Qualiopi certified Open-enrolment · In-house · Remote

Training by practitioners
who ship real code.

Every trainer at stat4decision is an active data scientist or consultant. They solve production problems, then teach what they've learnt. No theoretical slides — only applied, tool-first learning built around your team's actual stack.

training_facts.py copy
# what makes our training different

training = {
    "trainers":    "active practitioners",   # not career academics
    "max_group":   7,                       # open-enrolment sessions
    "hands_on":    True,                    # real datasets, real code
    "stack":       "open source only",      # python, R, no vendor lock-in
    "formats":     ["open-enrolment", "in-house", "remote"],
    "certified":   "Qualiopi",              # EU funding eligible
    "materials":   "github.com/stat4decision",
}

Three ways to train with us

All formats use the same trainers, the same materials and the same open-source stack. Pick what fits your organisation.

🏢Format
Open-enrolment

Sessions run on fixed dates at our Paris location. Max 7 participants — enough for discussion, small enough for individual follow-up. Ideal if you have 1–3 people to train.

ParisFixed datesMax 7 peoplePer-seat pricing
🏭Format
In-house

We come to your site. Dates agreed with you. Content adapted to your tools, your data and your business context. Max 8 participants. Priced per session — cost-effective for larger teams.

Your siteFlexible datesMax 8 peoplePer-session pricingCustom content
💻Format
Remote

Full live training via videoconference. Same hands-on approach — participants work in shared coding environments. Available for both open-enrolment and in-house formats.

Video callShared envAny timezoneEU funding eligible

What we teach

All courses use open-source tools. Materials are shared on GitHub after each session.

📊
R for data science
Foundations of data manipulation, visualisation and modelling with R. Tidyverse ecosystem, ggplot2, dplyr. For analysts and scientists moving from Excel or SAS.
3 daystidyverseggplot2dplyrrenv
🔍
Machine learning with R
Supervised and unsupervised learning with the tidymodels framework. Classification, regression, clustering, model evaluation and cross-validation.
3 daystidymodelsparsniprecipesyardstick
🖥️
Shiny app development
Build interactive web applications with R Shiny. From single-file apps to modular production-grade apps. Deployment via Posit Connect or Docker.
2 daysShinymodulesreactlogPosit ConnectDocker
📈
Time-series forecasting with R
ARIMA, ETS, Prophet and neural-network models for forecasting. Confidence intervals, scenario analysis, and model selection. Applied to your business data.
2 daysforecastfableprophettsibble
🎨
Data visualisation with ggplot2
Publication-quality charts with ggplot2. Layered grammar of graphics, custom themes, extensions (ggrepel, patchwork, gganimate). Output to HTML, PDF, Quarto.
2 daysggplot2Quartopatchworkplotly
🔄
SAS to R migration
Hands-on transition for SAS users. Mapping SAS concepts to R equivalents, rewriting PROC steps, validating results, and setting up a reproducible R environment with renv.
3 daystidyverserenvhavenSAS migration
🐍
Python for data science
Practical Python for data work: pandas, numpy, scikit-learn, matplotlib. Environment managed with uv. Exercises built around real datasets. From beginner to production-ready.
3 daysuvpandasscikit-learnmatplotlib
🤖
Generative AI & LLMs in practice
RAG pipeline construction, prompt engineering, open-source LLM deployment (Mistral AI, Llama via Ollama). On-premise setup, data privacy, AI Act awareness.
2 daysRAGLangChainMistral AIOllamavLLM
📱
Streamlit & Dash — data apps
Build and deploy interactive data applications with Streamlit or Dash. From prototype to production: authentication, Docker deployment, GitLab CI integration.
2 daysStreamlitDashPlotlyDockerGitLab CI
✏️
Custom programme
Need something that doesn't exist in our catalogue? We design bespoke programmes around your team's level, tools and business context. Contact us to discuss.
On requestAny topicAny levelYour data
Contact us →

What makes the difference

Training is only useful if participants can apply it immediately. Everything we do is optimised for that.

Practitioners teach
Trainers who debug at 2am
Every trainer is an active consultant. They know what breaks in production, what shortcuts cause technical debt, and which abstractions actually matter. That experience goes directly into the course content.
Real exercises
Your data, your problems
In-house sessions are built around your actual datasets and business context. Open-enrolment sessions use representative real-world data — not toy examples that have no analogue in practice.
Open source only
No proprietary runtime to maintain
We teach Python (with uv), R (with renv), and open LLMs. Environments are reproducible lockfile-based — participants can recreate them months later, on any machine.
Qualiopi certified
EU co-funding available
Our Qualiopi certification means training can be co-funded through European mechanisms — OPCO (skills development plans), CPF, or France Travail. We help you navigate the paperwork.

What to expect

👥Group size
Max 7 (open) / 8 (in-house)

Small groups mean participants can ask questions freely and get individual help during exercises. We cap all sessions strictly.

🕐Schedule
9:30–12:30 · 13:30–17:30

Open-enrolment sessions. In-house: 9:30–13:00 and 14:00–17:30. Confirmed 3 weeks before with location and materials link.

💻Equipment
Laptop required

Participants need a laptop. Loaner machines available on request for open-enrolment sessions. Setup guide sent in advance — environment takes 5 minutes with uv or renv.

📁Materials
GitHub + secure platform

Slides, notebooks and exercises on a secure participant platform. Code and additional materials on GitHub. Everything is available after the session — indefinitely.

💳Payment
Card, bank transfer, PO

Credit card (via PayPal link), bank transfer on purchase order, or direct OPCO billing. Registration up to the working day before, but 1 month recommended for OPCO processing.

Accessibility
Contact us to discuss needs

We work to make all sessions accessible. Disability contact: Coralie André — info@stat4decision.com · +33 1 72 25 40 82.

Ready to upskill your team?

Tell us your team's level, the tools you already use, and what you need to be able to do after the training. We'll propose the right format and dates within 48 hours.

Phone +33 1 72 25 40 82 Email info@stat4decision.com Response within 48h