---

lastmod: 2023-11-9

This project aims to process and analyze natural language data, a subfield of artificial intelligence known as Natural Language Processing (NLP).

Installation

Build from source

To install this program from the source, follow these steps:

# Clone the repository to your local machine.
git clone https://github.com/uname/repo.git

# Navigate to the cloned repository.
cd repo

# Install the required dependencies.
pip install -r requirements.txt

# Build the program.
python setup.py 

# Alternatively, you can run the program directly.
python launch.py 

Prerequisite

1. CUDA

   • Check if CUDA is already installed:

   ls /usr/local | grep cuda
   

2. CUDA Toolkit Driver

3. CUDA Toolkit

• https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=20.04&target_type=deb_local

1. CUDA Toolkit Driver

2. RAPIDS (instal with pip)

   RAPIDS is a suite of software libraries for executing end-to-end data science and analytics pipelines entirely on GPUs. Here’s how you can instal

   • If CUDA 12 is installed, use the following command to install RAPIDS:

   pip install \
       --extra-index-url=https://pypi.nvidia.com \
       cudf-cu12 dask-cudf-cu12 cuml-cu12 cugraph-cu12 cuspatial-cu12 cuproj-cu12 cuxfilter-cu12 cucim
   

   • To uninstall, use:

   pip uninstall cudf-cu12 dask-cudf-cu12 cuml-cu12 cugraph-cu12 cuspatial-cu12 cuproj-cu12 cuxfilter-cu12 cucim
   

   • If CUDA 11 is installed, use the following command to install RAPIDS:

   pip install \
       --extra-index-url=https://pypi.nvidia.com \
       cudf-cu11 dask-cudf-cu11 cuml-cu11 cugraph-cu11 cuspatial-cu11 cuproj-cu11 cuxfilter-cu11 cucim
   

   • To uninstall, use:

   pip uninstall cudf-cu11 dask-cudf-cu11 cuml-cu11 cugraph-cu11 cuspatial-cu11 cuproj-cu11 cuxfilter-cu11 cucim
   

   For more information, visit the RAPIDS documentation.

   • https://docs.rapids.ai/install#wsl2

Trouble shootings

If you encounter any issues during the installation or running of the program, here are a few things you can check:

• Check your CUDA installation correctly:

  nvcc -V
  which nvcc 
  

  • https://docs.nvidia.com/cuda/wsl-user-guide/index.html#cuda-support-for-wsl2

• Check your Windows driver.

• Check your PATH variable. If you encounter an error like ImportError: libcuda.so.1: cannot open shared object file: No such file or directory you might need to set your PATH variable in your shell configuration:

  • For fish shell

  sudo find /usr/ -name 'libcuda.so.*'
  
  set -gx CUDA_HOME /usr/local/cuda
  set -gx LD_LIBRARY_PATH /usr/lib/wsl/lib/
  set -gx NUMBAPRO_NVVM /usr/local/cuda/nvvm/lib64/libnvvm.so
  set -gx NUMBAPRO_LIBDEVICE /usr/local/cuda/nvvm/libdevice/
  

  • For bash shell:

  export CUDA_HOME=/usr/local/cuda
  export LD_LIBRARY_PATH=/usr/lib/wsl/lib/
  export NUMBAPRO_NVVM=/usr/local/cuda/nvvm/lib64/libnvvm.so
  export NUMBAPRO_LIBDEVICE=/usr/local/cuda/nvvm/libdevice/
  

Acknowledgements

These providers make this project possible:

• https://huggingface.co/cyberagent/calm2-7b-chat

• https://huggingface.co/datasets/wikipedia