torch.optim¶
torch.optim
is a package implementing various optimization algorithms.
Most commonly used methods are already supported, and the interface is general
enough, so that more sophisticated ones can be also easily integrated in the
future.
How to use an optimizer¶
To use torch.optim
you have to construct an optimizer object, that will hold
the current state and will update the parameters based on the computed gradients.
Constructing it¶
To construct an Optimizer
you have to give it an iterable containing the
parameters (all should be Variable
s) to optimize. Then,
you can specify optimizerspecific options such as the learning rate, weight decay, etc.
Note
If you need to move a model to GPU via .cuda()
, please do so before
constructing optimizers for it. Parameters of a model after .cuda()
will
be different objects with those before the call.
In general, you should make sure that optimized parameters live in consistent locations when optimizers are constructed and used.
Example:
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
optimizer = optim.Adam([var1, var2], lr=0.0001)
Perparameter options¶
Optimizer
s also support specifying perparameter options. To do this, instead
of passing an iterable of Variable
s, pass in an iterable of
dict
s. Each of them will define a separate parameter group, and should contain
a params
key, containing a list of parameters belonging to it. Other keys
should match the keyword arguments accepted by the optimizers, and will be used
as optimization options for this group.
Note
You can still pass options as keyword arguments. They will be used as defaults, in the groups that didn’t override them. This is useful when you only want to vary a single option, while keeping all others consistent between parameter groups.
For example, this is very useful when one wants to specify perlayer learning rates:
optim.SGD([
{'params': model.base.parameters()},
{'params': model.classifier.parameters(), 'lr': 1e3}
], lr=1e2, momentum=0.9)
This means that model.base
’s parameters will use the default learning rate of 1e2
,
model.classifier
’s parameters will use a learning rate of 1e3
, and a momentum of
0.9
will be used for all parameters.
Taking an optimization step¶
All optimizers implement a step()
method, that updates the
parameters. It can be used in two ways:
optimizer.step()
¶
This is a simplified version supported by most optimizers. The function can be
called once the gradients are computed using e.g.
backward()
.
Example:
for input, target in dataset:
optimizer.zero_grad()
output = model(input)
loss = loss_fn(output, target)
loss.backward()
optimizer.step()
optimizer.step(closure)
¶
Some optimization algorithms such as Conjugate Gradient and LBFGS need to reevaluate the function multiple times, so you have to pass in a closure that allows them to recompute your model. The closure should clear the gradients, compute the loss, and return it.
Example:
for input, target in dataset:
def closure():
optimizer.zero_grad()
output = model(input)
loss = loss_fn(output, target)
loss.backward()
return loss
optimizer.step(closure)
Algorithms¶

class
torch.optim.
Optimizer
(params, defaults)[source]¶ Base class for all optimizers.
Warning
Parameters need to be specified as collections that have a deterministic ordering that is consistent between runs. Examples of objects that don’t satisfy those properties are sets and iterators over values of dictionaries.
 Parameters
params (iterable) – an iterable of
torch.Tensor
s ordict
s. Specifies what Tensors should be optimized.defaults – (dict): a dict containing default values of optimization options (used when a parameter group doesn’t specify them).

add_param_group
(param_group)[source]¶ Add a param group to the
Optimizer
s param_groups.This can be useful when fine tuning a pretrained network as frozen layers can be made trainable and added to the
Optimizer
as training progresses. Parameters
param_group (dict) – Specifies what Tensors should be optimized along with group
optimization options. (specific) –

load_state_dict
(state_dict)[source]¶ Loads the optimizer state.
 Parameters
state_dict (dict) – optimizer state. Should be an object returned from a call to
state_dict()
.

state_dict
()[source]¶ Returns the state of the optimizer as a
dict
.It contains two entries:
 state  a dict holding current optimization state. Its content
differs between optimizer classes.
param_groups  a dict containing all parameter groups

step
(closure)[source]¶ Performs a single optimization step (parameter update).
 Parameters
closure (callable) – A closure that reevaluates the model and returns the loss. Optional for most optimizers.

zero_grad
()[source]¶ Clears the gradients of all optimized
torch.Tensor
s.

