DjangoCon Europe 2025 Keynote - Django for Data Science

These are the slides from my recent talk on training a machine learning model in a Jupyter notebook and then deploying it with Django. Includes links to all code repos and demos.

Transcript

1. Django for Data Science Will Vincent DjangoCon Europe (April 2025) Deploying Machine Learning Models with Django
3. Image: AlmaBetter.com Image: AlmaBetter.com
5. What is Data Science?
6. What do Data Scientists Do?
8. 1,000,000,000,000,000 1 Quadrillion 10^15 Tokens
11. Train a Model
13. Jupyter Notebook - Setup
15. Iris Dataset
16. Iris Dataset
17. requirements.txt pandas~=2.2.3 scikit-learn~=1.6.1
18. Jupyter Notebook # Load libraries import pandas as pd from sklearn.model_selection import train_test_split from sklearn.metrics import accuracy_score from sklearn.svm import SVC from joblib import dump, load # Load dataset df = pd.read_csv("data/iris.csv") # Extract features (X) and labels (y) X = df[["SepalLengthCm", "SepalWidthCm", "PetalLengthCm", "PetalWidthCm"]].values y = df["Species"].values # Split into training and testing data X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.20, random_state=1 )
19. Jupyter Notebook # Train the model model = SVC(gamma="auto") model.fit(X_train, y_train) # Make predictions on validation dataset predictions = model.predict(X_test) # Evaluate model accuracy accuracy = accuracy_score(y_test, predictions) print(f"Model Accuracy: {accuracy:.2f}") # Print accuracy score # Save model using joblib dump(model, "models/iris.joblib") # Load model using joblib model = load("models/iris.joblib")
20. # Get input from the user try: sepal_length = float(input("Enter SepalLengthCm: ")) sepal_width = float(input("Enter SepalWidthCm: ")) petal_length = float(input("Enter PetalLengthCm: ")) petal_width = float(input("Enter PetalWidthCm: ")) except ValueError: print("Invalid input. Please enter numeric values.") exit() # Make a prediction input_data = [[sepal_length, sepal_width, petal_length, petal_width]] result = model.predict(input_data) print("Prediction:", result[0]) # Iris Classification with SVM
21. Live Demo?
22. Visualize # requirements.txt pandas~=2.2.3 scikit-learn~=1.6.1 seaborn~=0.13.2 matplotlib~=3.10.0 # Import Seaborn and Matplotlib for visualizations import seaborn as sns import matplotlib.pyplot as plt # Pairplot to visualize relationships between features sns.pairplot(df, hue="Species", diag_kind="kde") plt.show()
25. Django Project
26. Gameplan • Create new Django project • Load joblib file • Forms for users to enter predictions -> see results • Store user info in the database • Deployment?
27. DjangoForDataScience.com
30. ├── django_project │ ├── __init__.py │ ├── asgi.py │ ├── settings.py │ ├── urls.py │ └── wsgi.py ├── manage.py ├── predict │ ├── __init__.py │ ├── admin.py │ ├── apps.py │ ├── migrations │ │ └── __init__.py │ ├── models.py │ ├── tests.py │ └── views.py └── templates
32. ├── django_project │ ├── __init__.py │ ├── asgi.py │ ├── settings.py │ ├── urls.py │ └── wsgi.py ├── iris.joblib # hi there! ├── manage.py ├── predict │ ├── __init__.py │ ├── admin.py │ ├── apps.py │ ├── migrations │ │ └── __init__.py │ ├── models.py │ ├── tests.py │ └── views.py └── templates
33. URLs # django_project/urls.py from django.contrib import admin from django.urls import path, include # new urlpatterns = [ path("admin/", admin.site.urls), path("", include("predict.urls")), # new ] # predict/urls.py from django.urls import path from .views import predict urlpatterns = [ path("", predict, name="predict"),
34. View + Template # predict/views.py from django.shortcuts import render def predict(request): return render(request, "predict.html") <!-- templates/predict.html --> Hello
35. Runserver (again)
36. # predict/views.py import joblib import numpy as np from django.shortcuts import render # Load the trained model model = joblib.load("iris.joblib") def predict(request): prediction = None if request.method == "POST": try: # Get input values from the form sepal_length = float(request.POST.get("sepal_length")) sepal_width = float(request.POST.get("sepal_width")) petal_length = float(request.POST.get("petal_length")) petal_width = float(request.POST.get("petal_width")) # Make a prediction features = np.array( [[sepal_length, sepal_width, petal_length, petal_width]] ) prediction = model.predict(features)[0] except Exception as e: prediction = f"Error: {str(e)}" return render(request, "predict.html", {"prediction": prediction})
