model: xxx
+    model: xxx ('gpt-4-turbo' if you use openai)
+    (api_base): same as text_request
   img_generate_request:
     request_type: qwen
     api_key: xxx
@@ -110,11 +111,19 @@ apphub_request:
 
 #### LLM_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**
-- If you want to use your backend models, set the **api_base** (only available when using **openai**)
-  - default value: "https://api.openai.com/v1"
+- **text_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**
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+    - default value: "https://api.openai.com/v1"
+- **img_understand_request**
+  - By now, we support **qwen** and **openai**
+  - If choose **openai**, then the **model** has to be '**gpt-4-turbo**'
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+      - default value: "https://api.openai.com/v1"
+- **img_generate_request**
+  - By now, only [**qwen**] is supported
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**

{pycityagent-1.1.8 → pycityagent-1.1.10}/README.md

@@ -32,9 +32,10 @@ llm_request:
     model: xxx
     (api_base): xxx (this is an optional config, if you use opanai and want to use your own backend LLM model, default to "https://api.openai.com/v1")
   img_understand_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
-    model: xxx
+    model: xxx ('gpt-4-turbo' if you use openai)
+    (api_base): same as text_request
   img_generate_request:
     request_type: qwen
     api_key: xxx
@@ -63,11 +64,19 @@ apphub_request:
 
 #### LLM_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**
-- If you want to use your backend models, set the **api_base** (only available when using **openai**)
-  - default value: "https://api.openai.com/v1"
+- **text_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**
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+    - default value: "https://api.openai.com/v1"
+- **img_understand_request**
+  - By now, we support **qwen** and **openai**
+  - If choose **openai**, then the **model** has to be '**gpt-4-turbo**'
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+      - default value: "https://api.openai.com/v1"
+- **img_generate_request**
+  - By now, only [**qwen**] is supported
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent/ac/__init__.py

@@ -5,4 +5,4 @@ from .action import *
 from .hub_actions import *
 from .sim_actions import *
 
-__all__ = [ActionController, Action, HubAction, SimAction, SendUserMessage, SendStreetview, SendPop, PositionUpdate, ShowPath, ShowPosition, SetSchedule, SendAgentMessage]
+__all__ = [ActionController, Action, HubAction, SimAction, SendUserMessage, SendStreetview, SendPop, ShowPath, ShowPosition, SetSchedule, SendAgentMessage]

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent/agent_func.py

@@ -1,10 +1,12 @@
 """FuncAgent: 功能性智能体及其定义"""
 
+from typing import Union
 from pycityagent.urbanllm import UrbanLLM
 from .urbanllm import UrbanLLM
 from .agent import Agent, AgentType
 from .image.image import Image
 from .ac.hub_actions import PositionUpdate
+from .utils import *
 
 class FuncAgent(Agent):
     """
@@ -63,23 +65,96 @@ class FuncAgent(Agent):
                 - x (double)
                 - y (double)
                 - z (double)
-        - direction (double): 方向角
+        - direction (double): 朝向-方向角
         """
 
