Merge pull request ilastik#2904 from btbest/remove-mamba

Stop using mamba / disallow pytorch2.4.1

.github/workflows/core-conda-bld.yml

@@ -21,14 +21,14 @@ jobs:
           auto-activate-base: true
           activate-environment: ""
           channel-priority: strict
-          miniforge-variant: Mambaforge
-          use-mamba: true
+          miniforge-version: latest
+          conda-solver: libmamba
       - name: install xvfb/deps
         run: |
           sudo apt-get update
           sudo apt-get install -yy libgl1-mesa-dev xvfb curl
       - name: install common conda dependencies
-        run: mamba install -n base -c conda-forge mamba boa setuptools_scm -y
+        run: conda install -n base -c conda-forge boa setuptools_scm -y
       - uses: actions/cache@v4
         with:
           path: |
@@ -84,15 +84,15 @@ jobs:
           auto-activate-base: true
           activate-environment: ""
           channel-priority: strict
-          miniforge-variant: Mambaforge
-          use-mamba: true
+          miniforge-version: latest
+          conda-solver: libmamba
       - name: install xvfb/deps
         if: matrix.os == 'ubuntu-latest'
         run: |
           sudo apt-get update
           sudo apt-get install -yy mesa-utils libgl1-mesa-dev xvfb curl
       - name: install common conda dependencies
-        run: mamba install -n base -c conda-forge boa setuptools_scm -y
+        run: conda install -n base -c conda-forge boa setuptools_scm -y
       - uses: actions/cache@v4
         with:
           path: |

.github/workflows/deploy.yml

@@ -18,8 +18,8 @@ jobs:
           auto-update-conda: true
           auto-activate-base: true
           activate-environment: ""
-          miniforge-variant: Mambaforge
-          use-mamba: true
+          miniforge-version: latest
+          conda-solver: libmamba
       - name: install build dependencies
         run: |
           conda install -n base -c conda-forge boa setuptools_scm anaconda-client -y

CONTRIBUTING.md

@@ -11,7 +11,7 @@ The following text equips you with knowledge that makes contributing to ilastik
 
 ## Development Environment
 
-1. Download and install [Git][git], [GitHub CLI][github-cli], and _the latest 64-bit_ [Mambaforge][mambaforge].
+1. Download and install [Git][git], [GitHub CLI][github-cli], and _the latest 64-bit_ [Miniforge][miniforge].
    - Windows: all following commands should be executed in the _Anaconda Prompt_ from the Start Menu.
    - Linux: Git might be already preinstalled, GitHub CLI might be available in your system package manager.
    - macOS: Git is already preinstalled, GitHub CLI is available in [Homebrew][homebrew].
@@ -29,20 +29,19 @@ The following text equips you with knowledge that makes contributing to ilastik
    gh repo clone ilastik
    ```
 
-1. If you have an existing [Miniconda][miniconda] installation, you can use it for ilastik development, but install [Mamba][mamba] and configure conda to use [strict channel priority][strict-channel-priority]:
+1. If you have an existing conda installation, you can use it for ilastik development, but configure it to use [strict channel priority][strict-channel-priority].
 
    ```
    conda config --set channel_priority strict
-   conda install --name base --channel conda-forge mamba
    ```
 
 1. Create a new environment and install dependencies:
 
    ```
    # from within the ilastik folder
-   mamba deactivate
-   mamba env remove --name ilastik
-   mamba env create --name ilastik --file dev/environment-dev.yml
+   conda deactivate
+   conda env remove --name ilastik
+   conda env create --name ilastik --file dev/environment-dev.yml
    ```
 
 1. Install repositories as packages in development mode:
@@ -71,7 +70,7 @@ The following text equips you with knowledge that makes contributing to ilastik
 1. Launch ilastik:
 
    ```
-   mamba activate ilastik
+   conda activate ilastik
    cd ilastik
    python ilastik_scripts/ilastik_startup.py
    ```
@@ -156,12 +155,10 @@ But please run those tools on the code you are contributing :)
 
 [git]: https://git-scm.com/
 [github-cli]: https://cli.github.com/
-[miniconda]: https://docs.conda.io/en/latest/miniconda.html
-[mambaforge]: https://github.com/conda-forge/miniforge#mambaforge
 [homebrew]: https://brew.sh/
 [conda-build]: https://docs.conda.io/projects/conda-build/en/latest/
 [conda-develop]: https://docs.conda.io/projects/conda-build/en/latest/resources/commands/conda-develop.html
-[mamba]: https://mamba.readthedocs.io/en/latest/
+[miniforge]: https://github.com/conda-forge/miniforge?tab=readme-ov-file#miniforge3
 [ilastik]: https://github.com/ilastik/ilastik
 [volumina]: https://github.com/ilastik/volumina
 [github-flow]: https://guides.github.com/introduction/flow/

