MCPcopy Create free account

hub / github.com/IPADS-SAI/MobiAgent / functions

Functions923 in github.com/IPADS-SAI/MobiAgent

⨍ Functions 923◇ Types & classes 124↳ Endpoints 16
↓ 3 callersMethodfrom_string
从字符串获取日志级别
MobiFlow/avdag/logger.py:46
↓ 3 callersMethodgetAction
()
app/app/src/main/java/com/mobi/agent/Message.java:39
↓ 3 callersMethodgetParameters
()
app/app/src/main/java/com/mobi/agent/Message.java:47
↓ 3 callersMethodget_incoming_action
(self)
agent_rr/action_cache/tree.py:182
↓ 3 callersMethodget_level
获取当前日志级别
MobiFlow/avdag/logger.py:321
↓ 3 callersFunctionget_model_output
(model_client, prompt, image_b64=None)
deployment/server.py:170
↓ 3 callersMethodget_text
获取所有文字内容的拼接字符串
utils/ocr_engine.py:83
↓ 3 callersFunctionget_weights_manager
获取全局权重管理器实例
utils/weights_manager.py:203
↓ 3 callersMethodhasPermission
检查是否有有效的截图权限
app/app/src/main/java/com/mobi/agent/ScreenshotManager.java:215
↓ 3 callersFunctionhideInputModal
()
collect/manual/static/js/script.js:1227
↓ 3 callersMethodhideKeyboard
隐藏软键盘
app/app/src/main/java/com/mobi/agent/MainActivity.java:1783
↓ 3 callersFunctionidentify_task_type
根据任务描述识别任务类型 Args: task_description: 任务描述 Returns: 任务类型
runner/mobiagent/preference_templates.py:54
↓ 3 callersMethodinputTextToFocusedNode
(AccessibilityNodeInfo node, String text)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:1231
↓ 3 callersMethodisAccessibilityServiceEnabled
检查无障碍服务是否已启用 @return true如果已启用，false如果未启用
app/app/src/main/java/com/mobi/agent/MainActivity.java:1878
↓ 3 callersFunctionis_inside
(box1, box2)
utils/omni_utils.py:185
↓ 3 callersFunctionis_inside
(box1, box2)
agent_rr/omniparser/utils.py:186
↓ 3 callersMethodonPermissionRequired
()
app/app/src/main/java/com/mobi/agent/ScreenshotManager.java:37
↓ 3 callersMethodremoveActiveThread
从活跃线程列表移除线程
app/app/src/main/java/com/mobi/agent/MainActivity.java:2316
↓ 3 callersMethodreset_cur_task
(self, account=False)
agent_rr/run_experiment.py:57
↓ 3 callersFunctionsave_state
保存State到文件
runner/mobiagent/multi_task/mobiagent_refactored.py:186
↓ 3 callersMethodsetTextDirectly
直接设置文本到节点
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:719
↓ 3 callersFunctionshowLoadingAnimation
* 显示加载动画
collect/manual/static/js/script.js:30
↓ 3 callersFunctionshowTaskDescriptionModal
(isNextData = false)
collect/manual/static/js/script.js:986
↓ 3 callersMethodstopCapture
()
app/app/src/main/java/com/mobi/agent/ScreenCaptureService.java:359
↓ 3 callersMethodstopFocusMonitoring
()
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:784
↓ 3 callersMethodswipe
Swipe from (start_x, start_y) to (end_x, end_y)
collect/manual/device.py:45
↓ 3 callersMethodtest_by_type
按类型测试
MobiFlow/universal_test_runner.py:411
↓ 3 callersMethodtopo_order
(self)
MobiFlow/avdag/dag.py:104
↓ 3 callersMethodtrace
(self, msg, *args, **kwargs)
MobiFlow/avdag/logger.py:71
↓ 2 callersMethod__init__
(self, data_path)
agent_rr/run_experiment.py:16
↓ 2 callersMethod__init__
(self, description)
agent_rr/action_cache/tree.py:27
↓ 2 callersMethod__init__
(self)
agent_rr/agent/env.py:10
↓ 2 callersMethod_build_index
从加载的模板构建 llama_index 并将其保存到存储中。
utils/local_experience.py:94
↓ 2 callersMethod_build_messages
构建发送给模型的消息
runner/UI-TARS-agent/ui_tars_automation/framework.py:94
↓ 2 callersFunction_calculate_total_score
计算匹配节点的总分数
MobiFlow/avdag/verifier.py:281
↓ 2 callersMethod_call_model
调用模型获取响应
runner/UI-TARS-agent/ui_tars_automation/framework.py:130
↓ 2 callersMethod_execute_action
执行具体的操作
runner/UI-TARS-agent/ui_tars_automation/framework.py:153
↓ 2 callersMethod_extract_app_name
从app_id提取应用名称
utils/icon_detection.py:311
↓ 2 callersMethod_load_or_build_index
如果存储中存在索引，则加载它，否则构建它。
utils/local_experience.py:74
↓ 2 callersFunction_node_needs_llm
判断节点是否需要LLM验证