class
torch.optim.
Adadelta
(params, lr=1.0, rho=0.9, eps=1e06, weight_decay=0)[source]¶ Implements Adadelta algorithm.
It has been proposed in ADADELTA: An Adaptive Learning Rate Method.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
rho (float, optional) – coefficient used for computing a running average of squared gradients (default: 0.9)
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e6)
lr (float, optional) – coefficient that scale delta before it is applied to the parameters (default: 1.0)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)

class
torch.optim.
Adagrad
(params, lr=0.01, lr_decay=0, weight_decay=0, initial_accumulator_value=0)[source]¶ Implements Adagrad algorithm.
It has been proposed in Adaptive Subgradient Methods for Online Learning and Stochastic Optimization.
 Parameters

class
torch.optim.
Adam
(params, lr=0.001, betas=(0.9, 0.999), eps=1e08, weight_decay=0, amsgrad=False)[source]¶ Implements Adam algorithm.
It has been proposed in Adam: A Method for Stochastic Optimization.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e3)
betas (Tuple[float, float], optional) – coefficients used for computing running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e8)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)
amsgrad (boolean, optional) – whether to use the AMSGrad variant of this algorithm from the paper On the Convergence of Adam and Beyond (default: False)

class
torch.optim.
AdamW
(params, lr=0.001, betas=(0.9, 0.999), eps=1e08, weight_decay=0.01, amsgrad=False)[source]¶ Implements AdamW algorithm.
The original Adam algorithm was proposed in Adam: A Method for Stochastic Optimization. The AdamW variant was proposed in Decoupled Weight Decay Regularization.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e3)
betas (Tuple[float, float], optional) – coefficients used for computing running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e8)
weight_decay (float, optional) – weight decay coefficient (default: 1e2)
amsgrad (boolean, optional) – whether to use the AMSGrad variant of this algorithm from the paper On the Convergence of Adam and Beyond (default: False)

class
torch.optim.
SparseAdam
(params, lr=0.001, betas=(0.9, 0.999), eps=1e08)[source]¶ Implements lazy version of Adam algorithm suitable for sparse tensors.
In this variant, only moments that show up in the gradient get updated, and only those portions of the gradient get applied to the parameters.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e3)
betas (Tuple[float, float], optional) – coefficients used for computing running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e8)

class
torch.optim.
Adamax
(params, lr=0.002, betas=(0.9, 0.999), eps=1e08, weight_decay=0)[source]¶ Implements Adamax algorithm (a variant of Adam based on infinity norm).
It has been proposed in Adam: A Method for Stochastic Optimization.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 2e3)
betas (Tuple[float, float], optional) – coefficients used for computing running averages of gradient and its square
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e8)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)

class
torch.optim.
ASGD
(params, lr=0.01, lambd=0.0001, alpha=0.75, t0=1000000.0, weight_decay=0)[source]¶ Implements Averaged Stochastic Gradient Descent.
It has been proposed in Acceleration of stochastic approximation by averaging.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e2)
lambd (float, optional) – decay term (default: 1e4)
alpha (float, optional) – power for eta update (default: 0.75)
t0 (float, optional) – point at which to start averaging (default: 1e6)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)

class
torch.optim.
LBFGS
(params, lr=1, max_iter=20, max_eval=None, tolerance_grad=1e05, tolerance_change=1e09, history_size=100, line_search_fn=None)[source]¶ Implements LBFGS algorithm, heavily inspired by minFunc <https://www.cs.ubc.ca/~schmidtm/Software/minFunc.html>.
Warning
This optimizer doesn’t support perparameter options and parameter groups (there can be only one).
Warning
Right now all parameters have to be on a single device. This will be improved in the future.
Note
This is a very memory intensive optimizer (it requires additional
param_bytes * (history_size + 1)
bytes). If it doesn’t fit in memory try reducing the history size, or use a different algorithm. Parameters
lr (float) – learning rate (default: 1)
max_iter (int) – maximal number of iterations per optimization step (default: 20)
max_eval (int) – maximal number of function evaluations per optimization step (default: max_iter * 1.25).
tolerance_grad (float) – termination tolerance on first order optimality (default: 1e5).
tolerance_change (float) – termination tolerance on function value/parameter changes (default: 1e9).
history_size (int) – update history size (default: 100).
line_search_fn (str) – either ‘strong_wolfe’ or None (default: None).

