Vins_Fusion或Vins_Mono统计各话题Topic的发布频率

news2024/12/14 4:49:51

文章目录

  • 统计各话题Topic的发布频率
    • 串行访问节点并输出
      • 脚本
      • 使用方法
      • 结果
    • 并行访问节点
      • 代码
      • 关键改进点
      • 结果

统计各话题Topic的发布频率

由于我想统计Vins_Fusion_gpu中各个话题的统计频率，看一下平均时间，所以写了一个脚本可以自动输出所有节点的所有Topic的单次的频率。

串行访问节点并输出

脚本

#!/usr/bin/env python3

import rospy
import rosnode
import rostopic
import subprocess
import threading
import time
from collections import deque
import re

class TopicFrequencyChecker:
    def __init__(self, topic, window_size=10, timeout=5):
        self.topic = topic
        self.window_size = window_size
        self.timeout = timeout
        self.timestamps = deque(maxlen=window_size)
        self.lock = threading.Lock()
        self.subscriber = None
        self.start_time = time.time()
        self.rate_info = None

    def callback(self, msg):
        with self.lock:
            current_time = time.time()
            self.timestamps.append(current_time)

    def measure_rate(self):
        try:
            # Get the message type and topic class
            topic_type, real_topic, _ = rostopic.get_topic_class(self.topic, blocking=True)
            if not topic_type:
                return "Topic not found or no messages published."

            # Subscribe to the topic
            self.subscriber = rospy.Subscriber(real_topic, topic_type, self.callback)

            # Wait for messages or until timeout
            start_time = time.time()
            while time.time() - start_time < self.timeout:
                if len(self.timestamps) >= self.window_size:
                    break
                time.sleep(0.1)

            # Calculate rate
            with self.lock:
                if len(self.timestamps) < 2:
                    self.rate_info = "No messages received within the timeout period."
                else:
                    time_diffs = [self.timestamps[i] - self.timestamps[i-1] for i in range(1, len(self.timestamps))]
                    avg_rate = len(self.timestamps) / (self.timestamps[-1] - self.timestamps[0]) if len(self.timestamps) > 1 else 0
                    min_delta = min(time_diffs)
                    max_delta = max(time_diffs)
                    std_dev = (sum((x - (1/avg_rate))**2 for x in time_diffs) / len(time_diffs)) ** 0.5 if len(time_diffs) > 1 else 0
                    window = len(self.timestamps)
                    self.rate_info = f"average rate: {avg_rate:.3f}\n\tmin: {min_delta:.3f}s max: {max_delta:.3f}s std dev: {std_dev:.5f}s window: {window}"

            # Unsubscribe from the topic
            if self.subscriber:
                self.subscriber.unregister()

        except Exception as e:
            self.rate_info = str(e)

        return self.rate_info

def get_published_topics(node_name):
    try:
        node_info = subprocess.check_output(['rosnode', 'info', node_name], stderr=subprocess.STDOUT, text=True)
        lines = node_info.splitlines()
        in_publications_section = False
        topics = []

        for line in lines:
            if "Publications:" in line:
                in_publications_section = True
                continue
            if in_publications_section and not line.strip():
                break
            if in_publications_section and line.strip().startswith('*'):
                topic = line.split()[1]
                topics.append(topic)

        return topics

    except subprocess.CalledProcessError:
        return []
def main():
    rospy.init_node('topic_frequency_checker', anonymous=True)
    nodes = rosnode.get_node_names()

    for node in nodes:
        print(f"Node: {node}")
        topics = get_published_topics(node)
        
        if not topics:
            print("  No published topics found for this node.")
            print("========================================")
            continue

        for topic in topics:
            print(f"  Topic: {topic}")
            checker = TopicFrequencyChecker(topic, window_size=10, timeout=3)
            result = checker.measure_rate()
            print(f"  {result}")
            print("  ----------------------------------------")
        
        print("========================================")