Readme.md

@@ -37,10 +37,10 @@ If you don't have a dataset to work with, download one of the example projects t
 ### Conda installation (experimental)
 
 ilastik is also available as a conda package on our `ilastik-forge` conda channel.
-We recommend using [mamba][mamba] instead of conda, for faster installation:
+You can install it from the commandline using [conda][conda]:
 
 ```bash
-mamba create -n ilastik --override-channels -c pytorch -c ilastik-forge -c conda-forge ilastik
+conda create -n ilastik --override-channels -c pytorch -c ilastik-forge -c conda-forge ilastik
 
 # activate ilastik environment and start ilastik
 conda activate ilastik
@@ -99,4 +99,4 @@ For more complex changes, see [CONTRIBUTING.md](CONTRIBUTING.md) for details.
 [imagesc-search]: https://forum.image.sc/search
 [issues]: https://github.com/ilastik/ilastik/issues
 [edit-files-on-github]: https://docs.github.com/en/free-pro-team@latest/github/managing-files-in-a-repository/editing-files-in-another-users-repository
-[mamba]: https://github.com/mamba-org/mamba
+[conda]: https://github.com/conda-forge/miniforge?tab=readme-ov-file#miniforge3

appveyor.yml

@@ -23,12 +23,12 @@ install:
         }
   - cmd: set PATH=%MINIFORGE%;%MINIFORGE%\Scripts;%MINIFORGE%\Library\bin;%PATH%
   - cmd: conda config --set always_yes yes --set changeps1 no --set channel_priority strict
-  - cmd: conda update -n base -c conda-forge conda mamba
-  - cmd: mamba install -n base -c conda-forge setuptools_scm
+  - cmd: conda update -n base -c conda-forge conda
+  - cmd: conda install -n base -c conda-forge setuptools_scm
   - |
-    mamba env create --name %ENV_NAME% --file dev\environment-dev.yml
-    mamba install --name %ENV_NAME% --freeze-installed -c ilastik-forge -c conda-forge volumina
-    mamba run -n %ENV_NAME% pip install -e .
+    conda env create --name %ENV_NAME% --file dev\environment-dev.yml
+    conda install --name %ENV_NAME% --freeze-installed -c ilastik-forge -c conda-forge volumina
+    conda run -n %ENV_NAME% pip install -e .
   - conda clean -p
 
 build: off

conda-recipe/meta.yaml

@@ -116,7 +116,7 @@ outputs:
       run:
         - python 3.9.*
         - ilastik-core {{ setup_py_data.version }}
-        - pytorch >=1.6
+        - pytorch >=1.6,!=2.4.1
         - tiktorch 23.11.0*
         - torchvision
         - volumina
@@ -168,7 +168,7 @@ outputs:
       run:
         - python 3.9.*
         - ilastik-core {{ setup_py_data.version }}
-        - pytorch >=1.6
+        - pytorch >=1.6,!=2.4.1
         - tiktorch 23.11.0*
         - torchvision
         - pytorch-cuda

dev/environment-dev.yml

@@ -53,7 +53,7 @@ dependencies:
   # can be changed to request gpu versions
   # pytorch 2.4.* will work for all platforms except osx-64, where currently the
   # latest version is 2.2.2
-  - pytorch 2.4.*
+  - pytorch 2.4.*,!=2.4.1
   - cpuonly  # comment out for pytorch with cuda
   # mamba does not "respect" track_features, which is the idea behind cpuonly,
   # with ilastik-pytorch-version-helper-cpu we help mamba on linux and windows

notebooks/Readme.md

@@ -12,26 +12,26 @@ notebooks/
 ├── ...
 ```
 
-We use _mamba/(conda)_ for Python-based projects and recommend it for scientific Python development.
-It is assumed that you have already installed _mamba_ and are familiar with using a Terminal.
+We use _conda_ for Python-based projects and recommend it for scientific Python development.
+It is assumed that you have already installed _conda_ and are familiar with using a Terminal.
 
 In order to run the notebook type the following in a Terminal/Command Line:
 
-Note: mamba has to be configured with [_strict channel priority_](https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-channels.html#strict-channel-priority) in order to produce consistent environments.
+Note: conda has to be configured with [_strict channel priority_](https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-channels.html#strict-channel-priority) in order to produce consistent environments.
 
 
 ```bash
 # make sure to use strict channel priority - this setting is global but in general a good idea
-mamba config --set channel_priority strict
+conda config --set channel_priority strict
 
 # create n environment - make sure to choose an appropriate <environment_name>
-$ mamba env create -n <environment_name> -f environment.yml
+$ conda env create -n <environment_name> -f environment.yml
 # this will produce a lot of output
 
-$ mamba activate <environment_name>
+$ conda activate <environment_name>
 
 # start the notebook server in the current folder
 $ jupyter notebook --notebook-dir .
 
-# to close the server hit `ctrl + c` in the terminal shut down the server by confirming with `y` 
+# to close the server hit `ctrl + c` in the terminal shut down the server by confirming with `y`
 ```