-    async def init_position_aoi(self, aoi_id:int):
+        self._posUpdate = PositionUpdate(self)
+
+    async def set_position_aoi(self, aoi_id:int):
         """
-        - 将agent的位置初始化到指定aoi
-        - 根据指定aoi设置aoi_position, longlat_position以及xy_position
+        - 将agent的位置设定到指定aoi
+
+        Args:
+        - aoi_id (int): AOI id
         """
         if aoi_id in self._simulator.map.aois:
             aoi = self._simulator.map.aois[aoi_id]
+            self.motion['position'] = {}
             self.motion['position']['aoi_position'] = {'aoi_id': aoi_id}
             self.motion['position']['longlat_position'] = {'longitude': aoi['shapely_lnglat'].centroid.coords[0][0], 'latitude': aoi['shapely_lnglat'].centroid.coords[0][1]}
             x, y = self._simulator.map.lnglat2xy(lng=self.motion['position']['longlat_position']['longitude'], 
                                                  lat=self.motion['position']['longlat_position']['latitude'])
             self.motion['position']['xy_position'] = {'x': x, 'y': y}
-        pos = PositionUpdate(self)
-        await pos.Forward(longlat=[self.motion['position']['longlat_position']['longitude'], self.motion['position']['longlat_position']['latitude']])
+            await self._posUpdate.Forward(longlat=[self.motion['position']['longlat_position']['longitude'], self.motion['position']['longlat_position']['latitude']])
+        else:
+            print("Error: wrong aoi id")
+
+    async def set_position_lane(self, lane_id:int, s:float=0.0, direction:Union[float, str]='front'):
+        """
+        - 将agent的位置设定到指定lane
+        
+        Args:
+        - lane_id (int): Lane id
+        - s (float): distance from the start point of the lane, default None, if None, then set to the start point
+        - direction (Union[float, str]): agent方向角, 默认值为'front'
+            - float: 直接设置为给定方向角(atan2计算得到)
+            - str: 可选项['front', 'back'], 指定agent的行走方向
+                - 对于driving lane而言, 只能朝一个方向通行, 只能是'front'
+                - 对于walking lane而言, 可以朝两个方向通行, 可以是'front'或'back'
+        """
+        if lane_id in self._simulator.map.lanes:
+            lane = self._simulator.map.lanes[lane_id]
+            if s > lane['length']:
+                print("Error: 's' too large")
+            self.motion['position'] = {}
+            self.motion['position']['lane_position'] = {'lane_id': lane_id, 's': s}
+            nodes = lane['center_line']['nodes']
+            x, y = get_xy_in_lane(nodes, s)
+            longlat = self._simulator.map.xy2lnglat(x=x, y=y)
+            self.motion['position']['longlat_position'] = {
+                'longitude': longlat[0],
+                'latitude': longlat[1]
+            }
+            self.motion['position']['xy_position'] = {
+                'x': x,
+                'y': y
+            }
+            if isinstance(direction, float):
+                self.motion['direction'] = direction
+            else:
+                # 计算方向角
+                direction_ = get_direction_by_s(nodes, s, direction)
+                self.motion['direction'] = direction_                    
+            await self._posUpdate.Forward(longlat=[self.motion['position']['longlat_position']['longitude'], self.motion['position']['longlat_position']['latitude']])
+        else:
+            print("Error: wrong lane id")
+
+    async def set_position_poi(self, poi_id:int):
+        """
+        - 将agent的位置设定到指定poi
+
+        Args:
+        - poi_id (int): Poi id
+        """
+        if poi_id in self._simulator.map.pois:
+            poi = self._simulator.map.pois[poi_id]
+            x = poi['position']['x']
+            y = poi['position']['y']
+            longlat = self._simulator.map.xy2lnglat(x=x, y=y)
+            aoi_id = poi['aoi_id']
+            self.motion['position'] = {}
+            self.motion['position']['aoi_position'] = {'aoi_id': aoi_id}
+            self.motion['position']['longlat_position'] = {
+                'longitude': longlat[0],
+                'latitude': longlat[1]
+            }
+            self.motion['position']['xy_position'] = {
+                'x': x,
+                'y': y
+            }
+            await self._posUpdate.Forward(longlat=[self.motion['position']['longlat_position']['longitude'], self.motion['position']['longlat_position']['latitude']])
+        else:
+            print("Error: wrong poi id")
 
     def Bind(self):
         """

