I’m currently working on a project that involves setting up a federated learning and split learning system using PyTorch. My goal is to run this project across multiple Raspberry Pi devices. However, for development and testing purposes, I’m trying to simulate the Raspberry Pi environment on my x86 machine using VirtualBox.

Here’s a brief overview of my project:

Project Structure:
server.py: Manages client connections and coordinates the training process.
client.py: Handles data loading, model training, and communication with the server.
nbaiot-dataset: Contains the dataset in CSV format.
models: Directory to save trained models.
Requirements:
Python 3.6 or higher
PyTorch, pandas, h5py, psutil, tqdm
The Issue:
I’m trying to install Conda in this simulated environment to manage dependencies. Since I’m running this in VirtualBox (which is x86 architecture), I’m using Miniforge for the installation. However, I want to ensure that my setup closely mimics the Raspberry Pi environment to avoid discrepancies when I deploy the project on actual Raspberry Pi devices.

What I’ve Done:
Set up a Debian-based virtual machine in VirtualBox.
Downloaded and installed Miniforge using the x86_64 installer.
Initialized Conda and created a test environment successfully.
My Questions:
Is using Miniforge in an x86 environment within VirtualBox a good way to simulate the Raspberry Pi for development purposes?
Are there any potential issues I should be aware of when transitioning this setup to actual Raspberry Pi hardware?
Is there a better approach to achieve a Raspberry Pi-like environment on my x86 machine?
Any advice or suggestions from those who have experience with similar setups would be greatly appreciated!

Thanks in advance for your help.