37. <!-- templates/predict.html --> <!DOCTYPE html> <html lang="en"> <head> <meta charset="UTF-8"> <meta name="viewport" content="width=device-width, initial-scale=1.0"> <title>Iris Prediction</title> <style> body { font-family: Arial, sans-serif; max-width: 400px; margin: 50px auto; text-align: center; } form { display: flex; flex-direction: column; gap: 10px; } input, button { padding: 10px; font-size: 16px; } button { cursor: pointer; background-color: #4CAF50; color: white; border: none; }
38. <!-- templates/predict.html --> ... <body> <h2>Iris Species Predictor</h2> <form method="post"> {% csrf_token %} <input type="number" step="any" name="sepal_length" placeholder="Sepal Length" required> <input type="number" step="any" name="sepal_width" placeholder="Sepal Width" required> <input type="number" step="any" name="petal_length" placeholder="Petal Length" required> <input type="number" step="any" name="petal_width" placeholder="Petal Width" required> <button type="submit">Predict</button> </form> {% if prediction %} <h3>Prediction: {{ prediction }}</h3> {% endif %} </body> </html>
41. Store User Inputs # predict/views.py ... def predict(request): prediction = None form_data = {} # To store user inputs if request.method == "POST": try: # Store form inputs form_data = { "sepal_length": request.POST.get("sepal_length"), "sepal_width": request.POST.get("sepal_width"), "petal_length": request.POST.get("petal_length"), "petal_width": request.POST.get("petal_width"), } # Convert inputs to float and make a prediction features = np.array([[float(v) for v in form_data.values()]]) prediction = model.predict(features)[0] except Exception as e: prediction = f"Error: {str(e)}" return render( request, "predict.html", {"prediction": prediction, "form_data": form_data}
42. Display User Inputs <!-- templates/predict.html --> {% if prediction %} <div class="result"> <h3>Prediction: {{ prediction }}</h3> <p><strong>Inputs:</strong></p> <ul> <li>Sepal Length: {{ form_data.sepal_length }}</li> <li>Sepal Width: {{ form_data.sepal_width }}</li> <li>Petal Length: {{ form_data.petal_length }}</li> <li>Petal Width: {{ form_data.petal_width }}</li> </ul> </div>] {% endif %}
44. Add a Model to Store Predictions # predict/models.py from django.db import models class IrisPrediction(models.Model): sepal_length = models.FloatField() sepal_width = models.FloatField() petal_length = models.FloatField() petal_width = models.FloatField() prediction = models.CharField(max_length=100) created_at = models.DateTimeField(auto_now_add=True) def __str__(self): return f"{self.prediction} ({self.created_at.strftime('%Y-%m-%d %H:%M:%S')})"
45. # predict/views.py import joblib import numpy as np from django.shortcuts import render from .models import IrisPrediction. # new # Load the trained model model = joblib.load("iris.joblib") def predict(request): prediction = None form_data = {} if request.method == "POST": try: ... # Save prediction to database IrisPrediction.objects.create( sepal_length=form_data["sepal_length"], sepal_width=form_data["sepal_width"], petal_length=form_data["petal_length"], petal_width=form_data["petal_width"], prediction=prediction, )
46. Update Admin # predict/admin.py from django.contrib import admin from .models import IrisPrediction admin.site.register(IrisPrediction)
47. Update Admin
48. Deployment...
49. Deployment Checklist • configure static files and install WhiteNoise • add environment variables with environs • create a .env file and update the .gitignore file • update DEBUG, ALLOWED_HOSTS, SECRET_KEY, and CSRF_TRUSTED_ORIGINS • update DATABASES to run PostgreSQL in production and install psycopg • install Gunicorn as a production WSGI server • create a Procfile • update the requirements.txt file • create a new Heroku project, push the code to it, and start a dyno web process
50. Takeaways ✅ Django is great for deploying ML models (web UI + APIs) ✅ Iris dataset is simple but powerful for learning ML deployment ✅ Try deploying a real-world ML model
51. Questions? @wsvincent.bsky.social @wsvincent -- django_irisml & iris_ml fosstodon.org/@wsvincent @william-s-vincent