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent/brain/sence.py

@@ -13,83 +13,7 @@ from citystreetview import (
 )
 from .brainfc import BrainFunction
 from .static import POI_TYPE_DICT, LEVEL_ONE_PRE
-
-def point_on_line_given_distance(start_node, end_node, distance):
-    """
-    Given two points (start_point and end_point) defining a line, and a distance s to travel along the line,
-    return the coordinates of the point reached after traveling s units along the line, starting from start_point.
-
-    Args:
-        start_point (tuple): Tuple of (x, y) representing the starting point on the line.
-        end_point (tuple): Tuple of (x, y) representing the ending point on the line.
-        distance (float): Distance to travel along the line, starting from start_point.
-
-    Returns:
-        tuple: Tuple of (x, y) representing the new point reached after traveling s units along the line.
-    """
-
-    x1, y1 = start_node['x'], start_node['y']
-    x2, y2 = end_node['x'], end_node['y']
-
-    # Calculate the slope m and the y-intercept b of the line
-    if x1 == x2:
-        # Vertical line, distance is only along the y-axis
-        return (x1, y1 + distance if distance >= 0 else y1 - abs(distance))
-    else:
-        m = (y2 - y1) / (x2 - x1)
-        b = y1 - m * x1
-
-        # Calculate the direction vector (dx, dy) along the line
-        dx = (x2 - x1) / math.sqrt((x2 - x1)**2 + (y2 - y1)**2)
-        dy = (y2 - y1) / math.sqrt((x2 - x1)**2 + (y2 - y1)**2)
-
-        # Scale the direction vector by the given distance
-        scaled_dx = dx * distance
-        scaled_dy = dy * distance
-
-        # Calculate the new point's coordinates
-        x = x1 + scaled_dx
-        y = y1 + scaled_dy
-
-        return [x, y]
-
-def get_xy_in_lane(nodes, distance, direction:str='front'):
-    temp_sum = 0
-    remain_s = 0
-    if direction == 'front':
-        # 顺道路方向前进
-        if distance == 0:
-            return [nodes[0]['x'], nodes[0]['y']]
-        key_index = 0  # first node
-        for i in range(1, len(nodes)):
-            x1, y1 = nodes[i-1]['x'], nodes[i-1]['y']
-            x2, y2 = nodes[i]['x'], nodes[i]['y']
-            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
-            if temp_sum > distance:
-                remain_s = distance - (temp_sum - math.sqrt((x2 - x1)**2 + (y2-y1)**2))
-                break;
-            key_index += 1
-        if remain_s < 0.5:
-            return [nodes[key_index]['x'], nodes[key_index]['y']]
-        longlat = point_on_line_given_distance(nodes[key_index], nodes[key_index+1], remain_s)
-        return longlat
-    else:
-        # 逆道路方向前进
-        if distance == 0:
-            return [nodes[-1]['x'], nodes[-1]['y']]
-        key_index = len(nodes)-1  # last node
-        for i in range(len(nodes)-1, 0, -1):
-            x1, y1 = nodes[i]['x'], nodes[i]['y']
-            x2, y2 = nodes[i-1]['x'], nodes[i-1]['y']
-            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
-            if temp_sum > distance:
-                remain_s = distance - (temp_sum - math.sqrt((x2 - x1)**2 + (y2-y1)**2))
-                break;
-            key_index -= 1
-        if remain_s < 0.5:
-            return [nodes[key_index]['x'], nodes[key_index]['y']]
-        longlat = point_on_line_given_distance(nodes[key_index], nodes[key_index-1], remain_s)
-        return longlat
+from ..utils import point_on_line_given_distance, get_xy_in_lane
 
 class SencePlug:
     """
@@ -146,10 +70,10 @@ class Sence(BrainFunction):
         SencePlug Buffer: used to store those sence plug content
         """
 
-        self.enable_streeview = False
+        self.enable_streeview = True
         """
-        街景感知功能接口, 默认为False
-        Interface of streetview function, defualt: False
+        街景感知功能接口, 默认为True
+        Interface of streetview function, defualt: True
         """
 
         self._lane_type_mapping = {1: 'driving', 2: 'walking'}

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent/urbanllm/urbanllm.py

@@ -3,12 +3,16 @@
 from openai import OpenAI
 from http import HTTPStatus
 import dashscope
-from urllib.parse import urlparse, unquote
-from pathlib import PurePosixPath
 import requests
 from dashscope import ImageSynthesis
 from PIL import Image
 from io import BytesIO
+from typing import Union
+import base64
+
+def encode_image(image_path):
+  with open(image_path, "rb") as image_file:
+    return base64.b64encode(image_file.read()).decode('utf-8')
 
 class LLMConfig:
     """
@@ -79,19 +83,19 @@ class UrbanLLM:
             if response.status_code == HTTPStatus.OK:
                 return response.output.choices[0]['message']['content']
             else:
-                return "Error: {}".format(response.status_code)
+                return "Error: {}, {}".format(response.status_code, response.message)
         else:
             print("ERROR: Wrong Config")
             return "wrong config"
 
