Data science and Machine learning Makefile for Python projects or in general :)

Makefile

# Thank you @Earthly https://www.youtube.com/watch?v=w2UeLF7EEwk
# Can be adapted to pipenv, and poetry
# Other languages coming soon especially R and Julia

# .ONESHELL tells make to run each recipe line in a single shell    
.ONESHELL:
.DEFAULT_GOAL := help

# --- Configuration ---
PYTHON := .venv/bin/python3
PIP := .venv/bin/pip3
DOCKER_IMAGE_NAME    := test_app
DOCKER_IMAGE_VERSION := v0.0.0
DOCKER_IMAGE_TAG     := $(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_VERSION)
REGISTRY_URL         := registry.digitalocean.com/my-registry

# ==============================================================================
# 1. PACKAGE MANAGER DETECTION & ROUTING
# ==============================================================================
# Priority: uv > pipenv > poetry > standard venv/pip

DETECT_UV     := $(shell command -v uv 2> /dev/null)
DETECT_PIPENV := $(shell command -v pipenv 2> /dev/null)
DETECT_POETRY := $(shell command -v poetry 2> /dev/null)

ifeq ($(DETECT_UV),)
  HAS_UV := 0
else
  HAS_UV := 1
endif

ifeq ($(DETECT_PIPENV),)
  HAS_PIPENV := 0
else
  HAS_PIPENV := 1
endif

ifeq ($(DETECT_POETRY),)
  HAS_POETRY := 0
else
  HAS_POETRY := 1
endif

# Route commands based on availability
ifeq ($(HAS_UV),1)
  PACKAGE_MANAGER := uv
  INSTALL_CMD     := uv pip install -e . -r requirements.txt
  RUN_CMD         := uv run
  LINT_CMD        := uv run ruff check .
else ifeq ($(HAS_PIPENV),1)
  PACKAGE_MANAGER := pipenv
  INSTALL_CMD     := pipenv install --dev
  RUN_CMD         := pipenv run
  LINT_CMD        := pipenv run ruff check .
else ifeq ($(HAS_POETRY),1)
  PACKAGE_MANAGER := poetry
  INSTALL_CMD     := poetry install
  RUN_CMD         := poetry run
  LINT_CMD        := poetry run ruff check .
else
  PACKAGE_MANAGER := pip (standard venv)
  INSTALL_CMD     := .venv/bin/pip install --upgrade pip && .venv/bin/pip install -r requirements.txt
  RUN_CMD         := .venv/bin/python
  LINT_CMD        := .venv/bin/python -m ruff check .
endif

# ==============================================================================
# 2. ENVIRONMENT & SETUP
# ==============================================================================

.PHONY: detect install clean setup_readme

detect: ## Detect and show the selected package manager
	@echo "🔍 Package manager selected: $(PACKAGE_MANAGER)"
	@echo "   UV detected: $(HAS_UV)"
	@echo "   Pipenv detected: $(HAS_PIPENV)"
	@echo "   Poetry detected: $(HAS_POETRY)"

install: ## Install project dependencies
	@echo "📦 Installing dependencies using $(PACKAGE_MANAGER)..."
	@if [ "$(PACKAGE_MANAGER)" = "pip (standard venv)" ] && [ ! -d ".venv" ]; then \
		python3 -m venv .venv; \
	fi
	$(INSTALL_CMD)
	@echo "✅ Installation complete."

clean: ## Clean cache directories and virtual environments
	@echo "🧹 Cleaning project..."
	rm -rf __pycache__ utils/__pycache__ testing/__pycache__
	rm -rf .pytest_cache .ruff_cache .venv
	@echo "✅ Clean complete."

setup_readme: ## Create a boilerplate README.md if one doesn't exist
	@if [ ! -f README.md ]; then \
		echo "# Project Name\n\nDescription of the project.\n\n## Installation\n\n## Usage\n\n## License" > README.md; \
		echo "✅ README.md created."; \
	else \
		echo "⚠️ README.md already exists."; \
	fi

# ==============================================================================
# 3. CODE QUALITY & TESTING
# ==============================================================================