if __name__ == '__main__':
    main()
  • 这个 Python 脚本用于检查 ROS (Robot Operating System) 系统中各个节点发布的话题的频率。它通过订阅每个话题，收集一段时间内的消息时间戳，并计算这些消息的平均发布频率、最小间隔、最大间隔和标准差。
  • __init__: 初始化类实例，设置话题名称、窗口大小（即最多记录多少个时间戳）和超时时间。
  • callback: 每当收到一条消息时，记录当前的时间戳到 timestamps 队列中。
  • measure_rate:
    • 获取话题的消息类型和实际话题名称。
    • 订阅该话题，并等待消息或直到超时。
    • 收集足够的时间戳后，计算平均发布频率、最小间隔、最大间隔和标准差。
    • 如果在超时时间内没有收到足够的消息，则返回提示信息。
    • 最后取消订阅话题并返回频率统计结果。
  • main 函数：
    • 初始化一个 ROS 节点，命名为 topic_frequency_checker。
    • 获取系统中所有活动节点的名称。
    • 对每个节点，调用 get_published_topics 获取其发布的话题。
    • 如果该节点没有发布任何话题，则打印提示信息并继续处理下一个节点。
    • 对每个话题，创建一个 TopicFrequencyChecker 实例，测量该话题的频率，并打印结果。
    • 每个节点处理完毕后，打印分隔符以区分不同节点的输出。

使用方法

  • 新建一个python文件，并将上述内容粘贴进去：
vim cal_feq.py
  • 修改权限并执行：
sudo chmod +x cal_feq.py
./cal_feq.py
  • 或者也可以不修改权限直接执行：
python3 cal_feq.py
  • 执行的时候确保已经运行了VINS节点。
roslaunch vins euroc.launch
rosbag play ~/catkin_ws/datasets/euroc/MH_02_easy.bag

