CONDA & ANACONDA
CHEAT SHEET.

What are Anaconda and Conda?

• Anaconda – a full‑featured distribution of Python/R with 250+ pre‑installed data science packages.
• Miniconda – a minimal installer that includes conda and Python, allowing you to install only what you need.
• Conda – the package manager that works across Linux, macOS, Windows; manages environments and dependencies.

---

Installation

Download Anaconda

• Official download page
• Graphical installer or command‑line (for Linux).

Install Miniconda (Command Line)

# Linux / macOS
curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh
# Follow prompts, then restart shell

# macOS (Apple Silicon)
curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh
bash Miniconda3-latest-MacOSX-arm64.sh

# Windows – download .exe and run

---

Basic Conda Commands

# Check version
conda --version

# Update conda
conda update conda

# Update all packages in base
conda update --all

# Show configuration
conda info
conda info --envs

---

Environment Management

Environments isolate dependencies for different projects.

Create, Clone, Remove

# Create environment with specific Python
conda create --name myenv python=3.10

# Create environment with packages
conda create --name myenv python=3.10 numpy pandas

# Clone environment
conda create --name newenv --clone myenv

# Remove environment
conda remove --name myenv --all

# List all environments
conda env list

# Activate / deactivate
conda activate myenv
conda deactivate

Export / Import Environment

# Export to YAML (reproducible)
conda env export --name myenv > environment.yml

# Export only explicit package versions (faster)
conda list --explicit > packages.txt

# Create environment from YAML
conda env create -f environment.yml

# Update environment from YAML
conda env update -f environment.yml --prune

---

Package Management

Search, Install, Update, Remove

# Search for package
conda search numpy

# Install package in active environment
conda install numpy
conda install numpy=1.24.0
conda install numpy pandas matplotlib

# Install from specific channel
conda install -c conda-forge scikit-learn

# Install with pip (inside conda environment)
pip install some-package

# Update a package
conda update numpy

# Remove a package
conda remove numpy

# List installed packages in active environment
conda list

# List packages in a specific environment
conda list -n myenv

---

Channels

Channels are repositories for conda packages.

• defaults – Anaconda's official channel (stable).
• conda‑forge – community‑driven, more up‑to‑date packages.
• bioconda – bioinformatics packages.
• pytorch – official PyTorch channel.

# Add a channel (priority order)
conda config --add channels conda-forge
conda config --set channel_priority strict

# Show configured channels
conda config --show channels

# Install from a specific channel
conda install -c conda-forge package-name

# Set default channel
conda config --set default_channel conda-forge

---

Managing Python Versions

# Create environment with specific Python
conda create -n py38 python=3.8
conda create -n py39 python=3.9
conda create -n py310 python=3.10

# Change Python version in an existing environment
conda install python=3.9  # (in active environment)

# Check Python version
python --version

---

Advanced Commands

Cleanup and Cache

# Remove unused packages and cache
conda clean --all

# Remove only package tarballs
conda clean -t

# Remove index cache
conda clean -i

Environment File with Pip Packages

environment.yml:
name: myenv
channels:
  - conda-forge
  - defaults
dependencies:
  - python=3.10
  - numpy
  - pandas
  - pip
  - pip:
    - requests
    - flask

Clone root environment (base)

conda create --name baseclone --clone base

Run a command in a specific environment without activating

conda run -n myenv python script.py

---

Common Use Cases

Data Science Setup

conda create -n ds python=3.10
conda activate ds
conda install numpy pandas matplotlib seaborn jupyter notebook
conda install -c conda-forge scikit-learn scipy

Deep Learning with GPU

conda create -n tf python=3.10
conda activate tf
conda install tensorflow-gpu  # for CUDA
# Or for PyTorch (with CUDA)
conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch

---

Best Practices

• Create a new environment for each project – avoid dependency conflicts.
• Export environment.yml – share with team / version control.
• Use conda‑forge for up‑to‑date packages (set as priority channel).
• Pin exact versions in environment.yml for reproducibility.
• Prefer conda install over pip when possible (conda handles dependencies better).
• Keep base environment minimal – avoid installing too many packages in base.
• Use conda clean periodically to free disk space.
• Use mamba as a faster drop‑in replacement for conda (install via conda).
• Name environments descriptively (e.g., project‑name, py310‑ml).
• Document dependencies – include both conda and pip packages in environment.yml.

---

Mamba (Faster Alternative)

# Install mamba
conda install mamba -c conda-forge

# Use mamba instead of conda for faster installs
mamba install numpy pandas
mamba create -n newenv python=3.10

---

Troubleshooting

• Slow installation? – use mamba or configure faster channels.
• Environment conflicts? – solve by using conda install --solver libmamba (new solver).
• Out of disk space? – run conda clean -a.
• Package not found? – check channel; use conda search.
• Mixing pip and conda? – install pip packages after conda packages to avoid conflicts.
• Permission issues? – install Miniconda/Anaconda for the user (no admin required).