Skip to content
This repository has been archived by the owner on Aug 31, 2021. It is now read-only.


Repository files navigation

qhoptim: Quasi-hyperbolic optimization

This repository contains PyTorch and TensorFlow implementations of the quasi-hyperbolic momentum (QHM) and quasi-hyperbolic Adam (QHAdam) optimization algorithms from Facebook AI Research.


Use this one-liner for installation:

$ pip install qhoptim

Then, you can instantiate the optimizers in PyTorch:

>>> from qhoptim.pyt import QHM, QHAdam

# something like this for QHM
>>> optimizer = QHM(model.parameters(), lr=1.0, nu=0.7, momentum=0.999)

# or something like this for QHAdam
>>> optimizer = QHAdam(
...     model.parameters(), lr=1e-3, nus=(0.7, 1.0), betas=(0.995, 0.999))

Or in TensorFlow:

>>> from import QHMOptimizer, QHAdamOptimizer

# something like this for QHM
>>> optimizer = QHMOptimizer(
...     learning_rate=1.0, nu=0.7, momentum=0.999)

# or something like this for QHAdam
>>> optimizer = QHAdamOptimizer(
...     learning_rate=1e-3, nu1=0.7, nu2=1.0, beta1=0.995, beta2=0.999)


Please refer to the documentation for installation instructions, usage information, and a Python API reference.

Direct link to installation instructions: here.


QHM and QHAdam were proposed in the ICLR 2019 paper "Quasi-hyperbolic momentum and Adam for deep learning". We recommend reading the paper for both theoretical insights into and empirical analyses of the algorithms.

If you find the algorithms useful in your research, we ask that you cite the paper as follows:

  title={Quasi-hyperbolic momentum and Adam for deep learning},
  author={Jerry Ma and Denis Yarats},
  booktitle={International Conference on Learning Representations},


Bugfixes and contributions are very much appreciated! Please see CONTRIBUTING.rst for more information.


This source code is licensed under the MIT license found in the LICENSE file in the root directory of this source tree.