StackOverflow Questions for Tag: automatic-differentiation

Zoe Zhao

Reputation: 11

Why do custom backward functions and torch.autograd yield different results after network training?

pytorchneural-networkbackpropagationautomatic-differentiation

Score: 1

Answers: 1

Jimut123

Reputation: 518

Any faster and memory-efficient alternative of torch.autograd.functional.jacobian(model.decoder, latent_l)?

pythonpytorchautomatic-differentiation

Score: 0

Answers: 1

Peter234

Reputation: 1052

How to use torch.unique to filter duplicate values, calculate an expensive function, map it back, and then calculate the gradient?

pytorchautomatic-differentiation

Score: 0

Answers: 1

Moody

Reputation: 1417

Difference between symbolic differentiation and automatic differentiation?

symbolic-mathautomatic-differentiation

Score: 43

Answers: 4

Julier

Reputation: 25

The analogue of torch.autograd in TensorFlow

tensorflowpytorchneural-networkautomatic-differentiation

Score: 0

Answers: 1

Andreas Schuldei

Reputation: 405

How to Properly Track Gradients with MyGrad When Using Scipy's RectBivariateSpline for Interpolation?

pythonnumpyscipyautomatic-differentiation

Score: 1

Answers: 1

James Yong

Reputation: 31

How to handle errors caused by non-JAX objects (such as scipy.sparse.csr_matrix) in JAX custom_vjp?

pythonscipyjaxautomatic-differentiation

Score: 1

Answers: 1

Nick Alger

Reputation: 1104

Forward-over-reverse mode Hessian-vector product in Jax: how smart is jax.jvp at re-using computations?

pythonjaxautomatic-differentiation

Score: 1

Answers: 0

Samuel Beaussant

Reputation: 3

Discrepancy in BatchNorm2d Gradient Calculation Between TensorFlow and PyTorch

tensorflowpytorchautomatic-differentiation

Score: 0

Answers: 1

HaineBaie

Reputation: 1

How to compute Higher-Order Gradients Across Two Networks in PyTorch? (ML, Autograd)

pytorchgradientautogradautomatic-differentiationautodiff

Score: 0

Answers: 0

Shawn

Reputation: 89

Taking derivatives with multiple inputs in JAX

pythonmachine-learningdeep-learningjaxautomatic-differentiation

Score: 1

Answers: 1

Nsss123

Reputation: 11

Error in automatic derivative calculation with pytorch

pythondeep-learningpytorchneural-networkautomatic-differentiation

Score: 1

Answers: 1

Jingyang Wang

Reputation: 323

JAX custom_jvp with 'None' output leads to TypeError

pythontypeerrorjaxautomatic-differentiation

Score: 1

Answers: 1

liang alex

Reputation: 1

In PyTorch, when using loss.backward() to compute gradients, how can I prevent it from overriding the gradients I've manually computed?

deep-learninggradientloss-functionautomatic-differentiation

Score: 0

Answers: 0

R S

Reputation: 11

Derivative of Gaussian Blur with respect to the pixels

pythonmachine-learningdeep-learninggaussianblurautomatic-differentiation

Score: 1

Answers: 0

Enigmatisms

Reputation: 175

Whether there is any need to modify the backward function in pytorch?

pytorchautomatic-differentiation

Score: 0

Answers: 1

0xbadf00d

Reputation: 18218

Computing the Jacobian of an image: How to reshape the numpy array properly?

pythonnumpyneural-networkjaxautomatic-differentiation

Score: 2

Answers: 1

fmg

Reputation: 925

Update step in PyTorch implementation of Newton's method

pythonmathematical-optimizationpytorchnewtons-methodautomatic-differentiation

Score: 3

Answers: 2

0xbadf00d

Reputation: 18218

How to generate jacobian of a tensor-valued function using torch.autograd?

pythonpytorchautogradautomatic-differentiation

Score: 0

Answers: 1

klpskp

Reputation: 61

Calculating two gradients in pytorch and reusing an intermediate gradient

pythonpytorchautogradautomatic-differentiation

Score: 4

Answers: 2

PreviousPage 1Next

StackOverflow Questions for Tag: automatic-differentiation

Why do custom backward functions and torch.autograd yield different results after network training?

Any faster and memory-efficient alternative of torch.autograd.functional.jacobian(model.decoder, latent_l)?

How to use torch.unique to filter duplicate values, calculate an expensive function, map it back, and then calculate the gradient?

Difference between symbolic differentiation and automatic differentiation?

The analogue of torch.autograd in TensorFlow

How to Properly Track Gradients with MyGrad When Using Scipy&#39;s RectBivariateSpline for Interpolation?

How to handle errors caused by non-JAX objects (such as scipy.sparse.csr_matrix) in JAX custom_vjp?

Forward-over-reverse mode Hessian-vector product in Jax: how smart is jax.jvp at re-using computations?

Discrepancy in BatchNorm2d Gradient Calculation Between TensorFlow and PyTorch

How to compute Higher-Order Gradients Across Two Networks in PyTorch? (ML, Autograd)

Taking derivatives with multiple inputs in JAX

Error in automatic derivative calculation with pytorch

JAX custom_jvp with &#39;None&#39; output leads to TypeError

In PyTorch, when using loss.backward() to compute gradients, how can I prevent it from overriding the gradients I&#39;ve manually computed?

Derivative of Gaussian Blur with respect to the pixels

Whether there is any need to modify the backward function in pytorch?

Computing the Jacobian of an image: How to reshape the numpy array properly?

Update step in PyTorch implementation of Newton&#39;s method

How to generate jacobian of a tensor-valued function using torch.autograd?

Calculating two gradients in pytorch and reusing an intermediate gradient

How to Properly Track Gradients with MyGrad When Using Scipy's RectBivariateSpline for Interpolation?

JAX custom_jvp with 'None' output leads to TypeError

In PyTorch, when using loss.backward() to compute gradients, how can I prevent it from overriding the gradients I've manually computed?

Update step in PyTorch implementation of Newton's method