随笔分类 - Python
Python zone
摘要：训练yolov5识别小黄球模型 1、标注数据 2、整理数据 3、训练：修改：myvoc.yaml myvoc.yaml train: VOC_2022042301/train.txt val: VOC_2022042301/val.txt # number of classes nc: 1 # cl
阅读全文
摘要：labelme coco 好久没用了 #!/usr/bin/env python # coding: utf-8 # In[12]: import argparse import json import os import os.path as osp import warnings import
阅读全文
摘要：安装 dowhy pip install dowhy 安装日志 (wind_2021) F:\TensorflowProject\BayesianCausalInference-master> (wind_2021) F:\TensorflowProject\BayesianCausalInfere
阅读全文
摘要：paddleDetection train log [04/02 14:10:27] ppdet.engine INFO: Epoch: [99] [14520/14658] learning_rate: 0.000001 loss_xy: 12.273930 loss_wh: 3.473294 l
阅读全文
摘要：python plot_one_box def plot_one_box_wall0401(x, img, color=None, label=None, line_thickness=None): # Plots one bounding box on image img tl = line_th
阅读全文
摘要：Pytorch tensor的值 赋值给python变量 pytorch下只打印tensor的数值不打印出device等信息 print(conf) str_1 = f'{conf:.5f}' conf_2 = float(str_1)print(conf_2) tensor(0.12781, de
阅读全文
摘要：python opencv读取摄像头 camera = cv2.VideoCapture(2) flag2 = camera.open(2) #for path, img, im0s, vid_cap in dataset: while flag2: flag, image = camera.rea
阅读全文
摘要：PaddleDetection配置记录 启动环境： (base) C:\Users\BIM>conda activate wind_paddle (wind_paddle) C:\Users\BIM> 下载程序(从github上下载)，进入目录： (wind_paddle) F:\> (wind_p
阅读全文
摘要：记录 #过程误差协方差矩阵 Q=zeros([5,5]) Q[0,0]=w1[0][0] Q[0,1]=w1[0][1] Q[1,0]=w1[1][0] Q[1,1]=w1[1][1] Q[2,2]=w2[0][0] Q[2,3]=w2[0][1] Q[3,2]=w2[1][0] Q[3,3]=w2
阅读全文
摘要：绘制关节点 def draw_bodypose3(canvas, candidate, subset): stickwidth = 2 limbSeq = [[2, 3], [2, 6], [3, 4], [4, 5], [6, 7], [7, 8], [2, 9], [9, 10], \ [10,
阅读全文
摘要：uwb实验数据记录 list_dis = [] list_t = [] for p in range(146): index_dis = sqrt((x[0][p]-list_new_data[0][p])*(x[0][p]-list_new_data[0][p]) + (x[1][p]-list_
阅读全文
摘要：CKF计算流程梳理 # 3：t-1时刻容积点计算 #分解成AT*A（下三角*上三角）取下三角 S1 = np.linalg.cholesky(Pk) #计算容积点,repmat()把元素复制m行n列 # t-1时刻的2n个容积点 # 4：传播容积点，把每一个容积点带入非线性运动状态转移方程，计算出状
阅读全文
摘要：plot_pr_curve2 def plot_pr_curve2(px, py, conf_p,conf_r,ap, save_dir='.', names=()): fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) p
阅读全文
摘要：python排列组合 import itertools for i in itertools.combinations('12345', 4): print(i," ") print(i[0]) print('\n') (wind_2021) F:\PytorchProject\yolov5-mas
阅读全文
摘要：python读取shp文件属性值并保存到csv 使用shapefile库 pip安装shapefile (E:\Anaconda3\install1) C:\Users\luo> (E:\Anaconda3\install1) C:\Users\luo> (E:\Anaconda3\install1
阅读全文
摘要：yolov5训练模型日志 早晨9点多开始训练，刚刚终于训练完了！ python train_20220110.py --img-size 640 --batch-size 1 --epochs 300 --data ./data/myvoc.yaml --cfg ./models/yolov5m.y
阅读全文
摘要：labelme VOC import os import numpy as np import codecs import json from glob import glob import cv2 import shutil from sklearn.model_selection import
阅读全文
摘要：realsense测量2 import pyrealsense2 as rs pipeline = rs.pipeline() config = rs.config() config.enable_stream(rs.stream.depth, 640, 480, rs.format.z16, 30
阅读全文
摘要：matlab chol() [S]=chol(Ppr2)'; np.linalg.cholesky print(" "*20) Ppr2=np.array([[0.0119,-0.0832,-0.0287,-0.0072,0.0011],[-0.0832,5.2764,1.8354,0.7185,-
阅读全文
摘要：numpy.random.randn() import numpy as np print(np.random.rand(5,5)) 随机生成的结果 [[0.30819634 0.56391944 0.00456188 0.34257512 0.92346913] [0.19165846 0.807
阅读全文
浙公网安备 33010602011771号