.PHONY: format lint ruff-fix docstring test run

format: ## Format Python code using black
	@echo "🎨 Formatting code..."
	$(RUN_CMD) black *.py utils/*.py testing/*.py || echo "⚠️ Black not found, skipping."

lint: ## Lint code using Ruff and Pylint
	@echo "🔎 Linting code..."
	$(LINT_CMD)
	$(RUN_CMD) pylint --disable=R,C --errors-only *.py utils/*.py testing/*.py || echo "⚠️ Pylint not found, skipping."

ruff-fix: ## Run Ruff with autofix enabled
	@echo "🔧 Fixing code with Ruff..."
	$(LINT_CMD) --fix

docstring: ## Format docstrings using pyment (NumPy style)
	@echo "📝 Formatting docstrings..."
	$(RUN_CMD) pyment -w -o numpydoc *.py

test: ## Run tests using pytest
	@echo "🧪 Running tests..."
	$(RUN_CMD) pytest testing/

run: ## Run the main application (Example: app.py)
	@echo "🚀 Running application..."
	$(RUN_CMD) app.py

# ==============================================================================
# 4. UX & SYSTEM AUDITS (PyConKE '26 Framework)
# 
# ==============================================================================

.PHONY: audit-ux check-certainty check-relief check-calm check-speed

audit-ux: check-certainty check-relief check-calm check-speed ## Run Edmin M's 4-Quadrant UX Audit
	@echo "\n✅ UX Audit Complete. Engineering is hard, but the user feels nothing but magic."
	@echo "==============================================================================\n"

check-certainty:
	@echo "\n[ Quadrant I - CERTAINTY ]"
	@echo " - Idempotency: If the pipeline retries, do we get duplicate rows or a clean MERGE?"
	@echo " - Consistency: If the user double-clicks 'Submit', does it mutate state twice?"

check-relief:
	@echo "\n[ Quadrant II - RELIEF ]"
	@echo " - Degradation: If the primary LLM fails, does the UI crash or fallback smoothly?"
	@echo " - State Recovery: If validation fails, is the user's configuration preserved?"

check-calm:
	@echo "\n[ Quadrant III - CALM ]"
	@echo " - Resource Cleanup: If the socket closes mid-query, do compute resources stop?"
	@echo " - GPU Cycles: Are cancellation tokens propagated to PyTorch inference queues?"

check-speed:
	@echo "\n[ Quadrant IV - SPEED ]"
	@echo " - Payload Size: Are we sending raw geometries or optimized payloads?"
	@echo " - Edge Constraints: How quickly does the edge device return TTFT?"

# ==============================================================================
# 5. PEER REVIEW & XAI PROMPTS
# ==============================================================================

.PHONY: review-code review-ds review-logging review-xai

review-code: ## Display general code review heuristics
	@echo "#### General Code Review ####"
	@echo "1. Architecture (SOLID, DRY, KISS)"
	@echo "2. Reliability (Failover, Thread Safety)"
	@echo "3. Performance (Algorithmic efficiency, Memory I/O, Caching)"

review-ds: ## Display Data Science specific review heuristics
	@echo "#### Data Science Review ####"
	@echo "1. Pipeline: Schema validation, Leakage checks, Imputation strategy."
	@echo "2. Modeling: Hyperparameters, CV strategy, Explainability (SHAP/LIME)."
	@echo "3. Production: TTFT, Drift Monitoring, Canary deployment readiness."
	@echo "4. Tracking: MLflow/W&B logging, artifact versioning."

review-logging: ## Display Logging/Observability heuristics
	@echo "#### Logging Review ####"
	@echo "1. Are we using structured (JSON) logging instead of raw print()?"
	@echo "2. Are log levels (DEBUG, INFO, ERROR) appropriate?"
	@echo "3. Is sensitive data masked?"
	@echo "4. Are distributed trace IDs propagated across service boundaries?"