class
torch.optim.
RMSprop
(params, lr=0.01, alpha=0.99, eps=1e08, weight_decay=0, momentum=0, centered=False)[source]¶ Implements RMSprop algorithm.
Proposed by G. Hinton in his course.
The centered version first appears in Generating Sequences With Recurrent Neural Networks.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e2)
momentum (float, optional) – momentum factor (default: 0)
alpha (float, optional) – smoothing constant (default: 0.99)
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e8)
centered (bool, optional) – if
True
, compute the centered RMSProp, the gradient is normalized by an estimation of its varianceweight_decay (float, optional) – weight decay (L2 penalty) (default: 0)

class
torch.optim.
Rprop
(params, lr=0.01, etas=(0.5, 1.2), step_sizes=(1e06, 50))[source]¶ Implements the resilient backpropagation algorithm.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e2)
etas (Tuple[float, float], optional) – pair of (etaminus, etaplis), that are multiplicative increase and decrease factors (default: (0.5, 1.2))
step_sizes (Tuple[float, float], optional) – a pair of minimal and maximal allowed step sizes (default: (1e6, 50))

class
torch.optim.
SGD
(params, lr=<required parameter>, momentum=0, dampening=0, weight_decay=0, nesterov=False)[source]¶ Implements stochastic gradient descent (optionally with momentum).
Nesterov momentum is based on the formula from On the importance of initialization and momentum in deep learning.
 Parameters
params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float) – learning rate
momentum (float, optional) – momentum factor (default: 0)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)
dampening (float, optional) – dampening for momentum (default: 0)
nesterov (bool, optional) – enables Nesterov momentum (default: False)
Example
>>> optimizer = torch.optim.SGD(model.parameters(), lr=0.1, momentum=0.9) >>> optimizer.zero_grad() >>> loss_fn(model(input), target).backward() >>> optimizer.step()
Note
The implementation of SGD with Momentum/Nesterov subtly differs from Sutskever et. al. and implementations in some other frameworks.
Considering the specific case of Momentum, the update can be written as
$v = \rho * v + g \\ p = p  lr * v$where p, g, v and $\rho$ denote the parameters, gradient, velocity, and momentum respectively.
This is in contrast to Sutskever et. al. and other frameworks which employ an update of the form
$v = \rho * v + lr * g \\ p = p  v$The Nesterov version is analogously modified.
How to adjust Learning Rate¶
torch.optim.lr_scheduler
provides several methods to adjust the learning
rate based on the number of epochs. torch.optim.lr_scheduler.ReduceLROnPlateau
allows dynamic learning rate reducing based on some validation measurements.
Learning rate scheduling should be applied after optimizer’s update; e.g., you should write your code this way:
>>> scheduler = ...
>>> for epoch in range(100):
>>> train(...)
>>> validate(...)
>>> scheduler.step()
Warning
Prior to PyTorch 1.1.0, the learning rate scheduler was expected to be called before
the optimizer’s update; 1.1.0 changed this behavior in a BCbreaking way. If you use
the learning rate scheduler (calling scheduler.step()
) before the optimizer’s update
(calling optimizer.step()
), this will skip the first value of the learning rate schedule.
If you are unable to reproduce results after upgrading to PyTorch 1.1.0, please check
if you are calling scheduler.step()
at the wrong time.

class
torch.optim.lr_scheduler.
LambdaLR
(optimizer, lr_lambda, last_epoch=1)[source]¶ Sets the learning rate of each parameter group to the initial lr times a given function. When last_epoch=1, sets initial lr as lr.
 Parameters
Example
>>> # Assuming optimizer has two groups. >>> lambda1 = lambda epoch: epoch // 30 >>> lambda2 = lambda epoch: 0.95 ** epoch >>> scheduler = LambdaLR(optimizer, lr_lambda=[lambda1, lambda2]) >>> for epoch in range(100): >>> train(...) >>> validate(...) >>> scheduler.step()

load_state_dict
(state_dict)[source]¶ Loads the schedulers state.
 Parameters
state_dict (dict) – scheduler state. Should be an object returned from a call to
state_dict()
.