结果

lijyhh@lijyhh-desktop:~/catkin_ws$ ./cal_frq.py 
Node: /rvizvisualisation
  Topic: /clicked_point
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /initialpose
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/match_image/mouse_click
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /move_base_simple/goal
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /rosout
  average rate: 25.041
	min: 0.000s max: 0.103s std dev: 0.04970s window: 10
  ----------------------------------------
========================================
Node: /rosout
  Topic: /rosout_agg
  average rate: 26.270
	min: 0.000s max: 0.107s std dev: 0.04804s window: 10
  ----------------------------------------
========================================
Node: /topic_frequency_checker_2947944_1733974050991
  Topic: /rosout
  average rate: 42.333
	min: 0.000s max: 0.070s std dev: 0.02975s window: 10
  ----------------------------------------
========================================
Node: /vins_estimator
  Topic: /rosout
  average rate: 25.469
	min: 0.000s max: 0.118s std dev: 0.04983s window: 10
  ----------------------------------------
  Topic: /tf
  average rate: 24.588
	min: 0.000s max: 0.109s std dev: 0.05110s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose
  average rate: 11.372
	min: 0.086s max: 0.116s std dev: 0.01289s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose_right
  average rate: 11.135
	min: 0.085s max: 0.115s std dev: 0.01285s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose_visual
  average rate: 11.110
	min: 0.089s max: 0.112s std dev: 0.01230s window: 10
  ----------------------------------------
  Topic: /vins_estimator/extrinsic
  average rate: 11.079
	min: 0.091s max: 0.108s std dev: 0.01166s window: 10
  ----------------------------------------
  Topic: /vins_estimator/imu_propagate
  average rate: 211.338
	min: 0.000s max: 0.047s std dev: 0.01477s window: 10
  ----------------------------------------
  Topic: /vins_estimator/key_poses
  average rate: 10.983
	min: 0.091s max: 0.116s std dev: 0.01254s window: 10
  ----------------------------------------
  Topic: /vins_estimator/keyframe_point
  average rate: 5.256
	min: 0.110s max: 0.293s std dev: 0.05564s window: 10
  ----------------------------------------
  Topic: /vins_estimator/keyframe_pose
  average rate: 8.386
	min: 0.093s max: 0.208s std dev: 0.04998s window: 10
  ----------------------------------------
  Topic: /vins_estimator/margin_cloud
  average rate: 11.069
	min: 0.085s max: 0.112s std dev: 0.01209s window: 10
  ----------------------------------------
  Topic: /vins_estimator/odometry
  average rate: 10.960
	min: 0.082s max: 0.119s std dev: 0.01565s window: 10
  ----------------------------------------
  Topic: /vins_estimator/path
  average rate: 11.230
	min: 0.040s max: 0.135s std dev: 0.02753s window: 10
  ----------------------------------------
  Topic: /vins_estimator/point_cloud
  average rate: 10.906
	min: 0.081s max: 0.125s std dev: 0.01739s window: 10
  ----------------------------------------
  Topic: /vins_estimator/rectify_pose_left
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /vins_estimator/rectify_pose_right
  No messages received within the timeout period.
  ----------------------------------------
========================================
Node: /play_1733973836412008018
  Topic: /cam0/image_raw
  average rate: 22.248
	min: 0.038s max: 0.063s std dev: 0.00772s window: 10
  ----------------------------------------
  Topic: /cam1/image_raw
  average rate: 22.250
	min: 0.028s max: 0.064s std dev: 0.01179s window: 10
  ----------------------------------------
  Topic: /clock
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /imu0
  average rate: 39309.316
	min: 0.000s max: 0.000s std dev: 0.00005s window: 10
  ----------------------------------------
  Topic: /leica/position
  average rate: 18.706
	min: 0.037s max: 0.090s std dev: 0.01946s window: 10
  ----------------------------------------
  Topic: /rosout
  average rate: 25.532
	min: 0.000s max: 0.121s std dev: 0.05059s window: 10
  ----------------------------------------
========================================
Node: /loop_fusion
  Topic: /loop_fusion/base_path
  average rate: 8.708
	min: 0.084s max: 0.329s std dev: 0.07340s window: 10
  ----------------------------------------
  Topic: /loop_fusion/camera_pose_visual
  average rate: 11.317
	min: 0.072s max: 0.140s std dev: 0.02086s window: 10
  ----------------------------------------
  Topic: /loop_fusion/margin_cloud_loop_rect
  average rate: 10.790
	min: 0.079s max: 0.143s std dev: 0.02340s window: 10
  ----------------------------------------
  Topic: /loop_fusion/match_image
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/odometry_rect
  average rate: 11.025
	min: 0.078s max: 0.124s std dev: 0.01662s window: 10
  ----------------------------------------
  Topic: /loop_fusion/path_1
  average rate: 10.835
	min: 0.082s max: 0.141s std dev: 0.02034s window: 10
  ----------------------------------------
  Topic: /loop_fusion/path_2
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_3
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_4
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_5
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_6
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_7
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_8
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_9
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/point_cloud_loop_rect
  average rate: 6.617
	min: 0.096s max: 0.211s std dev: 0.05019s window: 10
  ----------------------------------------
  Topic: /loop_fusion/pose_graph
  average rate: 9.023
	min: 0.084s max: 0.199s std dev: 0.03971s window: 10
  ----------------------------------------
  Topic: /loop_fusion/pose_graph_path
  average rate: 11.219
	min: 0.084s max: 0.110s std dev: 0.01339s window: 10
  ----------------------------------------
  Topic: /rosout
  average rate: 24.987
	min: 0.000s max: 0.113s std dev: 0.05054s window: 10
  ----------------------------------------
========================================

上述timeout是因为规定了时间，可以修改：checker = TopicFrequencyChecker(topic, window_size=10, timeout=3)中的timeout=3来修改时间。

并行访问节点

为了实现并行访问节点并输出结果，我们可以使用 Python 的 concurrent.futures 模块中的 ThreadPoolExecutor。这个模块允许我们创建一个线程池，并发地执行多个任务。这样可以显著提高脚本的效率，尤其是在处理大量节点时。

代码

以下是修改后的代码，使用 ThreadPoolExecutor 来并发地检查每个节点的话题频率：

#!/usr/bin/env python3

import rospy
import rosnode
import rostopic
import subprocess
import threading
import time
from collections import deque
import re
from concurrent.futures import ThreadPoolExecutor, as_completed

class TopicFrequencyChecker:
    def __init__(self, topic, window_size=10, timeout=5):
        self.topic = topic
        self.window_size = window_size
        self.timeout = timeout
        self.timestamps = deque(maxlen=window_size)
        self.lock = threading.Lock()
        self.subscriber = None
        self.start_time = time.time()
        self.rate_info = None

    def callback(self, msg):
        with self.lock:
            current_time = time.time()
            self.timestamps.append(current_time)

    def measure_rate(self):
        try:
            # Get the message type and topic class
            topic_type, real_topic, _ = rostopic.get_topic_class(self.topic, blocking=True)
            if not topic_type:
                return "Topic not found or no messages published."