review-xai: ## Display Explainability and Interpretability (XAI) guide
	@echo "#### Explainability & Interpretability (XAI) Check ####"
	@echo "Ref: PyConKE 2026 - Brian M (Building ML Pipelines)"
	@echo ""
	@echo "1. Baseline / Intrinsic Interpretability:"
	@echo "   - Btw, did you check intrinsic feature importance first?"
	@echo "   - (e.g., Have you looked at Gini impurity in your Random Forest before reaching for complex tools?)"
	@echo ""
	@echo "2. SHAP (Global & Local Explainability):"
	@echo "   - Btw, did you run Permutation or Kernel SHAP for your model?"
	@echo "   - Did you visualize the outputs using a SHAP beeswarm plot? "
	@echo "     (Crucial for seeing both the magnitude AND direction of feature impacts on the model)"
	@echo ""
	@echo "3. LIME (Local Surrogate Models):"
	@echo "   - Btw, did you use LIME to sample and fit a local surrogate model?"
	@echo "   - Did you look at how the model behaves at the specific decision boundary for edge-case predictions?"
	@echo ""
	@echo "4. Delivery & Audience Check:"
	@echo "   - Btw, did you explicitly ask the stakeholders how technical they are before presenting this?"
	@echo "   - (Don't present a beeswarm plot to an audience that just needs the Gini impurity summary!)"

# ==============================================================================
# 6. DOCKER OPERATIONS
# ==============================================================================

.PHONY: docker-build docker-test docker-clean docker-run docker-push docker-info

docker-build: Dockerfile ## Build the Docker container
	@echo "🐳 Building image: $(DOCKER_IMAGE_TAG)"
	docker build --platform linux/amd64 -t $(DOCKER_IMAGE_TAG) .

docker-test: Dockerfile ## Lint the Dockerfile using Hadolint
	@echo "🐳 Linting Dockerfile..."
	docker run --rm -i hadolint/hadolint < Dockerfile

docker-clean: ## Remove dangling images, volumes, and networks
	@echo "🐳 Pruning Docker system..."
	docker system prune -a

docker-run: docker-build ## Run the built Docker container
	@echo "🐳 Running container..."
	docker run --platform linux/amd64 -p 8080:8080 $(DOCKER_IMAGE_TAG)

docker-push: docker-build ## Push image to the configured registry
	@echo "🐳 Pushing $(DOCKER_IMAGE_TAG) to $(REGISTRY_URL)..."
	docker tag $(DOCKER_IMAGE_TAG) $(REGISTRY_URL)/$(DOCKER_IMAGE_TAG)
	docker push $(REGISTRY_URL)/$(DOCKER_IMAGE_TAG)

docker-info: ## Show Docker image size and build history
	@echo "🐳 Image size for $(DOCKER_IMAGE_TAG):"
	@docker image ls $(DOCKER_IMAGE_TAG) --format "table {{.Repository}}:{{.Tag}}\t{{.Size}}\t{{.CreatedSince}}"
	@echo "\n🐳 Layer history:"
	@docker history $(DOCKER_IMAGE_TAG)

# ==============================================================================
# 7. HELP (Self-Documenting) & ALL
# ==============================================================================

.PHONY: help all

help: ## Show this help message
	@echo "Usage: make [target]\n"
	@echo "Available targets:"
	@awk 'BEGIN {FS = ":.*##"; printf "\033[1m%-20s\033[0m %s\n", "Target", "Description"} \
	/^[a-zA-Z_-]+:.*?##/ { printf "\033[36m%-20s\033[0m %s\n", $$1, $$2 }' $(MAKEFILE_LIST)

all: install format lint test run docker-build docker-run docker-push ## Run the full CI/CD pipeline (Install -> Test -> Run -> Dockerize)
	@echo "🚀 Full pipeline execution complete!"

