torch-dxdt

Contents

  • Installation
    • From PyPI
    • From Source
    • Development Installation
    • Requirements
    • Conda Environment
    • Verifying Installation
  • Quick Start
    • Basic Usage
      • Functional Interface
      • Object-Oriented Interface
    • Available Methods
    • Working with Gradients
    • Batched Processing
    • Smoothing
    • Next Steps
  • Differentiation Methods
    • Finite Difference
      • Theory
      • Usage
      • Parameters
      • When to Use
    • Savitzky-Golay Filter
      • Theory
      • Usage
      • Parameters
      • Padding Mode Details
      • When to Use
    • Spectral Differentiation
      • Theory
      • Usage
      • Parameters
      • When to Use
      • Note
    • Spline Smoothing
      • Theory
      • Usage
      • Parameters
      • When to Use
    • Kernel (Gaussian Process)
      • Theory
      • Usage
      • Parameters
      • When to Use
      • Note
    • Kalman Smoother
      • Theory
      • Usage
      • Parameters
      • When to Use
    • Whittaker-Eilers Smoother
      • Theory
      • Usage
      • Parameters
      • When to Use
    • Method Comparison
    • Boundary Behavior
      • Why Boundaries Matter
      • Configuring Boundary Padding
      • Recommendations for Boundary-Sensitive Applications
  • API Reference
    • Functional Interface
      • dxdt()
      • dxdt_orders()
      • smooth_x()
      • available_methods()
    • Base Class
      • Derivative
        • Derivative.d()
        • Derivative.d_orders()
        • Derivative.smooth()
    • Differentiation Methods
      • Finite Difference
        • FiniteDifference
      • Savitzky-Golay
        • SavitzkyGolay
      • Spectral
        • Spectral
      • Spline
        • Spline
      • Kernel
        • Kernel
      • Kalman
        • Kalman
      • Whittaker
        • Whittaker
  • Examples
    • Jupyter Notebook
    • Basic Usage
      • Computing Derivatives of Smooth Functions
      • Handling Noisy Data
    • Autodiff Integration
      • Training a Neural Network with Derivative Loss
      • Computing Second Derivatives
      • Computing Multiple Derivative Orders Efficiently
    • GPU Acceleration
      • Moving Computations to GPU
    • Batch Processing
      • Processing Multiple Signals
    • Smoothing Data
      • Using smooth_x for Denoising
    • Class-Based Interface
      • Using Method Classes Directly
torch-dxdt
  • Overview: module code

All modules for which code is available

  • torch_dxdt
    • torch_dxdt.base
    • torch_dxdt.finite_difference
    • torch_dxdt.kalman
    • torch_dxdt.kernel
    • torch_dxdt.savitzky_golay
    • torch_dxdt.spectral
    • torch_dxdt.spline_method
    • torch_dxdt.whittaker

© Copyright 2024, Maximilian Stoelzle.

Built with Sphinx using a theme provided by Read the Docs.