            # Subscribe to the topic
            self.subscriber = rospy.Subscriber(real_topic, topic_type, self.callback)

            # Wait for messages or until timeout
            start_time = time.time()
            while time.time() - start_time < self.timeout:
                if len(self.timestamps) >= self.window_size:
                    break
                time.sleep(0.1)

            # Calculate rate
            with self.lock:
                if len(self.timestamps) < 2:
                    self.rate_info = "No messages received within the timeout period."
                else:
                    time_diffs = [self.timestamps[i] - self.timestamps[i-1] for i in range(1, len(self.timestamps))]
                    avg_rate = len(self.timestamps) / (self.timestamps[-1] - self.timestamps[0]) if len(self.timestamps) > 1 else 0
                    min_delta = min(time_diffs)
                    max_delta = max(time_diffs)
                    std_dev = (sum((x - (1/avg_rate))**2 for x in time_diffs) / len(time_diffs)) ** 0.5 if len(time_diffs) > 1 else 0
                    window = len(self.timestamps)
                    self.rate_info = f"average rate: {avg_rate:.3f}\n\tmin: {min_delta:.3f}s max: {max_delta:.3f}s std dev: {std_dev:.5f}s window: {window}"

            # Unsubscribe from the topic
            if self.subscriber:
                self.subscriber.unregister()

        except Exception as e:
            self.rate_info = str(e)

        return self.rate_info

def get_published_topics(node_name):
    try:
        node_info = subprocess.check_output(['rosnode', 'info', node_name], stderr=subprocess.STDOUT, text=True)
        lines = node_info.splitlines()
        in_publications_section = False
        topics = []

        for line in lines:
            if "Publications:" in line:
                in_publications_section = True
                continue
            if in_publications_section and not line.strip():
                break
            if in_publications_section and line.strip().startswith('*'):
                topic = line.split()[1]
                topics.append(topic)

        return topics

    except subprocess.CalledProcessError:
        return []

def check_node(node_name):
    print(f"Node: {node_name}")
    topics = get_published_topics(node_name)
    
    if not topics:
        print("  No published topics found for this node.")
        print("========================================")
        return

    for topic in topics:
        print(f"  Topic: {topic}")
        checker = TopicFrequencyChecker(topic, window_size=10, timeout=3)
        result = checker.measure_rate()
        print(f"  {result}")
        print("  ----------------------------------------")
    
    print("========================================")

def main():
    rospy.init_node('topic_frequency_checker', anonymous=True)
    nodes = rosnode.get_node_names()

    # Use ThreadPoolExecutor to run checks in parallel
    with ThreadPoolExecutor(max_workers=10) as executor:
        futures = [executor.submit(check_node, node) for node in nodes]

        # Wait for all tasks to complete
        for future in as_completed(futures):
            future.result()  # This will raise any exceptions that occurred during execution

if __name__ == '__main__':
    main()

关键改进点

  1. ThreadPoolExecutor：

    • 使用 ThreadPoolExecutor 创建一个线程池，默认最大工作线程数为 10（你可以根据需要调整 max_workers 参数）。
    • 对每个节点调用 check_node 函数，并将其提交给线程池执行。

  2. as_completed：

    • 使用 as_completed 来等待所有提交的任务完成，并按完成顺序处理结果。
    • future.result() 确保任何在子线程中抛出的异常都能被捕获并处理。

