Recommended Tools and Setup for Lab

 

Recommended Tools 

Core Stack 

Tool	Purpose
Python	Main language
Jupyter Notebook	Interactive lab work
NumPy	Numerical computation
Pandas	Data handling
Matplotlib / Seaborn	Visualization
Scikit-learn	ML algorithms

---

Advanced Tools 

Tool	Use
Google Colab	No-install lab
Kaggle Notebooks	datasets + practice
TensorFlow / PyTorch	neural networks
Streamlit	mini project deployment

 Best Offline Setup (Recommended)

 Use Anaconda (Easiest)

This is the best choice for students.

🔹 Steps:

1. Download Anaconda from https://www.anaconda.com/download
2. Install it
3. It includes:
   • Python
   • Jupyter Notebook
   • NumPy, Pandas, Matplotlib
4. Open Jupyter Notebook
5. Start coding ML programs

👉 No need to install packages separately

Two Main Ways to Manage Python Packages

✅ Method 1: Using pip (Standard)

• Comes with Python
• Simple and widely used

✅ Method 2: Using conda (Recommended for ML)

• Comes with Anaconda
• Better for handling dependencies

👉 Recommendation:
Use conda for teaching labs, pip for basic usage

---

 Installing Packages (Step-by-Step)

🔹 Using pip

pip install numpy pandas matplotlib seaborn scikit-learn

Install specific version:

pip install numpy==1.26.0

---

🔹 Using conda

conda install numpy pandas matplotlib seaborn scikit-learn

👉 Automatically handles compatibility (very useful for students)

---

Creating Separate Environments (VERY IMPORTANT)

👉 Avoids “it works on my system” problems

🔹 Using conda (Best method)

Create environment:

conda create -n ml_lab python=3.10

Activate:

conda activate ml_lab

Install packages:

conda install scikit-learn pandas matplotlib

---

🔹 Using pip (venv)

python -m venv ml_env

Activate:

• Windows:

ml_env\Scripts\activate

• Linux/Mac:

source ml_env/bin/activate

Then install:

pip install numpy pandas scikit-learn

---

💡 Why environments matter

Without environments:

• Package conflicts ❌
• Version issues ❌
• Broken code ❌

With environments:

• Clean setup ✅
• Reproducibility ✅
• Easy grading ✅

Keeping Track of Packages

🔹 Export environment (important for labs)

pip:

pip freeze > requirements.txt

conda:

conda env export > environment.yml

---

🔹 Install from file

pip:

pip install -r requirements.txt

conda:

conda env create -f environment.yml

👉 This is very useful for distributing lab setups to students

---

Updating Packages

pip:

pip install --upgrade numpy

conda:

conda update numpy

---

Removing Packages

pip:

pip uninstall numpy

conda:

conda remove numpy