Lesson 02: Getting Started with Docker, Conda, and Python Environments on Raspberry Pi 5

2.1 Installing Docker on Raspberry Pi 5

Installing Docker on Raspberry Pi 5

There are two main methods to install Docker on your Raspberry Pi 5. Choose the one that best suits your needs.

Option 1: Using Docker's Convenience Script

This is the quickest and simplest method to install Docker:

Download Docker's convenience script:

curl -fsSL https://get.docker.com -o get-docker.sh

Run the script to install Docker:
```
sudo sh get-docker.sh
```

Option 2: Manual Installation

This method offers more control but involves additional steps. Follow these instructions to install Docker using its official APT repository:

Update the package list and install prerequisites:


sudo apt-get update
sudo apt-get install ca-certificates curl

Add Docker's official GPG key:


sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/debian/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

Add Docker's repository to APT sources:


echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/debian $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt-get update

Install Docker packages:


sudo apt-get install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin

After completing these steps, Docker will be installed and ready to use on your Raspberry Pi 5.

Next Steps

After completing either installation method, Docker will be installed and ready to use on your Raspberry Pi 5. You can verify the installation by running:

docker --version

Add yourself to the docker group

sudo usermod -aG docker $USER

Changing the Docker Root Direction

Relocating the Docker root directory on a Raspberry Pi 5 involves changing the directory where Docker stores all its data, including images, containers, volumes, and configurations. Docker uses /var/lib/docker as the root directory by default. If you want to move this directory to a different location, follow these steps:

Step 1: Stop the Docker Service

First, stop the Docker service to ensure no Docker processes are running while you move the data.

sudo systemctl stop docker

Step 2: Create the New Docker Directory

Create the new directory where you want Docker to store its data. For example, if you want to move it to /mnt/docker-data, you can create this directory:

sudo mkdir -p /ssd/docker

Step 3: Move the Existing Docker Data

Now, move the existing Docker data from /var/lib/docker to the new directory:

sudo rsync -aP /var/lib/docker/ /ssd/docker/

The rsync command is used here to ensure that all files are copied correctly and that the ownership and permissions are preserved.

Step 4: Update Docker Configuration

Tell Docker to use the new directory. To do this, create or edit the Docker daemon configuration file:

sudo nano /etc/docker/daemon.json

Add or modify the following line to specify the new directory:


{
    "data-root": "/ssd/docker"
}

Step 5: Start the Docker Service

Now that Docker's configuration has been updated, restart the Docker service:


sudo systemctl start docker

Step 6: Verify the Change

To verify that Docker is now using the new directory, you can run:


sudo docker info | grep "Docker Root Dir"
 Docker Root Dir: /ssd/docker

This command should show the new directory, confirming that Docker uses it correctly.

Step 7: Clean Up (Optional)

If everything is working correctly, and you're confident that the data was successfully moved, you can delete the old Docker data directory to free up space:


sudo rm -rf /var/lib/docker

You have successfully relocated the Docker root directory on your Raspberry Pi 5. Docker should now store all its data in the new directory you specified. This can be particularly useful if you're running low on space on the default filesystem or have a faster or larger storage option.

2.2 Install Conda on Raspberry Pi 5

Install Conda on Raspberry Pi 5

1. Install Conda (Miniforge)

Download Miniforge:

wget https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-aarch64.sh

Install Miniforge:
```
bash Miniforge3-Linux-aarch64.sh
```
Restart the terminal or activate conda:
```
source ~/miniforge3/bin/activate
```
Verify installation:
```
conda --version
```

2. Update Conda

conda update -n base -c conda-forge conda

3. Create a Conda Environment

conda create -n hailo python=3.10
conda activate hailo

4. Install Python Packages

Install numpy and pip:
```
conda install numpy pip
```
Install TensorFlow Lite:
```
pip install tflite-runtime
```

Install HailoRT (Python bindings):

pip install /opt/hailo/extra/hailo_python/hailo*.whl

5. Test the Installation

Test HailoRT:

python -c "import hailo; print('HailoRT OK!')"

Test TensorFlow Lite:

python -c "import tflite_runtime.interpreter; print('TFLite OK!')"

Common Conda Commands Reference

Common Conda Commands and Parameters

1. Environment and Package Commands

Description

Command	Description
`conda create -n myenv python=3.10`	Create environment named myenv with Python 3.10
`conda create --prefix /path/to/myenv python=3.10`	Create environment at a specific folder
`conda env list`	List all environments
`conda activate myenv`	Activate environment by name
`conda activate /path/to/myenv`	Activate environment by folder path
`conda deactivate`	Deactivate current environment
`conda install numpy`	Install package (e.g., numpy)
`conda install -n myenv numpy`	Install package in specific environment
`conda install --prefix /path/to/myenv numpy`	Install package in environment by folder
`conda list`	List packages in current environment
`conda list -n myenv`	List packages in named environment
`conda list --prefix /path/to/myenv`	List packages in environment at folder
`conda update numpy`	Update package
`conda remove numpy`	Remove package
`conda remove -n myenv --all`	Delete environment by name
`conda remove --prefix /path/to/myenv --all`	Delete environment at specific folder
`conda env export > env.yml`	Export environment to YAML file
`conda env create -f env.yml`	Create environment from YAML file
`conda search numpy`	Search for package (numpy)
`conda info`	Show Conda installation info
`conda update conda`	Update Conda itself
`conda config --add channels conda-forge`	Add conda-forge channel

2. Key Parameters

Parameter	Description
`-n myenv`	Specify environment name
`--prefix /path/to/myenv`	Specify environment folder path
`-c conda-forge`	Install from conda-forge channel
`-y`	Automatic yes to prompts
`--all`	Apply to all (e.g., delete all packages in env)
`-f env.yml`	Use YAML file for environment creation

3. Usage Examples


# Create environment by name
conda create -n testenv python=3.9

# Create environment at a specified folder
conda create --prefix /home/pi/project_env python=3.10

# Activate named environment
conda activate testenv

# Activate environment by folder
conda activate /home/pi/project_env

# Install numpy in current environment
conda install numpy

# Install matplotlib in environment at folder
conda install --prefix /home/pi/project_env matplotlib

# Remove environment at folder
conda remove --prefix /home/pi/project_env --all