  3. check_node 函数：

    • 将原来的节点检查逻辑封装到 check_node 函数中，以便可以在多个线程中并发执行。

通过这些改进，脚本将并发地检查每个节点的话题频率，显著提高了处理速度。即使某些节点无法访问或响应较慢，也不会阻塞其他节点的检查。

结果

并行输出可能有些显示问题，会错行，这是正常的。

lijyhh@lijyhh-desktop:~/catkin_ws$ ./parallel_cal_frq.py 
Node: /rosout
Node: /loop_fusion
Node: /vins_estimator
Node: /topic_frequency_checker_2956131_1733974476667
Node: /rvizvisualisation
Node: /play_1733974476168986985
  Topic: /rosout
  Topic: /rosout_agg
  Topic: /clicked_point
  Topic: /cam0/image_raw
  Topic: /rosout
  Topic: /loop_fusion/base_path
  average rate: 22.355
	min: 0.030s max: 0.061s std dev: 0.01066s window: 10
  ----------------------------------------
  Topic: /cam1/image_raw
  average rate: 22.568
	min: 0.039s max: 0.061s std dev: 0.01054s window: 10
  ----------------------------------------
  Topic: /clock
  average rate: 19.122
	min: 0.000s max: 0.286s std dev: 0.08556s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose
  average rate: 19.123
	min: 0.000s max: 0.286s std dev: 0.08559s window: 10
  ----------------------------------------
========================================
  average rate: 41.538
	min: 0.000s max: 0.061s std dev: 0.02991s window: 10
  ----------------------------------------
========================================
  average rate: 19.031
	min: 0.050s max: 0.065s std dev: 0.00737s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose_right
  average rate: 17.956
	min: 0.045s max: 0.103s std dev: 0.01937s window: 10
  ----------------------------------------
  Topic: /loop_fusion/camera_pose_visual
  average rate: 18.924
	min: 0.051s max: 0.066s std dev: 0.00694s window: 10
  ----------------------------------------
  Topic: /vins_estimator/camera_pose_visual
  average rate: 19.454
	min: 0.048s max: 0.064s std dev: 0.00702s window: 10
  ----------------------------------------
  Topic: /loop_fusion/margin_cloud_loop_rect
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /initialpose
  average rate: 19.143
	min: 0.048s max: 0.067s std dev: 0.00775s window: 10
  ----------------------------------------
  Topic: /vins_estimator/extrinsic
  average rate: 18.730
	min: 0.048s max: 0.064s std dev: 0.00783s window: 10
  ----------------------------------------
  Topic: /loop_fusion/match_image
  average rate: 19.981
	min: 0.050s max: 0.064s std dev: 0.00703s window: 10
  ----------------------------------------
  Topic: /vins_estimator/imu_propagate
  average rate: 1023000.976
	min: 0.000s max: 0.000s std dev: 0.00000s window: 10
  ----------------------------------------
  Topic: /vins_estimator/key_poses
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /imu0
  average rate: 26698.307
	min: 0.000s max: 0.000s std dev: 0.00007s window: 10
  ----------------------------------------
  Topic: /leica/position
  average rate: 17.805
	min: 0.052s max: 0.072s std dev: 0.00865s window: 10
  ----------------------------------------
  Topic: /vins_estimator/keyframe_point
  average rate: 16.714
	min: 0.056s max: 0.117s std dev: 0.01910s window: 10
  ----------------------------------------
  Topic: /vins_estimator/keyframe_pose
  average rate: 21.437
	min: 0.039s max: 0.088s std dev: 0.01580s window: 10
  ----------------------------------------
  Topic: /rosout
  average rate: 23.558
	min: 0.000s max: 0.109s std dev: 0.05290s window: 10
  ----------------------------------------
========================================
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/match_image/mouse_click
  average rate: 9.791
	min: 0.059s max: 0.193s std dev: 0.04560s window: 10
  ----------------------------------------
  Topic: /vins_estimator/margin_cloud
  average rate: 10.856
	min: 0.110s max: 0.142s std dev: 0.03367s window: 4
  ----------------------------------------
  Topic: /loop_fusion/odometry_rect
  average rate: 11.175
	min: 0.086s max: 0.111s std dev: 0.01229s window: 10
  ----------------------------------------
  Topic: /vins_estimator/odometry
  average rate: 11.100
	min: 0.087s max: 0.111s std dev: 0.01251s window: 10
  ----------------------------------------
  Topic: /loop_fusion/path_1
  average rate: 10.991
	min: 0.094s max: 0.112s std dev: 0.01194s window: 10
  ----------------------------------------
  Topic: /vins_estimator/path
  average rate: 11.347
	min: 0.045s max: 0.203s std dev: 0.04640s window: 10
  ----------------------------------------
  Topic: /loop_fusion/path_2
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /move_base_simple/goal
  average rate: 11.246
	min: 0.087s max: 0.120s std dev: 0.01348s window: 10
  ----------------------------------------
  Topic: /vins_estimator/point_cloud
  average rate: 11.160
	min: 0.081s max: 0.112s std dev: 0.01356s window: 10
  ----------------------------------------
  Topic: /vins_estimator/rectify_pose_left
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_3
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /rosout
  average rate: 24.035
	min: 0.000s max: 0.113s std dev: 0.05193s window: 10
  ----------------------------------------
========================================
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /vins_estimator/rectify_pose_right
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_4
  No messages received within the timeout period.
  ----------------------------------------
========================================
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_5
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_6
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_7
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_8
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/path_9
  No messages received within the timeout period.
  ----------------------------------------
  Topic: /loop_fusion/point_cloud_loop_rect
  average rate: 22.517
	min: 0.089s max: 0.089s std dev: 0.00000s window: 2
  ----------------------------------------
  Topic: /loop_fusion/pose_graph
  average rate: 10.780
	min: 0.095s max: 0.117s std dev: 0.01202s window: 10
  ----------------------------------------
  Topic: /loop_fusion/pose_graph_path
  average rate: 10.005
	min: 0.058s max: 0.136s std dev: 0.02569s window: 10
  ----------------------------------------
  Topic: /rosout
  average rate: 24.986
	min: 0.000s max: 0.114s std dev: 0.05046s window: 10
  ----------------------------------------
========================================

