Hi Idan, thanks, that's already quite useful. This is typically the kind of default values I encounter when using GDAL in the command line indeed. This does not match the values I find when using

Hi Amaury, I'm not aware of any "default" nodata value per dtype in gdal. Though, you might find the defaults in gdal_calc

Hi, what is the default nodata used by GDAL, and hence rasterio, for any given dtype? For example, rasterio.warp.reproject states: dst_nodata (int or float, optional) – The nodata value used to

Hi all, on the same topic, is there an easy way to get the destination parameters (transform, width, height) exactly as GDAL would do, considering the user might provide any of, or a combination of,

Replacing your FTP code with direct access using the /vsicurl/ virtual filesystem: import rasterio vrt = ''' <VRTDataset rasterXSize="6935" rasterYSize="3351"> <SRS>EPSG:4326</SRS>

You could use a raw raster VRT combined with the /vsitar and /vsigzip virtual filesystems (you could probably use /vsicurl as well to replace your FTP code, but I couldn't get that to work immediately

Hi, The xy method always returns projected coordinates. If you want long and lat, you must reproject the xy values using

Hi Ashley, Yes, I think that should be the standard way. AWS would have to remain a special case until we deprecate the special parameters properly, but we can already do what you outlined above for

Hi, We are working on a process that extracts values from raster data and returns them as GeoJSON geometries. Our use cases are point, line, and polygon. For the line use case, we are trying to

Hi All, I am processing SNODAS data archives that store each variable in a single-band ESRI BIL format raster. rasterio.open() works like a charm when a) fp references a file on disk and b) when a hdr

Hi, Rasterio is compatible with the "oldest supported numpy" version defined in https://github.com/scipy/oldest-supported-numpy/blob/master/setup.cfg. This means that rasterio wheels on PyPI are built

I am using tensorflow 2.5 which requires a specific version of NumPy 1.19.2. I am also using rasterio. Which version of rasterio that requires the same version of numpy ?

Thanks Luke. And everyone else for all great the suggestions. Much appreciated.

You don't need to pre-populate a dataset with a massive empty array, you can just open it in write mode and write chunks of data via windowed writing: profile = { 'driver': 'GTiff', 'tiled':

Thanks for your reply, Sean. > are you proposing that potentially any cloud platform would be made first class in rasterio, concretely, as in GDAL and as with rasterio/S3 today? Yes, thats right. >

Thanks Yon. I want to avoid making the 12 GB numpy zero array and the associated memory demand so solution two is the best bet. Greatly appreciated. Geordie

Thank you Alan. Good to know. Kind regards. Geordie

Reading & writing with windows can be done with rioxarray (rasterio + xarray): import rioxarray rds = rioxarray.open_rasterio("input.tif", lock=False) rds.rio.to_raster("output.tif",

Hey Hobart, I've had success using Rasterio's windowed reading and writing to create a huge raster in chunks. You could do something like this: # Create blank raster that will hold the data. Make sure

Hi there. I was wondering if there was a way to create a multiband 12 GB geotif file with rasterio without consuming a large amount of memory. I’m working on a shared system where resources are