MobiFlow/avdag/verifier.py:264
↓ 2 callersFunction_node_needs_ocr
判断节点是否需要OCR验证
MobiFlow/avdag/verifier.py:256
↓ 2 callersMethod_resize_image_if_needed
如果图像尺寸超过最大边长限制，则缩放图像
utils/ocr_engine.py:185
↓ 2 callersFunction_update_successor_reachability
动态更新后继节点的可达性
MobiFlow/avdag/verifier.py:214
↓ 2 callersMethodaddActiveThread
添加线程到活跃线程列表
app/app/src/main/java/com/mobi/agent/MainActivity.java:2306
↓ 2 callersMethodaddToChat
添加消息到聊天列表 @param message 消息内容 @param sentBy 发送者类型（Message.SENT_BY_ME 或 Message.SENT_BY_BOT）
app/app/src/main/java/com/mobi/agent/MainActivity.java:1677
↓ 2 callersMethodanalyzeCurrentScreen
(AccessibilityNodeInfo node, int depth)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:479
↓ 2 callersMethodanalyzeUIForInputElements
(AccessibilityNodeInfo rootNode)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:1040
↓ 2 callersFunctionannotate
This function annotates an image with bounding boxes and labels. Parameters: image_source (np.ndarray): The source image to be annot
utils/omni_utils.py:310
↓ 2 callersMethodapp_stop
(self, package_name)
runner/mobiagent/mobiagent.py:52
↓ 2 callersFunctionbacktrack
(start: str)
MobiFlow/avdag/verifier.py:437
↓ 2 callersFunctionbacktrack_path
(nid: str, visited: set)
MobiFlow/avdag/verifier.py:173
↓ 2 callersMethodcallAPI
调用API方法 - 一次性处理整个JSON数据并执行所有命令 @param question 用户输入的问题 功能说明： 1. 显示"正在执行"提示 2. 遍历整个reactData数组 3. 解析并执行所有function命令 4. 综合显示所有执行结果
app/app/src/main/java/com/mobi/agent/MainActivity.java:373
↓ 2 callersMethodcaptureScreen
(ScreenCaptureCallback callback)
app/app/src/main/java/com/mobi/agent/ScreenCaptureService.java:378
↓ 2 callersFunctionceil_by_factor
Returns the smallest integer greater than or equal to 'number' that is divisible by 'factor'.
runner/UI-TARS-agent/ui_tars_automation/coordinate_processor.py:27
↓ 2 callersFunctioncheckDeviceStatus
()
collect/manual/static/js/script.js:50
↓ 2 callersMethodconvertImageToBase64
将图片文件转换为Base64编码字符串 @param imagePath 图片文件路径 @return Base64编码的字符串，如果转换失败返回null
app/app/src/main/java/com/mobi/agent/MainActivity.java:2104
↓ 2 callersMethodconvert_model_coords_to_actual
将模型输出的坐标转换为实际设备坐标 参考README_coordinates.md的实现
runner/UI-TARS-agent/ui_tars_automation/coordinate_processor.py:68
↓ 2 callersFunctionconvert_qwen3_coordinates_to_absolute
将 Qwen3 模型返回的相对坐标（0-1000范围）转换为绝对坐标 Args: bbox_or_coords: 相对坐标或边界框，范围为 0-1000 img_width: 图像宽度 img_height: 图像高度
runner/mobiagent/mobiagent.py:394
↓ 2 callersFunctionconvert_qwen3_coordinates_to_absolute
将 Qwen3 模型返回的相对坐标（0-1000范围）转换为绝对坐标 Args: bbox_or_coords: 相对坐标或边界框，范围为 0-1000 img_width: 图像宽度 img_height: 图像高度
runner/mobiagent/multi_task/executor.py:167
↓ 2 callersFunctioncreate_decider_entries_for_one_task
(task, react_data, actions, root, data_path, out_path, factor, rules, unexpected_img_safe_abspaths, is_train,
collect/construct_sft.py:373
↓ 2 callersFunctioncreate_swipe_visualization
为滑动动作创建可视化图像
runner/mobiagent/mobiagent.py:422
↓ 2 callersMethodcreate_visualization_image
创建带有点击位置标记的可视化图像
runner/UI-TARS-agent/ui_tars_automation/coordinate_processor.py:97
↓ 2 callersFunctioncross_app_step1_reranker
(path)
agent_rr/train/prepare_data.py:182
↓ 2 callersMethoddetect_icon
在图像中检测指定图标 Args: image: 目标图像（numpy数组或文件路径） icon_path: 图标模板路径 threshold: 相似度阈值，None时使用默认值
utils/icon_detection.py:177
↓ 2 callersMethoddump_hierarchy
(self)
runner/mobiagent/multi_task/mobiagent_refactored.py:76
↓ 2 callersMethodembed
(self, tasks, steps=None)
agent_rr/action_cache/embedder.py:13
↓ 2 callersFunctionembedding_main
(train_path, test_path)
agent_rr/train/prepare_data.py:131
↓ 2 callersMethodexecuteDoneCommand
执行动作前截图命令 - 使用新的MediaProjection前台服务架构 在每次动作执行前截取一张当前屏幕状态图
app/app/src/main/java/com/mobi/agent/MainActivity.java:1474
↓ 2 callersMethodexecuteScreenshot
