Search results

Support of AdamW

Status: Open.

#140 In facebookresearch/higher;

· ridiculouz opened on May 6, 2023

Does higher work with huggingface (hugging face, HF) models? e.g. ViT?

Status: Open.

#139 In facebookresearch/higher;

· brando90 opened on Mar 17, 2023

CUDA out of memory

Status: Open.

#138 In facebookresearch/higher;

· aooating opened on Mar 1, 2023

once_differentiable

Status: Open.

#137 In facebookresearch/higher;

· aooating opened on Mar 1, 2023

grad clip correctness

Status: Open.

#136 In facebookresearch/higher;

· Whalefishin opened on Jan 6, 2023

Using higher for hyperparameter optimization

Status: Open.

#135 In facebookresearch/higher;

· aruniyer opened on Dec 21, 2022

Non scalar loss

Status: Open.

#134 In facebookresearch/higher;

· janglinko-dac opened on Dec 14, 2022

Intialize Differentiable Optimizer with non-leaf tensort

Status: Open.

#133 In facebookresearch/higher;

· andrearosasco opened on Aug 8, 2022

higher for dpt architectures?

Status: Open.

#132 In facebookresearch/higher;

· Ainaz99 opened on Jul 22, 2022

How to Update/Optimize subset of params

Status: Open.

#130 In facebookresearch/higher;

· blake-camp opened on Mar 11, 2022

Potential bug with first order MAML using only higher, setting track_higher_grads = False leads to .grad field to not be populated and be None, is that a bug?

Status: Open.

#129 In facebookresearch/higher;

· brando90 opened on Feb 2, 2022

Would making the gradient "data" by detaching them implement first order MAML?

Status: Open.

#128 In facebookresearch/higher;

· brando90 opened on Feb 1, 2022