hello-neural-networks

Repository to keep notebooks from Deep Learning A-Z 2024 course

Setup with Anaconda

Install Anaconda (Debian)

Refer: https://docs.anaconda.com/anaconda/install/linux/

1. Install dependent libraries

   apt-get install libgl1-mesa-glx libegl1-mesa libxrandr2 libxrandr2 libxss1 libxcursor1 libxcomposite1 libasound2 libxi6 libxtst6

2. To download the installer, open a terminal and use the following command, depending on your Linux architecture:

   # Replace <INSTALLER_VERSION> with the version of the installer file you want to download
   # For example, https://repo.anaconda.com/archive/Anaconda3-2023.09-0-Linux-x86_64.sh
   # All installers can be found at repo.anaconda.com/archive/
   curl -O https://repo.anaconda.com/archive/Anaconda3-<INSTALLER_VERSION>-Linux-x86_64.sh

3. To install, run the following command (Linux x86)

   # Include the bash command even if you aren't using the Bash shell
   # Replace ~/Downloads with the path to the installer file, if necessary
   # Replace <INSTALLER_VERSION> with the version of the installer file
   # For example, Anaconda3-2023.09-0-Linux-x86_64.sh
   bash ~/Downloads/Anaconda3-<INSTALLER_VERSION>-Linux-x86_64.sh

4. Initialize conda

   # Replace <PATH_TO_CONDA> with the path to your conda install
   source <PATH_TO_CONDA>/bin/activate
   conda init

Setup Tenserflow environment in Anaconda

conda create -n tf-gpu tensorflow-gpu
conda activate tf-gpu

Install dependent packages in Anaconda

(tdqm is optional)

pip install pandas matplotlib scikit-learn scipy tdqm keras
conda install -c conda-forge jupyterlab

Setup without Anaconda

Install tenserflow (https://www.tensorflow.org/install/pip)

1. Verify gpu (or skip for cpu only)

   nvidia-smi

2. Create a virtual environment with venv

   python3 -m venv tf-gpu
   source tf-gpu/bin/activate

3. Install tensorflow (gpu)

   pip install --upgrade pip
   pip install tensorflow[and-cuda]

   or install tensorflow (cpu only)

   pip install tensorflow

4. Verify the installation

   • [cpu]

     python3 -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"

   • [gpu]

     python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

5. [GPU only] Virtual environment configuration

   Create symbolic links to NVIDIA shared libraries:

   pushd $(dirname $(python -c 'print(__import__("tensorflow").__file__)'))
   ln -svf ../nvidia/*/lib/*.so* .
   popd

   Create a symbolic link to ptxas:

   ln -sf $(find $(dirname $(dirname $(python -c "import nvidia.cuda_nvcc;         
   print(nvidia.cuda_nvcc.__file__)"))/*/bin/) -name ptxas -print -quit) $VIRTUAL_ENV/bin/ptxas

   Verify the GPU setup:

   python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

Install dependencies

pip install pandas matplotlib scikit-learn scipy tdqm keras

Either install Jupyter Lab or just Jupyter Notebook as per the requirement

pip install jupyterlab

or

pip install notebook

Run with Anaconda

Activate Anaconda Tenserflow environment

conda activate tf-gpu

Start Jupyter Lab

jupyter lab

---

Run without Anaconda

Activate Tenserflow environment

source tf-gpu/bin/activate

Start Jupyter Lab

jupyter lab

or

jupyter notebook