实际执行截图操作
app/app/src/main/java/com/mobi/agent/ScreenshotManager.java:172
↓ 2 callersMethodexecuteTerminateCheatCode
执行terminate异常终止代码 - 在无障碍权限未启用等非正常情况时调用
app/app/src/main/java/com/mobi/agent/MainActivity.java:1914
↓ 2 callersFunctionexecute_single_task
执行单个任务的通用函数 Args: task_description: 任务描述 device: 设备对象 data_dir: 数据保存目录 use_experience: 是否使用经验改写任务
runner/mobiagent/mobiagent.py:945
↓ 2 callersFunctionextract_all_bounds
提取截图中的所有边界框信息，优化参数以识别更多可点击元素
utils/parse_omni.py:12
↓ 2 callersMethodextract_from_directory
从指定目录及其子目录中提取所有任务描述 Args: directory_path: 目标目录路径 Returns: 包含任务描述信息的字典列表
MobiFlow/auto_rules/task_extractor.py:16
↓ 2 callersMethodextract_full_description
从结果中的多个 JSON 对象中提取 full_experience 内容，支持单任务和多任务格式。
utils/local_experience.py:173
↓ 2 callersMethodfindAllEditableElementsEnhanced
(AccessibilityNodeInfo node, List<AccessibilityNodeInfo> editableNodes)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:1022
↓ 2 callersMethodfindAnyInputElement
更宽松的输入框检测方法，适用于现代应用
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:1162
↓ 2 callersMethodfindEditableElementNearCoordinate
(AccessibilityNodeInfo rootNode, int targetX, int targetY, int radius)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:1868
↓ 2 callersFunctionfindElementAtPosition
(x, y)
collect/manual/static/js/script.js:1486
↓ 2 callersMethodfindScreenshotButton
(AccessibilityNodeInfo node)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:2893
↓ 2 callersFunctionfloor_by_factor
Returns the largest integer less than or equal to 'number' that is divisible by 'factor'.
runner/UI-TARS-agent/ui_tars_automation/coordinate_processor.py:31
↓ 2 callersMethodformatGestureLog
格式化手势操作日志 @param operation 操作类型 @param startX 起始x坐标 @param startY 起始y坐标 @param endX 结束x坐标（可选） @param endY 结束y坐标（可选） @return 格式化的日志字符串
app/app/src/main/java/com/mobi/agent/GestureUtils.java:72
↓ 2 callersFunctiongenerate_preference_extraction_prompt
生成偏好提取的prompt Args: task_data: 任务数据 task_type: 任务类型 Returns: 生成的prompt
runner/mobiagent/preference_templates.py:106
↓ 2 callersMethodgetSentBy
()
app/app/src/main/java/com/mobi/agent/Message.java:23
↓ 2 callersMethodgetService
()
app/app/src/main/java/com/mobi/agent/ScreenCaptureService.java:512
↓ 2 callersMethodget_app_task_trajectories
(self)
agent_rr/run_experiment.py:36
↓ 2 callersMethodget_cached_action
(self, task)
agent_rr/action_cache/tree.py:151
↓ 2 callersFunctionget_icon_detection_service
获取图标检测服务实例
utils/icon_detection.py:559
↓ 2 callersFunctionget_lshare
(trajectory1, trajectory2)
agent_rr/train/prepare_data.py:13
↓ 2 callersFunctionget_ocr_processor
(*args, **kwargs)
MobiFlow/avdag/ocr_processor.py:71
↓ 2 callersFunctionget_screenshot
(img_path, factor=0.4)
collect/auto/server.py:56
↓ 2 callersFunctionget_task_templates
(raw_template)
agent_rr/train/task_template.py:6
↓ 2 callersFunctionget_verifier_logger
获取验证器日志器
MobiFlow/avdag/logger.py:384
↓ 2 callersFunctionget_xywh
(input)
utils/omni_utils.py:473
↓ 2 callersFunctionget_xywh
(input)
agent_rr/omniparser/utils.py:452
↓ 2 callersFunctionhandleClickAction
(x, y)
collect/manual/static/js/script.js:582
↓ 2 callersMethodhandleMediaProjectionPermissionResult
统一处理MediaProjection权限申请结果 @param resultCode 结果码 @param data 结果数据 @param source 权限申请来源（用于日志）
app/app/src/main/java/com/mobi/agent/MainActivity.java:1626
↓ 2 callersFunctionhandleSwipeAction
(startX, startY, endX, endY, deltaX, deltaY)
collect/manual/static/js/script.js:658
↓ 2 callersFunctionhistory_str
(history)
collect/construct_sft.py:72
↓ 2 callersMethodinput
Input text to the device
collect/manual/device.py:40
↓ 2 callersMethodinputTextSafely
(String text)
app/app/src/main/java/com/mobi/agent/MyAccessibilityService.java:510
← previousnext →101–200 of 923, ranked by callers