-    def img_understand(self, img_path:str, prompt:str=None) -> str:
+    def img_understand(self, img_path:Union[str, list[str]], prompt:str=None) -> str:
         """
         图像理解
         Image understanding
 
         Args:
-        - img_path: 目标图像的路径. The path of selected Image
-        - prompt: 理解提示词 - 例如理解方向. The understanding prompts
+        - img_path (Union[str, list[str]]): 目标图像的路径, 既可以是一个路径也可以是包含多张图片路径的list. The path of selected Image
+        - prompt (str): 理解提示词 - 例如理解方向. The understanding prompts
 
         Returns:
         - (str): the understanding content
@@ -99,22 +103,70 @@ class UrbanLLM:
         ppt = "如何理解这幅图像？"
         if prompt != None:
             ppt = prompt
-        dialog = [{
-            'role': 'user',
-            'content': [
-                {'image': 'file://' + img_path},
-                {'text': ppt}
-            ]
-        }]
-        response = dashscope.MultiModalConversation.call(
-                    model=self.config.image_u['model'],
-                    api_key=self.config.image_u['api_key'],
-                    messages=dialog
-                )
-        if response.status_code == HTTPStatus.OK:
-            return response.output.choices[0]['message']['content']
+        if self.config.image_u['request_type'] == 'openai':
+            if 'api_base' in self.config.image_u.keys():
+                api_base = self.config.image_u['api_base']
+            else:
+                api_base = None
+            client = OpenAI(
+                api_key=self.config.text['api_key'], 
+                base_url=api_base,
+            )
+            content = []
+            content.append({'type': 'text', 'text': ppt})
+            if isinstance(img_path, str):
+                base64_image = encode_image(img_path)
+                content.append({
+                    'type': 'image_url', 
+                    'image_url': {
+                            'url': f"data:image/jpeg;base64,{base64_image}"
+                        }
+                })
+            elif isinstance(img_path, list) and all(isinstance(item, str) for item in img_path):
+                for item in img_path:
+                    base64_image = encode_image(item)
+                    content.append({
+                        'type': 'image_url',
+                        'image_url': {
+                            'url': f"data:image/jpeg;base64,{base64_image}"
+                        }
+                    })
+            response = client.chat.completions.create(
+                model=self.config.image_u['model'],
+                messages=[{
+                    'role': 'user',
+                    'content': content
+                }]
+            )
+            return response.choices[0].message.content
+        elif self.config.image_u['request_type'] == 'qwen':
+            content = []
+            if isinstance(img_path, str):
+                content.append({'image': 'file://' + img_path})
+                content.append({'text': ppt})
+            elif isinstance(img_path, list) and all(isinstance(item, str) for item in img_path):
+                for item in img_path:
+                    content.append({
+                        'image': 'file://' + item
+                    })
+                content.append({'text': ppt})
+
+            dialog = [{
+                'role': 'user',
+                'content': content
+            }]
+            response = dashscope.MultiModalConversation.call(
+                        model=self.config.image_u['model'],
+                        api_key=self.config.image_u['api_key'],
+                        messages=dialog
+                    )
+            if response.status_code == HTTPStatus.OK:
+                return response.output.choices[0]['message']['content']
+            else:
+                print(response.code)  # The error code.
+                return "Error"
         else:
-            print(response.code)  # The error code.
+            print("ERROR: wrong image understanding type, only 'openai' and 'openai' is available")
             return "Error"
 
     def img_generate(self, prompt:str, size:str='512*512', quantity:int = 1):