class
torch.optim.lr_scheduler.
StepLR
(optimizer, step_size, gamma=0.1, last_epoch=1)[source]¶ Sets the learning rate of each parameter group to the initial lr decayed by gamma every step_size epochs. When last_epoch=1, sets initial lr as lr.
 Parameters
Example
>>> # Assuming optimizer uses lr = 0.05 for all groups >>> # lr = 0.05 if epoch < 30 >>> # lr = 0.005 if 30 <= epoch < 60 >>> # lr = 0.0005 if 60 <= epoch < 90 >>> # ... >>> scheduler = StepLR(optimizer, step_size=30, gamma=0.1) >>> for epoch in range(100): >>> train(...) >>> validate(...) >>> scheduler.step()

class
torch.optim.lr_scheduler.
MultiStepLR
(optimizer, milestones, gamma=0.1, last_epoch=1)[source]¶ Set the learning rate of each parameter group to the initial lr decayed by gamma once the number of epoch reaches one of the milestones. When last_epoch=1, sets initial lr as lr.
 Parameters
Example
>>> # Assuming optimizer uses lr = 0.05 for all groups >>> # lr = 0.05 if epoch < 30 >>> # lr = 0.005 if 30 <= epoch < 80 >>> # lr = 0.0005 if epoch >= 80 >>> scheduler = MultiStepLR(optimizer, milestones=[30,80], gamma=0.1) >>> for epoch in range(100): >>> train(...) >>> validate(...) >>> scheduler.step()

class
torch.optim.lr_scheduler.
ExponentialLR
(optimizer, gamma, last_epoch=1)[source]¶ Set the learning rate of each parameter group to the initial lr decayed by gamma every epoch. When last_epoch=1, sets initial lr as lr.

class
torch.optim.lr_scheduler.
CosineAnnealingLR
(optimizer, T_max, eta_min=0, last_epoch=1)[source]¶ Set the learning rate of each parameter group using a cosine annealing schedule, where $\eta_{max}$ is set to the initial lr and $T_{cur}$ is the number of epochs since the last restart in SGDR:
$\eta_t = \eta_{min} + \frac{1}{2}(\eta_{max}  \eta_{min})(1 + \cos(\frac{T_{cur}}{T_{max}}\pi))$When last_epoch=1, sets initial lr as lr.
It has been proposed in SGDR: Stochastic Gradient Descent with Warm Restarts. Note that this only implements the cosine annealing part of SGDR, and not the restarts.

class
torch.optim.lr_scheduler.
ReduceLROnPlateau
(optimizer, mode='min', factor=0.1, patience=10, verbose=False, threshold=0.0001, threshold_mode='rel', cooldown=0, min_lr=0, eps=1e08)[source]¶ Reduce learning rate when a metric has stopped improving. Models often benefit from reducing the learning rate by a factor of 210 once learning stagnates. This scheduler reads a metrics quantity and if no improvement is seen for a ‘patience’ number of epochs, the learning rate is reduced.
 Parameters
optimizer (Optimizer) – Wrapped optimizer.
mode (str) – One of min, max. In min mode, lr will be reduced when the quantity monitored has stopped decreasing; in max mode it will be reduced when the quantity monitored has stopped increasing. Default: ‘min’.
factor (float) – Factor by which the learning rate will be reduced. new_lr = lr * factor. Default: 0.1.
patience (int) – Number of epochs with no improvement after which learning rate will be reduced. For example, if patience = 2, then we will ignore the first 2 epochs with no improvement, and will only decrease the LR after the 3rd epoch if the loss still hasn’t improved then. Default: 10.
verbose (bool) – If
True
, prints a message to stdout for each update. Default:False
.threshold (float) – Threshold for measuring the new optimum, to only focus on significant changes. Default: 1e4.
threshold_mode (str) – One of rel, abs. In rel mode, dynamic_threshold = best * ( 1 + threshold ) in ‘max’ mode or best * ( 1  threshold ) in min mode. In abs mode, dynamic_threshold = best + threshold in max mode or best  threshold in min mode. Default: ‘rel’.
cooldown (int) – Number of epochs to wait before resuming normal operation after lr has been reduced. Default: 0.
min_lr (float or list) – A scalar or a list of scalars. A lower bound on the learning rate of all param groups or each group respectively. Default: 0.
eps (float) – Minimal decay applied to lr. If the difference between new and old lr is smaller than eps, the update is ignored. Default: 1e8.
Example
>>> optimizer = torch.optim.SGD(model.parameters(), lr=0.1, momentum=0.9) >>> scheduler = ReduceLROnPlateau(optimizer, 'min') >>> for epoch in range(10): >>> train(...) >>> val_loss = validate(...) >>> # Note that step should be called after validate() >>> scheduler.step(val_loss)

