Project Commands

Commands for project initialisation and setup.

---

Quick Reference

# Create new project (wizard)
project init my-project

# Launch existing project (smart - builds image if needed)
project launch my-project

# Individual steps
dir-create my-project
git-init my-project
readme-create my-project

---

Recommended Workflow

# One-time: Create project
project init my-thesis --type=cv

# Daily: Launch project
project launch my-thesis

# When done: Retire container
container retire my-thesis

---

project-init

Complete project initialisation (L4 wizard)

project-init [project-name] [OPTIONS]

Options:

Option	Description
--type=TYPE	Project type: ml, cv, nlp, rl, audio, ts, llm, custom
--quick	Skip interactive prompts, use defaults (type=ml)
--no-git	Skip Git initialization
--guided	Educational mode

Examples:

project-init                      # Interactive
project-init my-research          # Specify name
project-init my-thesis --type=cv  # Computer vision project
project-init test --quick         # Quick setup with defaults

What it does:

1. dir-create - Creates ~/workspace/<project>/
2. git-init - Initializes Git repository
3. readme-create - Generates README.md
4. image-create - Builds custom Docker image
5. container-deploy - Deploys first container

Equivalent to running all steps individually.

---

project-launch

Launch container for existing project (L4 wizard)

project-launch [project-name] [OPTIONS]

Options:

Option	Description
--guided	Educational mode
--open	Start and enter terminal
--background	Start in background
--rebuild	Force rebuild image

Examples:

project-launch              # Interactive (select from list)
project-launch my-thesis    # Launch specific project
project-launch my-thesis --open  # Launch and enter

What it does:

1. Shows menu of projects in ~/workspace/ (if no name given)
2. Checks if Docker image exists
3. If no image: runs image-create automatically
4. Runs container-deploy to start container

Key difference from container-deploy:

• project-launch = Smart (handles image creation automatically)
• container-deploy = Direct (requires image to exist)

---

dir-create

Create workspace directory (L2 atomic)

dir-create <project-name> [OPTIONS]

Options:

Option	Description
--type=TYPE	Structure type: data-science (default), blank
--force	Overwrite existing directory
--guided	Educational mode

Examples:

dir-create my-project               # Create with data-science structure
dir-create my-project --type=blank  # Create empty directory
dir-create my-project --guided      # With explanations

Creates standard directory structure:

~/workspace/my-project/
├── data/{raw,processed,external}/
├── models/checkpoints/
├── notebooks/
├── scripts/
├── configs/
├── outputs/{logs,figures}/
└── tests/

---

git-init

Initialize Git repository (L2 atomic)

git-init <project-name> [OPTIONS]

Options:

Option	Description
--user=NAME	Git user name (prompts if not set)
--email=EMAIL	Git user email (prompts if not set)
--remote=URL	Add remote repository URL
--no-lfs	Skip Git LFS setup
--guided	Educational mode

Examples:

git-init my-project                           # Initialize Git
git-init my-project --user="Jane" --email="jane@example.com"
git-init my-project --remote=git@github.com:user/repo.git
git-init my-project --guided                  # With explanations

Features:

• Comprehensive ML/DS .gitignore
• Git LFS for large model files (.pth, .pt, .h5, .ckpt, .bin)
• User config setup
• Optional remote repository

---

readme-create

Generate README.md (L2 atomic)

readme-create <project-name> [OPTIONS]

Options:

Option	Description
--type=TYPE	Project type: ml (default), cv, nlp, rl
--desc="TEXT"	Project description
--structure=TYPE	Structure: data-science (default), blank
--commit	Create initial Git commit (requires Git repo)
--guided	Educational mode

Examples:

readme-create my-project                                # Basic README
readme-create cv-project --type=cv --desc="Image classification"
readme-create my-project --commit                       # Create and commit
readme-create my-project --guided                       # With explanations

Creates a template README with sections for project description, setup, usage, and license.

---

Workflow Example

Using project-init (recommended)

project-init my-thesis
# Handles everything automatically

Manual setup

dir-create my-thesis
git-init my-thesis
readme-create my-thesis
image-create my-thesis
container-deploy my-thesis --open

---

Project Structure

After project-init:

~/workspace/my-project/
├── .git/              # Git repository
├── .gitignore         # Python/DS ignores
├── README.md          # Project documentation
├── data/              # Data files (gitignored)
├── notebooks/         # Jupyter notebooks
├── src/               # Source code
└── models/            # Saved models (gitignored)

~/dockerfiles/
└── my-project.Dockerfile  # Custom image definition

---

See Also

• Container Commands
• Image Commands
• Creating Projects Guide