本文来自互联网用户投稿，该文观点仅代表作者本人，不代表本站立场。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如若转载，请注明出处：http://www.coloradmin.cn/o/2259112.html

如若内容造成侵权/违法违规/事实不符，请联系多彩编程网进行投诉反馈，一经查实，立即删除！

相关文章

【机器学习与数据挖掘实战】案例01：基于支持向量回归的市财政收入分析

【机器学习与数据挖掘实战】案例01：基于支持向量回归的市财政收入分析

【作者主页】Francek Chen 【专栏介绍】 ⌈ ⌈ ⌈机器学习与数据挖掘实战 ⌋ ⌋ ⌋ 机器学习是人工智能的一个分支&#xff0c;专注于让计算机系统通过数据学习和改进。它利用统计和计算方法&#xff0c;使模型能够从数据中自动提取特征并做出预测或决策。数据挖掘则是从大型数…
阅读更多...
Spire.PDF for .NET【页面设置】演示：向 PDF 文档添加页码

Spire.PDF for .NET【页面设置】演示：向 PDF 文档添加页码

在 PDF 文档中添加页码不仅实用&#xff0c;而且美观&#xff0c;因为它提供了类似于专业出版材料的精美外观。无论您处理的是小说、报告还是任何其他类型的长文档的数字副本&#xff0c;添加页码都可以显著提高其可读性和实用性。在本文中&#xff0c;您将学习如何使用Spire.P…
阅读更多...
【iOS】OC高级编程 iOS多线程与内存管理阅读笔记——自动引用计数（三）

【iOS】OC高级编程 iOS多线程与内存管理阅读笔记——自动引用计数（三）

目录 ARC规则 概要 所有权修饰符 __strong修饰符 __weak修饰符 __unsafe_unretained修饰符 __autoreleasing修饰符 ARC规则 概要 “引用计数式内存管理”的本质部分在ARC中并没有改变&#xff0c;ARC只是自动地帮助我们处理“引用计数”的相关部分。 在编译单位上可以…
阅读更多...
An error happened while trying to locate the file on the Hub and we cannot f

An error happened while trying to locate the file on the Hub and we cannot f

An error happened while trying to locate the file on the Hub and we cannot find the requested files in the local cache. Please check your connection and try again or make sure your Internet connection is on. 关于上述comfy ui使用control net预处理器的报错问…
阅读更多...
angular19-官方教程学习

angular19-官方教程学习

周日了解到angular已经更新到19了&#xff0c;想按官方教程学习一遍&#xff0c;工欲善其事必先利其器&#xff0c;先更新工具&#xff1a; 安装新版版本 卸载老的nodejs 20.10.0&#xff0c;安装最新的LTS版本 https://nodejs.org 最新LTS版本已经是22.12.0 C:\Program File…
阅读更多...
计算机毕业设计Python+Vue.js游戏推荐系统 Steam游戏推荐系统 Django Flask 游 戏可视化 游戏数据分析 游戏大数据 爬虫 机

