why pip instead of conda?

[nick-youngblut]

I'm wondering why you use pip instead of conda/mamba for the following install:

pip install absl-py==0.13.0 biopython==1.79 chex==0.0.7 dm-haiku==0.0.4 dm-tree==0.1.6 immutabledict==2.0.0 jax==0.2.14 ml-collections==0.1.0 numpy==1.19.5 scipy==1.7.0 tensorflow==2.5.0 pandas==1.3.4 tensorflow-cpu==2.5.0

Have you tried: https://conda-forge.org/blog/posts/2021-11-03-tensorflow-gpu/

[sanjaysrikakulam]

Hi @nick-youngblut

No reason. If you prefer installing using conda/mamba please feel free to do so.

AF2 updated their setup and requirements file to use tensorflow-cpu package, so it is reflected here as well.

[nick-youngblut]

Thanks for the quick feedback! Also, thanks for providing the helpful guidelines on installing AF2!

[nick-youngblut]

For anyone who cares, this conda env yaml works for me (Ubuntu 18.04.6):

channels:
- conda-forge
- bioconda
dependencies:
- python=3.8
- pip
- conda-forge::openmm=7.5.1
- conda-forge::cudnn=8.2.1.32
- conda-forge::cudatoolkit=11.0.3
- conda-forge::pdbfixer=1.7
- conda-forge::numpy=1.19.5
- conda-forge::scipy=1.7.0
- conda-forge::pandas=1.3.4
- conda-forge::absl-py=0.13.0
- conda-forge::chex=0.0.7
- conda-forge::dm-haiku=0.0.4
- conda-forge::dm-tree=0.1.6
- conda-forge::immutabledict=2.0.0
- conda-forge::biopython=1.79
- bioconda::hmmer=3.3.2
- bioconda::hhsuite=3.3.0
- bioconda::kalign2=2.04
- pip:
  - jax==0.2.14
  - ml-collections==0.1.0
  - tensorflow==2.5.0
  - tensorflow-cpu==2.5.0

...and then

conda activate MY_AF2_ENV
pip install --upgrade jax jaxlib==0.1.69+cuda111 -f https://storage.googleapis.com/jax-releases/jax_releases.html
pip install jax=0.2.25

...and then apply the openmm patch.

conda info:

       user config file : /ebio/abt3/nyoungblut/.condarc
 populated config files : /ebio/abt3_projects/software/dev/miniconda3_dev/.condarc
                          /ebio/abt3/nyoungblut/.condarc
          conda version : 4.10.1
    conda-build version : not installed
         python version : 3.7.4.final.0
       virtual packages : __linux=5.4.89=0
                          __glibc=2.27=0
                          __unix=0=0
                          __archspec=1=x86_64
       base environment : /ebio/abt3_projects/software/dev/miniconda3_dev  (writable)
      conda av data dir : /ebio/abt3_projects/software/dev/miniconda3_dev/etc/conda
  conda av metadata url : https://repo.anaconda.com/pkgs/main
           channel URLs : https://conda.anaconda.org/conda-forge/linux-64
                          https://conda.anaconda.org/conda-forge/noarch
                          https://conda.anaconda.org/bioconda/linux-64
                          https://conda.anaconda.org/bioconda/noarch
                          https://repo.anaconda.com/pkgs/main/linux-64
                          https://repo.anaconda.com/pkgs/main/noarch
                          https://repo.anaconda.com/pkgs/r/linux-64
                          https://repo.anaconda.com/pkgs/r/noarch
                          https://conda.anaconda.org/leylabmpi/linux-64
                          https://conda.anaconda.org/leylabmpi/noarch
                          https://conda.anaconda.org/r/linux-64
                          https://conda.anaconda.org/r/noarch
                          https://conda.anaconda.org/qiime2/linux-64
                          https://conda.anaconda.org/qiime2/noarch
          package cache : /ebio/abt3_projects/temp_data/nyoungblut/conda_cache
                          /ebio/abt3_projects/software/dev/miniconda3_dev/pkgs
       envs directories : /ebio/abt3_projects/software/dev/miniconda3_dev/envs
                          /ebio/abt3/nyoungblut/.conda/envs
               platform : linux-64
             user-agent : conda/4.10.1 requests/2.25.1 CPython/3.7.4 Linux/5.4.89 ubuntu/18.04.6 glibc/2.27
                UID:GID : 6354:350
             netrc file : None
           offline mode : False