How to install - Githubissues

(This is the private account of t-oda-ic) /home/ubuntu7/anaconda3/ is my conda install dir. I'm using a commercial subscription licensed version. /data/disk0/alphafold_check/afm_refiner/ is where I cloned this repo. Sorry for contamination of pip and pip3.

The following is the result of nvidia-smi.

Thu Dec 29 11:28:33 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.141.03   Driver Version: 470.141.03   CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0  On |                  N/A |
| 18%   44C    P8    18W / 250W |    573MiB / 11019MiB |     10%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

The following is the command that I used to install. cudatoolkit==XXX and jaxlib==0.1.69+cudaXXX will depend on your system, I guess.

git clone https://github.com/t-oda-ic/afm_refiner.git

cd afm_refiner/

conda create -n af2_old

conda activate af2_old

conda install -y -c nvidia -c conda-forge \
      openmm=7.5.1 \
      cudatoolkit==11.4 \
      pdbfixer \
      pip \
      python=3.7

pip install --upgrade pip 
pip install -r requirements.txt
pip install "jax[cuda11_cudnn82]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
pip3 install --upgrade jax==0.2.14 jaxlib==0.1.69+cuda111 -f \
      https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

cd /home/ubuntu7/anaconda3/envs/af2_old/lib/python3.7/site-packages 
patch -p0 < /data/disk0/alphafold_check/afm_refiner/docker/openmm.patch

t-oda-ic / afm_refiner

How to install #1