# reasons you might want to use `environment.yaml` instead of `requirements.txt`:
# - pip installs packages in a loop, without ensuring dependencies across all packages
#   are fulfilled simultaneously, but conda achieves proper dependency control across
#   all packages
# - conda allows for installing packages without requiring certain compilers or
#   libraries to be available in the system, since it installs precompiled binaries
# in case of errors look here: https://pytorch.org/get-started/previous-versions/ 

name: dnabind2

channels:
  - conda-forge
  - defaults
  - nvidia  # GH200
  - pytorch

# it is strongly recommended to specify versions of packages installed through conda
# to avoid situation when version-unspecified packages install their latest major
# versions which can sometimes break things

# current approach below keeps the dependencies in the same major versions across all
# users, but allows for different minor and patch versions of packages where backwards
# compatibility is usually guaranteed

dependencies:
  - python=3.10
  - dask[complete]
  - pip>=23
  - lightning=2.5.1
  - cudnn=9.10.2.21
  - torchmetrics=0.11.4
  - pip:
    - torch==2.6.0+cu124
    - rootutils==1.0.7
    - hydra-core==1.3.2         # Hydra for config management
    - hydra-colorlog==1.2.0     # Allow colorful logging in Hydra
    - omegaconf==2.3.0          # Required by hydra-core
    - pandas==2.2.3
    - lxml==5.3.0
    - pymex==0.9.31
    - gitpython==3.1.44
    - black==25.1.0   # code formatter
    - tqdm==4.67.1
    - matplotlib==3.10.3
    - transformers==4.55.2
    - huggingface_hub==0.34.4
    - biopython==1.85
    - ortools==9.14.6206
    - fair-esm==2.0.0
    - scikit-learn==1.7.1
    - rich==14.1.0
    - wandb==0.21.1
    - --extra-index-url https://download.pytorch.org/whl/cu124
    - -e .

# conda install -c nvidia -c conda-forge cuda-toolkit=12.4 ninja cmake -y
# use the toolkit inside the conda env
#export CUDA_HOME="$CONDA_PREFIX"
#export PATH="$CUDA_HOME/bin:$PATH"
#export LD_LIBRARY_PATH="$CUDA_HOME/lib64:$LD_LIBRARY_PATH"
# recommended by many CUDA builds
#export CUDACXX="$CUDA_HOME/bin/nvcc"
#which nvcc && nvcc -V   # should now show 12.4 under $CONDA_PREFIX/bin/nvcc
#pip install --no-build-isolation mamba_ssm