Les modèles

<mermaid> graph TD

  A[Time Series Models] --> B[Statistical Models]
  A --> C[Machine Learning]
  A --> D[Deep Learning]
  A --> E[Other Approaches]
  
  B --> ARIMA[ARIMA\n(Trend, Non-seasonal)]
  B --> SARIMA[SARIMA\n(Trend, Seasonal)]
  B --> HW[Exponential Smoothing\n(Trend, Seasonal)]
  
  C --> Prophet[Prophet\n(Flexible, Daily observations)]
  C --> GP[Gaussian Processes\n(Flexible regression)]
  C --> HMM[Hidden Markov Models\n(Sequence of states)]
  C --> DTW[Dynamic Time Warping\n(Sequence similarity)]
  
  D --> LSTM[LSTM Networks\n(Long-term dependencies)]
  D --> CNN[Convolutional Neural Networks\n(Patterns in segments)]
  D --> Transformer[Transformer Models\n(Long-range dependencies)]
  
  E --> Note1[Note: Gaussian Processes, Hidden Markov Models, and Dynamic Time Warping are classified under 'Other Approaches' due to their unique methodologies not fitting neatly into the conventional ML or DL categories. They are versatile and can be applied across various contexts.]

  classDef statistical fill:#f9f,stroke:#333,stroke-width:2px;
  classDef machineLearning fill:#ccf,stroke:#333,stroke-width:2px;
  classDef deepLearning fill:#fcf,stroke:#333,stroke-width:2px;
  classDef otherApproaches fill:#cfc,stroke:#333,stroke-width:4px,stroke-dasharray: 5, 5;
  
  class B statistical;
  class C machineLearning;
  class D deepLearning;
  class E otherApproaches;

</mermaid>