计算机毕业设计Python+Vue.js游戏推荐系统 Steam游戏推荐系统 Django Flask 游 戏可视化 游戏数据分析 游戏大数据 爬虫 机

温馨提示&#xff1a;文末有 CSDN 平台官方提供的学长联系方式的名片&#xff01; 温馨提示&#xff1a;文末有 CSDN 平台官方提供的学长联系方式的名片&#xff01; 温馨提示&#xff1a;文末有 CSDN 平台官方提供的学长联系方式的名片&#xff01; 作者简介&#xff1a;Java领…
阅读更多...
上海亚商投顾：创业板指震荡调整 机器人概念股再度爆发

上海亚商投顾：创业板指震荡调整 机器人概念股再度爆发

上海亚商投顾前言&#xff1a;无惧大盘涨跌&#xff0c;解密龙虎榜资金&#xff0c;跟踪一线游资和机构资金动向&#xff0c;识别短期热点和强势个股。 一.市场情绪 沪指昨日冲高回落&#xff0c;深成指、创业板指盘中跌超1%&#xff0c;尾盘跌幅有所收窄。机器人概念股逆势爆…
阅读更多...
粘贴可运行：Java调用大模型(LLM) 流式Flux stream 输出；基于spring ai alibaba

粘贴可运行：Java调用大模型(LLM) 流式Flux stream 输出；基于spring ai alibaba

在Java中&#xff0c;使用Spring AI Alibaba框架调用国产大模型通义千问&#xff0c;实现流式输出&#xff0c;是一种高效的方式。通过Spring AI Alibaba&#xff0c;开发者可以轻松地集成通义千问模型&#xff0c;并利用其流式处理能力&#xff0c;实时获取模型生成的文本。这…
阅读更多...
【CSS in Depth 2 精译_070】11.3 利用 OKLCH 颜色值来处理 CSS 中的颜色问题（下）：从页面其他颜色衍生出新颜色

【CSS in Depth 2 精译_070】11.3 利用 OKLCH 颜色值来处理 CSS 中的颜色问题（下）：从页面其他颜色衍生出新颜色

当前内容所在位置&#xff08;可进入专栏查看其他译好的章节内容&#xff09; 第四部分 视觉增强技术 ✔️【第 11 章 颜色与对比】 ✔️ 11.1 通过对比进行交流 11.1.1 模式的建立11.1.2 还原设计稿 11.2 颜色的定义 11.2.1 色域与色彩空间11.2.2 CSS 颜色表示法 11.2.2.1 RGB…
阅读更多...
Ajax--实现检测用户名是否存在功能

Ajax--实现检测用户名是否存在功能

目录 &#xff08;一&#xff09;什么是Ajax &#xff08;二&#xff09;同步交互与异步交互 &#xff08;三&#xff09;AJAX常见应用情景 &#xff08;四&#xff09;AJAX的优缺点 &#xff08;五&#xff09;使用jQuery实现AJAX 1.使用JQuery中的ajax方法实现步骤&#xf…
阅读更多...
【PSINS】以速度和位置作为观测量（即6维观测量）的组合导航滤波，EKF实现，提供可直接运行的MATLAB代码

【PSINS】以速度和位置作为观测量（即6维观测量）的组合导航滤波，EKF实现，提供可直接运行的MATLAB代码

原有的代码是以位置作为观测量的，这里提供位置+速度，共6维的观测量，状态量还是15维不变 文章目录 源代码运行结果PS源代码 源代码如下： % 【PSINS】位置与速度为观测的153，EKF。从速度观测的EKF153改进而来 % 2024-12-11/Ver1：位置与速度为观测量% 清空工作空间，清除命…
阅读更多...
探索云原生安全解决方案的未来