pycityagent-1.1.10/pycityagent/utils.py

@@ -0,0 +1,148 @@
+import math
+
+def get_angle(x, y):
+    return math.atan2(y, x)*180/math.pi
+
+def point_on_line_given_distance(start_node, end_node, distance):
+    """
+    Given two points (start_point and end_point) defining a line, and a distance s to travel along the line,
+    return the coordinates of the point reached after traveling s units along the line, starting from start_point.
+
+    Args:
+        start_point (tuple): Tuple of (x, y) representing the starting point on the line.
+        end_point (tuple): Tuple of (x, y) representing the ending point on the line.
+        distance (float): Distance to travel along the line, starting from start_point.
+
+    Returns:
+        tuple: Tuple of (x, y) representing the new point reached after traveling s units along the line.
+    """
+
+    x1, y1 = start_node['x'], start_node['y']
+    x2, y2 = end_node['x'], end_node['y']
+
+    # Calculate the slope m and the y-intercept b of the line
+    if x1 == x2:
+        # Vertical line, distance is only along the y-axis
+        return (x1, y1 + distance if distance >= 0 else y1 - abs(distance))
+    else:
+        m = (y2 - y1) / (x2 - x1)
+        b = y1 - m * x1
+
+        # Calculate the direction vector (dx, dy) along the line
+        dx = (x2 - x1) / math.sqrt((x2 - x1)**2 + (y2 - y1)**2)
+        dy = (y2 - y1) / math.sqrt((x2 - x1)**2 + (y2 - y1)**2)
+
+        # Scale the direction vector by the given distance
+        scaled_dx = dx * distance
+        scaled_dy = dy * distance
+
+        # Calculate the new point's coordinates
+        x = x1 + scaled_dx
+        y = y1 + scaled_dy
+
+        return [x, y]
+
+def get_xy_in_lane(nodes, distance, direction:str='front'):
+    temp_sum = 0
+    remain_s = 0
+    if direction == 'front':
+        # 顺道路方向前进
+        if distance == 0:
+            return [nodes[0]['x'], nodes[0]['y']]
+        key_index = 0  # first node
+        for i in range(1, len(nodes)):
+            x1, y1 = nodes[i-1]['x'], nodes[i-1]['y']
+            x2, y2 = nodes[i]['x'], nodes[i]['y']
+            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
+            if temp_sum > distance:
+                remain_s = distance - (temp_sum - math.sqrt((x2 - x1)**2 + (y2-y1)**2))
+                break;
+            key_index += 1
+        if remain_s < 0.5:
+            return [nodes[key_index]['x'], nodes[key_index]['y']]
+        longlat = point_on_line_given_distance(nodes[key_index], nodes[key_index+1], remain_s)
+        return longlat
+    else:
+        # 逆道路方向前进
+        if distance == 0:
+            return [nodes[-1]['x'], nodes[-1]['y']]
+        key_index = len(nodes)-1  # last node
+        for i in range(len(nodes)-1, 0, -1):
+            x1, y1 = nodes[i]['x'], nodes[i]['y']
+            x2, y2 = nodes[i-1]['x'], nodes[i-1]['y']
+            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
+            if temp_sum > distance:
+                remain_s = distance - (temp_sum - math.sqrt((x2 - x1)**2 + (y2-y1)**2))
+                break;
+            key_index -= 1
+        if remain_s < 0.5:
+            return [nodes[key_index]['x'], nodes[key_index]['y']]
+        longlat = point_on_line_given_distance(nodes[key_index], nodes[key_index-1], remain_s)
+        return longlat
+    
+def get_direction_by_s(nodes, distance, direction:str='front'):
+    temp_sum = 0
+    if direction == 'front':
+        # 顺道路方向前进
+        if distance == 0:
+            return [nodes[0]['x'], nodes[0]['y']]
+        key_index = 0  # first node
+        for i in range(1, len(nodes)):
+            x1, y1 = nodes[i-1]['x'], nodes[i-1]['y']
+            x2, y2 = nodes[i]['x'], nodes[i]['y']
+            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
+            if temp_sum > distance:
+                break;
+            key_index += 1
+        if key_index == len(nodes)-1:
+            # 端点
+            x = nodes[key_index]['x']-nodes[key_index-1]['x']
+            y = nodes[key_index]['y']-nodes[key_index-1]['y']
+            return get_angle(x, y)
+        else:
+            # 中间点
+            x = nodes[key_index+1]['x'] - nodes[key_index]['x']
+            y = nodes[key_index+1]['y'] - nodes[key_index]['y']
+            return get_angle(x, y)
+    elif direction == 'back':
+        # 逆道路方向前进
+        if distance == 0:
+            return [nodes[-1]['x'], nodes[-1]['y']]
+        key_index = len(nodes)-1  # last node
+        for i in range(len(nodes)-1, 0, -1):
+            x1, y1 = nodes[i]['x'], nodes[i]['y']
+            x2, y2 = nodes[i-1]['x'], nodes[i-1]['y']
+            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
+            if temp_sum > distance:
+                break;
+            key_index -= 1
+        if key_index == 0:
+            x = nodes[key_index]['x'] - nodes[key_index+1]['x']
+            y = nodes[key_index]['y'] - nodes[key_index+1]['y']
+            return get_angle(x, y)
+        else:
+            x = nodes[key_index-1]['x'] - nodes[key_index]['x']
+            y = nodes[key_index-1]['y'] - nodes[key_index]['y']
+            return get_angle(x, y)
+    else:
+        print("Warning: wroing direction, 'front' instead")
+        if distance == 0:
+            return [nodes[0]['x'], nodes[0]['y']]
+        key_index = 0  # first node
+        for i in range(1, len(nodes)):
+            x1, y1 = nodes[i-1]['x'], nodes[i-1]['y']
+            x2, y2 = nodes[i]['x'], nodes[i]['y']
+            temp_sum += math.sqrt((x2 - x1)**2 + (y2-y1)**2)
+            if temp_sum > distance:
+                break;
+            key_index += 1
+        if key_index == len(nodes)-1:
+            # 端点
+            x = nodes[key_index]['x']-nodes[key_index-1]['x']
+            y = nodes[key_index]['y']-nodes[key_index-1]['y']
+            return get_angle(x, y)
+        else:
+            # 中间点
+            x = nodes[key_index+1]['x'] - nodes[key_index]['x']
+            y = nodes[key_index+1]['y'] - nodes[key_index]['y']
+            return get_angle(x, y)

