- CONDA INSTALL PACKAGE FOR SPECIFIC ENVIRONMENT UPDATE
- CONDA INSTALL PACKAGE FOR SPECIFIC ENVIRONMENT DOWNLOAD
- CONDA INSTALL PACKAGE FOR SPECIFIC ENVIRONMENT WINDOWS
Make sure you download the wheel files for both gdal and rasterio. Note that this website is kind of the official unofficial source for python wheel files. You can access unofficial wheel files from this site. To install with pip we just need to download the wheel files for gdal and rasterio and use them for the install. Provide a path to gdal-config using a GDAL_CONFIG environment variable or use a GDAL_VERSION environment variable.ĭon’t worry. More information is available in the README.ĮRROR: A GDAL API version must be specified.
CONDA INSTALL PACKAGE FOR SPECIFIC ENVIRONMENT WINDOWS
INFO:root:Building on Windows requires extra options to setup.py to locate needed GDAL files. ERROR: Command errored out with exit status 1:
If you do run the command pip install rasterio you’ll probably get an error message that contains at least some of the following text. Second, rasterio must also be installed from a wheel file or compiled from source ( see documentation here). First, rasterio depends on gdal and if you’ve ever installed gdal for python (on Windows) you know it must be installed from a wheel file or compiled from source code.
There are a couple of reasons why you will get errors if you try to run the command pip install rasterio as you would with most python packages. This article describes each method below in detail. There are dependable ways to install rasterio using both pip and conda. The rasterio package wraps around the Geospatial Data Abstraction Library (GDAL, also available as a python package) to provide core functionality for accessing raster data.Īs with gdal (guide for GDAL installation here), installing rasterio for python on Windows is not as simple as running pip install rasterio. foo_package -> python[version='>=2.7,=3.5,=2.7,=3.The most used python package for reading and writing raster data is probably rasterio. To be incompatible with the existing python installation in your environment: UnsatisfiableError: The following specifications were found Retrying with flexible solve.įound conflicts! Looking for incompatible packages. Solving environment: failed with initial frozen solve. When I try to install it: (scratch) PS C:\Users\griffin> conda install foo_packageĬollecting package metadata (repodata.json): done I have a package foo_package, which conda sees on my channel: (scratch) PS C:\Users\griffin> conda search foo_package It contains: # packages in environment at C:\ProgramData\Anaconda3\envs\scratch: To keep things simple, I have created a "scratch" environment with python only. Installing standard packages like numpy still works. Now, when I try to update/install a package from my channel to any environment it fails. Neither of which helped, so I reverted to my previous conda config settings. Sometimes I have to kill it after a few hours.Īfter looking for solutions to speed up environment solve, I edited my conda config by doing: conda config -set channel_priority strict
CONDA INSTALL PACKAGE FOR SPECIFIC ENVIRONMENT UPDATE
It has the highest priority.Įverything was working to my satisfaction, except that conda update -n base -c defaults conda has become extremely slow. I have a custom channel, on a network share, that I host my conda packages on. I've inherited a machine, with an Anaconda installation, on which I must manage my python environments.
0 kommentar(er)