探索云原生安全解决方案的未来

我们是否充分意识到云端所面临的网络安全威胁&#xff1f; 在当今互联互通的世界中&#xff0c;维护安全的环境至关重要。云的出现扩大了潜在威胁的范围&#xff0c;因为它催生了机器身份&#xff08;称为非人类身份 (NHI)&#xff09;及其秘密。随着组织越来越多地转向云原生…
阅读更多...
无法正常启动此程序，因为计算机丢失wlanapi.dll

无法正常启动此程序，因为计算机丢失wlanapi.dll

wlanapi.dll丢失怎么办&#xff1f;有没有什么靠谱的修复wlanapi.dll方法_无法启动此程序,因为计算机中丢失wlanapi.dll-CSDN博客 wlanapi.dll是 Windows 操作系统中的一个动态链接库文件&#xff0c;主要与 Windows 无线 LAN (WLAN) API 相关。该DLL提供了许多必要的函数&…
阅读更多...
ADC -DMA

ADC -DMA

文章目录 前言一、ADC配置修改二、配置DMA三、DMA中断四、完整参考代码总结 前言 提示&#xff1a;这里可以添加本文要记录的大概内容&#xff1a; 项目需要&#xff1a; 在原有基础上改进&#xff0c;采用DMA中断采集数据 提示&#xff1a;以下是本篇文章正文内容&#xff0…
阅读更多...
Referer头部在网站反爬虫技术中的运用

Referer头部在网站反爬虫技术中的运用

网站数据的安全性和完整性至关重要。爬虫技术&#xff0c;虽然在数据收集和分析中发挥着重要作用&#xff0c;但也给网站管理员带来了挑战。为了保护网站数据不被恶意爬取&#xff0c;反爬虫技术应运而生。本文将探讨HTTP头部中的Referer字段在反爬虫技术中的应用&#xff0c;并…
阅读更多...
docker搭建haproxy实现负载均衡

docker搭建haproxy实现负载均衡

华子目录 获取haproxy镜像建立haproxy容器的数据卷获取haproxy的配置文件编写yaml文件运行测试 获取haproxy镜像 [rootdocker-node1 ~]# docker pull haproxy:2.3建立haproxy容器的数据卷 conf目录为数据卷 [rootdocker-node1 ~]# mkdir /var/lib/docker/volumes/conf/获取h…
阅读更多...
阿里云数据库MongoDB版助力极致游戏高效开发

阿里云数据库MongoDB版助力极致游戏高效开发

客户简介 成立于2010年的厦门极致互动网络技术股份有限公司&#xff08;以下简称“公司”或“极致游戏”&#xff09;&#xff0c;是一家集网络游戏产品研发与运营为一体的重点软件企业&#xff0c;公司专注于面向全球用户的网络游戏研发与运营。在整个产业链中&#xff0c;公…
阅读更多...
深入探索前端调试神器vConsole

深入探索前端调试神器vConsole

深入探索前端调试神器vConsole 在前端开发过程中&#xff0c;调试工具的重要性不言而喻。而vConsole作为一款轻量级的、可嵌入的JavaScript调试面板&#xff0c;为前端开发者提供了一个便捷的调试解决方案。本文将带你深入了解vConsole的基本概念、作用&#xff0c;并通过丰富…
阅读更多...
子网划分实例

子网划分实例

看到有人问这个问题&#xff1a; 想了一下&#xff0c;这是一个子网划分的问题&#xff1a; 处理方法如图&#xff1a; 这是一个子网划分的问题 设备1用三层交换机&#xff0c;端口设置为路由模式&#xff0c;设备2和设备3为傻瓜交换机模式 设备2和设备3下挂设备都是26为掩码&…
阅读更多...
【机器人】振动分析和控制工具之Bode图

【机器人】振动分析和控制工具之Bode图

Bode 图完整介绍 Bode 图由两个部分组成&#xff1a; 幅值图 (Magnitude Plot)&#xff1a;描述系统对不同频率输入信号的增益大小&#xff08;幅值响应&#xff09;。相位图 (Phase Plot)&#xff1a;描述系统输出信号相对于输入信号的相位差。 Bode 图的横轴是频率&#x…
阅读更多...
推荐文章
最新文章

Copyright @ 2022~2023