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycityagent
-Version: 1.1.8
+Version: 1.1.10
 Summary: LLM-based城市模拟器agent构建库
 Author-email: Yuwei Yan <pinkgranite86@gmail.com>
 License: MIT License
@@ -79,9 +79,10 @@ llm_request:
     model: xxx
     (api_base): xxx (this is an optional config, if you use opanai and want to use your own backend LLM model, default to "https://api.openai.com/v1")
   img_understand_request:
-    request_type: qwen
+    request_type: openai / qwen
     api_key: xxx
-    model: xxx
+    model: xxx ('gpt-4-turbo' if you use openai)
+    (api_base): same as text_request
   img_generate_request:
     request_type: qwen
     api_key: xxx
@@ -110,11 +111,19 @@ apphub_request:
 
 #### LLM_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**
-- If you want to use your backend models, set the **api_base** (only available when using **openai**)
-  - default value: "https://api.openai.com/v1"
+- **text_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**
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+    - default value: "https://api.openai.com/v1"
+- **img_understand_request**
+  - By now, we support **qwen** and **openai**
+  - If choose **openai**, then the **model** has to be '**gpt-4-turbo**'
+  - If you want to use your backend models, set the **api_base** (only available when using **openai**)
+      - default value: "https://api.openai.com/v1"
+- **img_generate_request**
+  - By now, only [**qwen**] is supported
 
 #### CITYSIM_REQUEST
 - Most of the configuration options in this part are determined, such as **simulator.server**, **map_request.mongo_coll**, **route_request.server**

{pycityagent-1.1.8 → pycityagent-1.1.10}/pycityagent.egg-info/SOURCES.txt

@@ -6,6 +6,7 @@ pycityagent/agent.py
 pycityagent/agent_citizen.py
 pycityagent/agent_func.py
 pycityagent/simulator.py
+pycityagent/utils.py
 pycityagent.egg-info/PKG-INFO
 pycityagent.egg-info/SOURCES.txt
 pycityagent.egg-info/dependency_links.txt