Machine Learning Crash Course
Google Machine Learning Crash Course: linear and logistic regression, classification, numerical and categorical data, overfitting, neural networks, embeddings, large language models, fairness, production ML systems and AutoML. With interactive and programming (Colab) exercises. SciTechBiz Institute training based on open materials from Google: text under CC BY 4.0 and code under Apache 2.0.
- Google · CC BY 4.0 (texto) · Apache 2.0 (código)
- Lecciones: 76
- ≈ 14 h
- inglés

Lo que aprenderás
- Prerequisites
- Linear regression
- Logistic regression
- Classification
- Working with numerical data
- Working with categorical data
- Datasets, generalization, and overfitting
- Neural networks
- Embeddings
- Intro to Large Language Models
- Fairness
- Production ML systems
- Automated machine learning
Programa
PrerequisitesLecciones: 1
- Prerequisites and prework
Linear regressionLecciones: 7
- Linear regression
- Linear regression: Loss
- Linear regression: Parameters exercise
- Linear regression: Gradient descent
- Linear regression: Hyperparameters
- Linear regression: Gradient descent exercise
- Linear regression: Programming exercise
Logistic regressionLecciones: 3
- Logistic Regression
- Logistic regression: Calculating a probability with the sigmoid function
- Logistic regression: Loss and regularization
ClassificationLecciones: 7
- Classification
- Thresholds and the confusion matrix
- Classification: Accuracy, recall, precision, and related metrics
- Classification: ROC and AUC
- Classification: Prediction bias
- Classification: Multi-class classification
- Classification: Programming exercise
Working with numerical dataLecciones: 10
- Working with numerical data
- Numerical data: How a model ingests data using feature vectors
- Numerical data: First steps
- Numerical data: Programming exercises
- Numerical data: Normalization
- Numerical data: Binning
- Numerical data: Scrubbing
- Numerical data: Qualities of good numerical features
- Numerical data: Polynomial transforms
- Numerical data: Conclusion
Working with categorical dataLecciones: 5
- Working with categorical data
- Categorical data: Vocabulary and one-hot encoding
- Categorical data: Common issues
- Categorical data: Feature crosses
- Categorical data: Feature cross exercises
Datasets, generalization, and overfittingLecciones: 11
- Datasets, generalization, and overfitting
- Datasets: Data characteristics
- Datasets: Labels
- Datasets: Class-imbalanced datasets
- Datasets: Dividing the original dataset
- Datasets: Transforming data
- Generalization
- Overfitting
- Overfitting: Model complexity
- Overfitting: L2 regularization
- Overfitting: Interpreting loss curves
Neural networksLecciones: 6
- Neural networks
- Neural networks: Nodes and hidden layers
- Neural networks: Activation functions
- Neural Networks: Training using backpropagation
- Neural networks: Interactive exercises
- Neural networks: Multi-class classification
EmbeddingsLecciones: 4
- Embeddings
- Embeddings: Embedding space and static embeddings
- Embeddings: Interactive exercises
- Embeddings: Obtaining embeddings
Intro to Large Language ModelsLecciones: 3
- Introduction to Large Language Models
- LLMs: What's a large language model?
- LLMs: Fine-tuning, distillation, and prompt engineering
FairnessLecciones: 9
- Fairness
- Fairness: Types of bias
- Fairness: Identifying bias
- Fairness: Mitigating bias
- Fairness: Evaluating for bias
- Fairness: Demographic parity
- Fairness: Equality of opportunity
- Fairness: Counterfactual fairness
- Fairness: Programming exercise
Production ML systemsLecciones: 7
- Production ML systems
- Production ML systems: Static versus dynamic training
- Production ML systems: Static versus dynamic inference
- Production ML systems: When to transform data?
- Production ML systems: Deployment testing
- Production ML systems: Monitoring pipelines
- Production ML systems: Questions to ask
Automated machine learningLecciones: 3
- Automated Machine Learning (AutoML)
- AutoML: Benefits and limitations
- AutoML: Getting started
Certificado y tutor de IA
Los miembros estudian estos cursos sin coste adicional. El certificado de aprovechamiento del SciTechBiz Institute y el tutor de IA se incluyen con la licencia.
Ver la licenciaAtribución
- Fuente
- Licencia
- CC BY 4.0 (texto) · Apache 2.0 (código)
- Versión
- web-2026-11-03
Based on the "Machine Learning Crash Course" by Google (developers.google.com; text under CC BY 4.0, code under Apache 2.0)
Google does not endorse or certify this training.
Las fuentes citadas no respaldan ni certifican estos cursos; se usan bajo sus licencias abiertas, con atribución en cada lección.