class
torch.optim.lr_scheduler.
CyclicLR
(optimizer, base_lr, max_lr, step_size_up=2000, step_size_down=None, mode='triangular', gamma=1.0, scale_fn=None, scale_mode='cycle', cycle_momentum=True, base_momentum=0.8, max_momentum=0.9, last_epoch=1)[source]¶ Sets the learning rate of each parameter group according to cyclical learning rate policy (CLR). The policy cycles the learning rate between two boundaries with a constant frequency, as detailed in the paper Cyclical Learning Rates for Training Neural Networks. The distance between the two boundaries can be scaled on a periteration or percycle basis.
Cyclical learning rate policy changes the learning rate after every batch. step should be called after a batch has been used for training.
This class has three builtin policies, as put forth in the paper: “triangular”:
A basic triangular cycle w/ no amplitude scaling.
 “triangular2”:
A basic triangular cycle that scales initial amplitude by half each cycle.
 “exp_range”:
A cycle that scales initial amplitude by gamma**(cycle iterations) at each cycle iteration.
This implementation was adapted from the github repo: bckenstler/CLR
 Parameters
optimizer (Optimizer) – Wrapped optimizer.
base_lr (float or list) – Initial learning rate which is the lower boundary in the cycle for each parameter group.
max_lr (float or list) – Upper learning rate boundaries in the cycle for each parameter group. Functionally, it defines the cycle amplitude (max_lr  base_lr). The lr at any cycle is the sum of base_lr and some scaling of the amplitude; therefore max_lr may not actually be reached depending on scaling function.
step_size_up (int) – Number of training iterations in the increasing half of a cycle. Default: 2000
step_size_down (int) – Number of training iterations in the decreasing half of a cycle. If step_size_down is None, it is set to step_size_up. Default: None
mode (str) – One of {triangular, triangular2, exp_range}. Values correspond to policies detailed above. If scale_fn is not None, this argument is ignored. Default: ‘triangular’
gamma (float) – Constant in ‘exp_range’ scaling function: gamma**(cycle iterations) Default: 1.0
scale_fn (function) – Custom scaling policy defined by a single argument lambda function, where 0 <= scale_fn(x) <= 1 for all x >= 0. If specified, then ‘mode’ is ignored. Default: None
scale_mode (str) – {‘cycle’, ‘iterations’}. Defines whether scale_fn is evaluated on cycle number or cycle iterations (training iterations since start of cycle). Default: ‘cycle’
cycle_momentum (bool) – If
True
, momentum is cycled inversely to learning rate between ‘base_momentum’ and ‘max_momentum’. Default: Truebase_momentum (float or list) – Lower momentum boundaries in the cycle for each parameter group. Note that momentum is cycled inversely to learning rate; at the peak of a cycle, momentum is ‘base_momentum’ and learning rate is ‘max_lr’. Default: 0.8
max_momentum (float or list) – Upper momentum boundaries in the cycle for each parameter group. Functionally, it defines the cycle amplitude (max_momentum  base_momentum). The momentum at any cycle is the difference of max_momentum and some scaling of the amplitude; therefore base_momentum may not actually be reached depending on scaling function. Note that momentum is cycled inversely to learning rate; at the start of a cycle, momentum is ‘max_momentum’ and learning rate is ‘base_lr’ Default: 0.9
last_epoch (int) – The index of the last batch. This parameter is used when resuming a training job. Since step() should be invoked after each batch instead of after each epoch, this number represents the total number of batches computed, not the total number of epochs computed. When last_epoch=1, the schedule is started from the beginning. Default: 1
Example
>>> optimizer = torch.optim.SGD(model.parameters(), lr=0.1, momentum=0.9) >>> scheduler = torch.optim.lr_scheduler.CyclicLR(optimizer, base_lr=0.01, max_lr=0.1) >>> data_loader = torch.utils.data.DataLoader(...) >>> for epoch in range(10): >>> for batch in data_loader: >>> train_batch(...) >>> scheduler.step()