AI Under the Hood (Beta)

Appearance

Setup ​

This guide helps you set up your environment for the AI/ML tutorials.
We support two paths:

  • Rust for performance and safety.
  • Python for ecosystem richness and beginner-friendliness.

Choose the language you prefer (or follow both for cross-learning).
Both setups include installing the language, libraries, IDE, and verifying with a simple program.

Setup with Rust ​

Step 1: Install Rust ​

  1. Visit rust-lang.org.
  2. Install rustup:
    On Unix-like systems (macOS/Linux):
    bash
    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
    On Windows, follow the instructions on the website.
  3. Update Rust:
    bash
    rustup update
  4. Verify installation:
    bash
    rustc --version
cargo --version

Step 2: IDE Setup ​

Use Visual Studio Code (VS Code):

  • Install VS Code.
  • Add the rust-analyzer extension.

Step 3: Create a Rust Project ​

bash
cargo new rust_ml_tutorial
cd rust_ml_tutorial
cargo run

Expect “Hello, world!” output.

Step 4: Install ML Libraries ​

Add dependencies in Cargo.toml:

toml
[dependencies]
linfa = "0.7"
nalgebra = "0.32"
tch = "0.13"
polars = "0.32"
rust-bert = "0.24"
actix-web = "4"
plotters = "0.3"
ndarray = "0.15"
rand = "0.8"
statrs = "0.16"

Run:

bash
cargo build

Step 5: Verify Rust Setup ​

Test with a matrix:

rust
use nalgebra::Matrix2;

fn main() {
    let matrix = Matrix2::new(1.0, 2.0, 3.0, 4.0);
    println!("Matrix:\n{}", matrix);
}
bash
cargo run

Setup with Python ​

Step 1: Install Python ​

  • Install Python 3.11+ from python.org or use Anaconda.
  • Verify installation:
    bash
    python --version
pip --version

Step 2: Create a Virtual Environment ​

bash
python -m venv venv
source venv/bin/activate   # macOS/Linux
venv\Scripts\activate      # Windows

Step 3: Install ML Libraries ​

bash
pip install numpy pandas scikit-learn torch tensorflow matplotlib seaborn jupyter

Step 4: IDE Setup ​

Use VS Code or Jupyter Notebook:

  • Install VS Code.
  • Add the Python extension.
  • Or, run notebooks:
    bash
    jupyter notebook

Step 5: Verify Python Setup ​

python
import numpy as np

matrix = np.array([[1, 2], [3, 4]])
print("Matrix:\n", matrix)

Run:

bash
python main.py

Troubleshooting ​

  • Rust issues: reinstall via rust-lang.org.
  • Python issues: ensure pip is updated (pip install --upgrade pip).
  • Library conflicts: use cargo clean (Rust) or pip uninstall/pip freeze (Python).
  • IDE issues: check extensions (rust-analyzer, Python).

Next Steps ​

Further Reading ​