Template for Julia needs edits

# Thank you @Earthly https://www.youtube.com/watch?v=w2UeLF7EEwk
# Can be adapted to pipenv, and poetry
# Other languages coming soon especially R and Julia

# .ONESHELL tells make to run each recipe line in a single shell
.ONESHELL:

# .DEFAULT_GOAL tells make which target to run when no target is specified
.DEFAULT_GOAL := all

# Specify julia version virtual environment
venv/bin/activate: 
	# create virtual environment
	julia -e 'using Pkg; Pkg.add("VirtualEnvironments"); using VirtualEnvironments; VirtualEnvironments.build()'
  
activate:
	# activate virtual environment
	julia -e 'using Pkg; Pkg.activate("."); Pkg.instantiate()'

install: venv/bin/activate  # prerequisite
	# install common packages
	julia -e 'using Pkg; Pkg.add("PyCall"); Pkg.add("PyPlot"); Pkg.add("Statistics")
    
docstring: activate
	# format docstrings
	julia -e 'using Pkg; Pkg.add("Documenter"); using Documenter; Documenter.generate("docs")'
  
format: activate 
	# format code
	julia -e 'using Pkg; Pkg.add("JuliaFormatter"); using JuliaFormatter; format("src")'
  
clean:
	# clean directory of cache files
	julia -e 'using Pkg; Pkg.gc()'

lint: activate install format
	# lint code
	julia -e 'using Pkg; Pkg.add("Lint"); using Lint; lint("src")'
  
test: activate install format
	# test code
	julia -e 'using Pkg; Pkg.add("Test"); using Test; include("test/runtests.jl")'
  
run: activate install format
	# run julia script
	julia src/main.jl
  
docker_build: Dockerfile
	# build julia container
	#docker build -t plot-timeseries-app:v0 .
  
docker_run_test: Dockerfile
	# linting Dockerfile
	docker run --rm -i hadolint/hadolint < Dockerfile

docker_clean: Dockerfile docker_build
	# remove dangling images, containers, volumes and networks
	docker system prune -a
  
docker_run: build
	# run docker
	# docker run -e ENDPOINT_URL -e SECRET_KEY -e SPACES_ID -e SPACES_NAME plot-timeseries-app:v0

docker_push: docker_build
  # push to registry
  # docker tag <my-image> registry.digitalocean.com/<my-registry>/<my-image>
  # docker push registry.digitalocean.com/<my-registry>/<my-image>  


# .PHONY tells make that these targets do not represent actual files
.PHONY: activate format clean lint test build run docker_build docker_run docker_push docker_clean docker_run_test
  
all: install format lint test run docker_build docker_run docker_push

An attempt for java script

# Thank you @Earthly https://www.youtube.com/watch?v=w2UeLF7EEwk
# Can be adapted to pipenv, and poetry
# Other languages coming soon especially R and Julia

# .ONESHELL tells make to run each recipe line in a single shell
.ONESHELL:

# .DEFAULT_GOAL tells make which target to run when no target is specified
.DEFAULT_GOAL := all

# Specify node location in virtual environment
NODE := node

# Specify npm location in virtual environment
NPM := npm

# Specify Docker container name
DOCKER_CONTAINER_NAME := test_app:v0.0.0

venv/bin/activate: requirements.txt
    # create virtual environment
    npm install -g venv
    # make command executable
    chmod +x venv/bin/activate
    # activate virtual environment
    . venv/bin/activate

activate:
    # activate virtual environment
    . venv/bin/activate

install: venv/bin/activate requirements.txt # prerequisite
    # install commands
    npm ci

docstring: activate
    # format docstring
    javascriptdoc -w -o numpydoc *.js

format: activate 
    # format code
    prettier --write *.js utils/*.js testing/*.js

clean:
    # clean directory of cache
    rm -rf __pycache__ &&\
    rm -rf utils/__pycache__ &&\
    rm -rf testing/__pycache__ &&\
    rm -rf .pytest_cache &&\
    rm -rf .venv

lint: activate install format
    # eslint or #jshint
    # see the errors with
    npm run lint
    # fix the errors with npm run lint:fix
setup_readme:  ## Create a README.md
    @if [ ! -f README.md ]; then \
        echo "# Project Name\n\
Description of the project.\n\n\
## Installation\n\
- Step 1\n\
- Step 2\n\n\
## Usage\n\
Explain how to use the project here.\n\n\
## Contributing\n\
Explain how to contribute to the project.\n\n\
## License\n\
License information." > README.md; \
        echo "README.md created."; \
    else \
        
test: activate install format
    # test
    npm test

run: activate install format lint
    # run application
    # example node app.js

docker_build: Dockerfile
    # build container
    docker build -t plot-timeseries-app:v0 .

