inPath 输入影像outPath 输出影像times 重采样多少倍，一般下采样比较多，这里是下采样import osfrom osgeo import gdaldef resample(inPath, outPath, times):in_ds = gdal.Open(inPath)out_rows = int(in_ds.RasterYSize / times)out_columns = i

from _ctypes import Union, Structure, ArrayImportError: DLL load failed while importing _ctypes: 找不到指定的模块。

具体错误如下：RuntimeError: CUDA error: CUBLAS_STATUS_ALLOC_FAILED when calling `cublasCreate(handle)`Exception raised from createCublasHandle at ..\aten\src\ATen\cuda\CublasHandlePool.cpp:8 (most recent cal

测试链接：GitHub - LongguangWang/DASR: [CVPR 2021] Unsupervised Degradation Representation Learning for Blind Super-Resolutionhttps://github.com/LongguangWang/DASR0.环境windows10系统，python3.7.0，我这个是pytorch的虚拟

稀疏深度图补偿

基于Transformer 的变化检测模型测试

1.OSM数据下载下载就不多说了，去官网地址OpenStreetMaphttps://www.openstreetmap.org/需要登陆账号，没有的自己注册下，登录后如下，手动选择想要下载的区域，不能太大，否则无法下载下载好后的文件是map.osm数据2.OSM数据转化这里用gdal来转化，参考链接：Vector Layers — Python GDAL/OGR Cookbook 1.0 doc

Traceback (most recent call last):  File "dataDeal.py", line 28, in    createJson("/home/wcs/train2014.json","/home/wcs/train")  File "dataDeal.py", line 8, in createJson    data=json.load(f)

EfficientDet复现

wget -O opencv.zip https://github.com/opencv/opencv/archive/3.4.4.zipunzip opencv3.4.4.zipcd opencv3.4.4mkdir buildcd build/cmake -D CMAKE_BUILD_TYPE=Release -D OPENCV_GENERATE_PKGCONFIG=YES -D CMAKE_