- 博客(13)
- 资源 (1)
- 收藏
- 关注
RSS订阅原创 环境安装步骤
5.2.2安装NVIDIA驱动1.输入sudo gedit /etc/modprobe.d/blacklist.conf 用编辑器打开blacklist.conf配置文件图5-16 修改blacklist2.在文件的最后一行加入下面的命令，屏蔽有影响的驱动包blacklist nouveauoptions nouveau modeset=0加入后按ctrl+s保存，之后关闭图5-17 需要加入的内容3.输入sudo update-initramfs –u 更新linu
2020-12-14 09:45:58
2441
原创 向test.txt文件中写入0-99并修改文件名
# 向test.txt文件中写入0-99import ospath = 'home/hdkj/dataset/ImageSets/Main/test.txt'with open(path, 'w') as f: for i in range(100): f.write(str(i) + '\n')# 修改从平台下载的图片文件名称import ospath = 'home/hdkj/1002/data/'f = os.listdir(path)n = 0for i...
2020-12-14 00:01:12
338
原创 开发环境搭建过程资料
注：文档中所用的mmdetection工具包存放在home下，其余的安装包在 home/package一、安装NVIDIA驱动1.输入 sudogedit/etc/modprobe.d/blacklist.conf 用编辑器打开 blacklist.conf 配置文件2.在文件的最后一行加入下面的命令，屏蔽有影响的驱动包 blacklistnouveau optionsnouveaumodeset=0 加入后按 ctrl+s 保存，之后关闭3.输入 sudoupdate-initramfs–
2020-12-13 23:45:38
244
原创 voc
from .registry import DATASETSfrom .xml_style import XMLDataset@DATASETS.register_moduleclass VOCDataset(XMLDataset): # CLASSES = ('aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', # 'cat', 'chair', 'cow', 'diningtabl...
2020-12-13 23:35:24
188
原创 usb2rtmp
import cv2import subprocess as spcam = cv2.VideoCapture(0)img_counter = 0ret, frame = cam.read()WIDTH = frame.shape[1]HEIGHT = frame.shape[0]RTMP_URL = 'rtmp://192.168.0.115:1935/live/test'FPS = 24def ffpemgtortmp(width, height, recivedrtmp_addr,
2020-12-13 23:34:52
244
原创 train
from __future__ import divisionimport argparseimport osimport torchfrom mmcv import Configfrom mmdet import __version__from mmdet.apis import (get_root_logger, init_dist, set_random_seed, train_detector)from mmdet.datasets ...
2020-12-13 23:34:18
293
原创 test
import argparseimport osimport os.path as ospimport shutilimport tempfileimport mmcvimport torchimport torch.distributed as distfrom mmcv.parallel import MMDataParallel, MMDistributedDataParallelfrom mmcv.runner import get_dist_info, load_checkpo
2020-12-13 23:33:48
966
原创 my_dataset
from .registry import DATASETSfrom .voc import VOCDataset@DATASETS.register_moduleclass MyDataset(VOCDataset): # CLASSES = ('aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', # 'cat', 'chair', 'cow', 'diningtable', 'do...
2020-12-13 23:33:12
368
原创 HDserver-dahua
# -*- coding: utf-8 -*-from flask import Flask, requestimport base64import numpy as npimport jsonfrom PIL import Imageimport PIL.Imagefrom mmdet.apis import init_detector, inference_detectorimport mmcvimport cv2config_file = '/home/hdkj/mmdetect
2020-12-13 23:32:35
1020
原创 faster_rcnn_r50_fpn_1x-dahua
# model settingsmodel = dict( type='FasterRCNN', pretrained='torchvision://resnet50', backbone=dict( type='ResNet', depth=50, num_stages=4, out_indices=(0, 1, 2, 3), frozen_stages=1, style='pytor...
2020-12-13 23:32:00
752
原创 retail-dahua
#!/usr/bin/env python# encoding: utf-8# author: zs# license: (C) Copyright 2019-2020# software: pycharm# file: HDliveObjDetection.py# time: 2020/4/10 下午4:16import cv2import timeimport requestsimport base64from PIL import Image,ImageDraw,ImageFo
2020-12-13 23:31:16
215
原创 init-dahua
from .builder import build_datasetfrom .cityscapes import CityscapesDatasetfrom .coco import CocoDatasetfrom .custom import CustomDatasetfrom .dataset_wrappers import ConcatDataset, RepeatDatasetfrom .loader import DistributedGroupSampler, GroupSample
2020-12-13 23:29:09
131
code人工智能代码
2020-12-14
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人