docker_run_test: Dockerfile
    # linting Dockerfile
    docker run --rm -i hadolint/hadolint < Dockerfile

docker_clean: Dockerfile
    # remove dangling images, containers, volumes and networks
    docker system prune -a

docker_run: Dockerfile docker_build
    # run docker
    # docker run -e ENDPOINT_URL -e SECRET_KEY -e SPACES_ID -e SPACES_NAME plot-timeseries-app:v0

docker_push: docker_build
  # push to registry
  # docker tag <my-image> registry.digitalocean.com/<my-registry>/<my-image>
  # docker push registry.digitalocean.com/<my-registry>/<my-image>  
  
 
# .PHONY tells make that these targets do not represent actual files
.PHONY: activate format clean lint test build run docker_build docker_run docker_push docker_clean docker_run_test
  
all: install format lint test run docker_build docker_run docker_push

For rust

# Makefile for a Rust project
.PHONY: help build build_release run run_release test lint fmt doc clean check_all docker_build docker_run docker_lint

# Default target to show help
.DEFAULT_GOAL := help

TARGET_NAME := $(shell basename $(CURDIR)) # Assumes project name is current directory name
TARGET_PATH_DEBUG := target/debug/$(TARGET_NAME)
TARGET_PATH_RELEASE := target/release/$(TARGET_NAME)

help: ## Show this help message
    @echo "Available targets for Rust project:"
    @grep -E '^[a-zA-Z0-9_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\\033[36m%-20s\\033[0m %s\\n", $$1, $$2}'

build: ## Build the Rust project (debug mode)
    @echo "Building $(TARGET_NAME) (debug)..."
    cargo build

build_release: ## Build the Rust project (release mode)
    @echo "Building $(TARGET_NAME) (release)..."
    cargo build --release

run: build ## Run the debug build of the project
    @echo "Running $(TARGET_NAME) (debug)..."
    cargo run

run_release: build_release ## Run the release build of the project
    @echo "Running $(TARGET_NAME) (release)..."
    cargo run --release
    # Or directly: $(TARGET_PATH_RELEASE)

test: ## Run tests
    @echo "Running tests..."
    cargo test

lint: ## Lint code using clippy (errors on warnings)
    @echo "Linting with clippy..."
    cargo clippy --all-targets --all-features -- -D warnings

fmt: ## Format code using rustfmt
    @echo "Formatting with rustfmt..."
    cargo fmt --all

doc: ## Generate and open documentation
    @echo "Generating documentation..."
    cargo doc --open

clean: ## Clean build artifacts
    @echo "Cleaning build artifacts..."
    cargo clean

check_all: fmt lint test ## Run all checks (format, lint, test)
    @echo "All checks passed."

# --- Docker Targets (Optional) ---
DOCKER_IMAGE_NAME ?= $(TARGET_NAME)-app
DOCKER_IMAGE_TAG ?= latest
DOCKERFILE ?= Dockerfile

docker_build: ## Build the Docker image
ifeq (,$(wildcard $(DOCKERFILE)))
    @echo "Dockerfile not found. Skipping docker_build."
else
    @echo "Building Docker image $(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG)..."
    docker build -t $(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG) -f $(DOCKERFILE) .
endif

docker_run: ## Run the app in a Docker container (requires docker_build)
ifeq (,$(wildcard $(DOCKERFILE)))
    @echo "Dockerfile not found. Skipping docker_run."
else
    @echo "Running Docker container $(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG)..."
    docker run --rm -it $(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG)
endif

docker_lint: ## Lint the Dockerfile using hadolint
ifeq (,$(wildcard $(DOCKERFILE)))
    @echo "Dockerfile not found. Skipping docker_lint."
else
    @echo "Linting $(DOCKERFILE) with hadolint..."
    docker run --rm -i hadolint/hadolint < $(DOCKERFILE)
endif