Users of ArcGIS Pro who wish to customize advanced features often rely on Python scripts, as many functions of ArcGIS Pro itself are implemented using Python. However, directly installing various Python packages in the built-in environment can easily lead to compatibility issues (especially since Python packages often have compatibility problems between versions). Therefore, ArcGIS Pro comes with a built-in Conda environment and packages the environment that ArcGIS Pro depends on as a read-only environment. This ensures that ArcGIS Pro remains stable and does not fail to run due to arbitrary modifications by users. As shown in the figure below, the software clearly indicates that the default environment cannot be modified.
Cloning the Environment
So, how can you make modifications? The official method is to clone an environment based on the default environment. All modifications are made in the cloned environment, so even if it gets corrupted, switching back to the default environment ensures that ArcGIS Pro can start normally. However, testing shows that without any measures, the cloning process is extremely slow. After researching online, it was found that cloning an environment in ArcGIS Pro does not copy the local environment but downloads the relevant packages from the network. Unfortunately, the default Esri repository is not friendly in China due to network issues, resulting in a very slow cloning process that takes a long time. How to solve this?
Method 1: Using Tools (Understood by Those in the Know)
This method cannot be described in detail, but those who are familiar with it should understand: it involves accelerating the network speed when Conda installs packages. The advantage of this method is that it does not require modifying any files in ArcGIS Pro and has no side effects.
Method 2: Configuring Domestic Mirrors
The first method has a bit of a learning curve and may not be accessible to every user. Therefore, a more general method is introduced: configuring domestic mirrors for the Python environment. The idea is similar to Method 1, but the difference is that Method 1 does not change the source of the packages but only accelerates the network speed, while Method 2 involves synchronizing all packages from abroad to domestic servers in advance, so that installations are done directly from domestic sources without going through international networks.
PS: Setting up domestic mirrors requires significant hardware and network investments with little revenue, so it is usually done by universities or large companies. Well-known examples include Alibaba Cloud and Tsinghua University.
Back to the topic: since directly downloading packages from abroad is slow, we need to modify the configuration to change the download addresses to domestic ones. The method is as follows:
Open the Conda environment configuration file for ArcGIS Pro. The default directory is:
C:\Program Files\ArcGIS\Pro\bin\Python
Then, copy the .condarc file and rename it to .condarc2 (this is just for backup).
Then, replace the original .condarc file with the following content:
channels:
- esri
- defaults
show_channel_urls: true
default_channels:
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
- https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
custom_channels:
conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
pytorch: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
pytorch-lts: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
deepmodeling: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Then, restart ArcGIS Pro and try cloning an environment again. The speed should be much faster.
If it still doesn't work, open the directory C:\Program Files\ArcGIS\Pro\bin\Python\Scripts, right-click in the blank area of the current directory (hold Shift in Windows 10), and open the command line. Enter .\proenv.bat to activate the Python environment. If it still doesn't work, run conda clean -i to clear the existing Conda cache, and then restart ArcGIS Pro. Generally, modifying the configuration file should suffice.
Summary
Conflicts in the ArcGIS Pro Python environment are common issues that users often encounter. If you face problems with ArcGIS Python conflicts, you can refer to the previous article 'What Version of Python is Used in ArcGIS?' to reset to the factory version.