Skip to content

GitLab

"...git@developer.sourcefind.cn:chenpangpang/transformers.git" did not exist on "a75c64d80c76c3dc71f735d9197a4a601847e0cd"
Commit 8ca767f1 authored by thomwolf's avatar thomwolf
Browse files

clean up optimization

parent 74a24f0f
Hide whitespace changes
Inline Side-by-side
Showing with 49 additions and 59 deletions
pytorch_transformers/optimization.py
View file @ 8ca767f1
......@@ -24,109 +24,99 @@ from torch.optim.lr_scheduler import LambdaLR
logger = logging.getLogger(__name__)
class ConstantLRSchedule(LambdaLR):
""" Constant learning rate schedule.
"""
def __init__(self, optimizer, last_epoch=-1):
super(ConstantLRSchedule, self).__init__(optimizer, lambda _: 1.0, last_epoch=last_epoch)
class WarmupCosineSchedule(LambdaLR):
"""
Linearly increases learning rate from 0 to 1 over `warmup` training steps.
Decreases learning rate from 1. to 0. over remaining `t_total - warmup` steps following a cosine curve.
If `cycles` (default=0.5) is different from default, learning rate follows cosine function after warmup.
:param warmup: see LRSchedule
:param t_total: see LRSchedule
:param cycles: number of cycles. Default: 0.5, corresponding to cosine decay from 1. at progress==warmup and 0 at progress==1.
:param kw:
class WarmupConstantSchedule(LambdaLR):
""" Linear warmup and then constant.
Linearly increases learning rate schedule from 0 to 1 over `warmup_steps` training steps.
Keeps learning rate schedule equal to 1. after warmup_steps.
"""
warn_t_total = True
def __init__(self, optimizer, warmup_steps, t_total, cycles=.5, last_epoch=-1):
def __init__(self, optimizer, warmup_steps, last_epoch=-1):
def lr_lambda(step):
if step < warmup_steps:
return float(step) / float(max(1.0, warmup_steps))
else:
progress = float(step - warmup_steps) / float(max(1, t_total - warmup_steps)) # progress after warmup
return 0.5 * (1. + math.cos(math.pi * float(cycles) * 2.0 * progress))
return 1.
super(WarmupCosineSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
super(WarmupConstantSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
class WarmupCosineWithHardRestartsSchedule(LambdaLR):
"""
Linearly increases learning rate from 0 to 1 over `warmup` fraction of training steps.
If `cycles` (default=1.) is different from default, learning rate follows `cycles` times a cosine decaying
learning rate (with hard restarts).
class WarmupLinearSchedule(LambdaLR):
""" Linear warmup and then linear decay.
Linearly increases learning rate from 0 to 1 over `warmup_steps` training steps.
Linearly decreases learning rate from 1. to 0. over remaining `t_total - warmup_steps` steps.
"""
def __init__(self, optimizer, warmup_steps, t_total, cycles=1., last_epoch=-1):
def __init__(self, optimizer, warmup_steps, t_total, last_epoch=-1):
def lr_lambda(step):
if step < warmup_steps:
return float(step) / float(max(1, warmup_steps))
else:
progress = float(step - warmup_steps) / float(max(1, t_total - warmup_steps)) # progress after warmup
if progress >= 1.0:
return 0.0
return 0.5 * (1. + math.cos(math.pi * ((float(cycles) * progress) % 1.0)))
return max(0.0, float(t_total - step) / float(max(1.0, t_total - warmup_steps)))
super(WarmupCosineWithHardRestartsSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
super(WarmupLinearSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
class WarmupConstantSchedule(LambdaLR):
"""
Linearly increases learning rate from 0 to 1 over `warmup` fraction of training steps.
Keeps learning rate equal to 1. after warmup.
class WarmupCosineSchedule(LambdaLR):
""" Linear warmup and then cosine decay.
Linearly increases learning rate from 0 to 1 over `warmup_steps` training steps.
Decreases learning rate from 1. to 0. over remaining `t_total - warmup_steps` steps following a cosine curve.
If `cycles` (default=0.5) is different from default, learning rate follows cosine function after warmup.
"""
def __init__(self, optimizer, warmup_steps, last_epoch=-1):
warn_t_total = True
def __init__(self, optimizer, warmup_steps, t_total, cycles=.5, last_epoch=-1):
def lr_lambda(step):
if step < warmup_steps:
return float(step) / float(max(1.0, warmup_steps))
return 1.
super(WarmupConstantSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
else:
progress = float(step - warmup_steps) / float(max(1, t_total - warmup_steps)) # progress after warmup
return max(0.0, 0.5 * (1. + math.cos(math.pi * float(cycles) * 2.0 * progress)))
super(WarmupCosineSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
class WarmupLinearSchedule(LambdaLR):
"""
Linearly increases learning rate from 0 to 1 over `warmup` fraction of training steps.
Linearly decreases learning rate from 1. to 0. over remaining `1 - warmup` steps.
class WarmupCosineWithHardRestartsSchedule(LambdaLR):
""" Linear warmup and then cosine cycles with hard restarts.
Linearly increases learning rate from 0 to 1 over `warmup_steps` training steps.
If `cycles` (default=1.) is different from default, learning rate follows `cycles` times a cosine decaying
learning rate (with hard restarts).
"""
def __init__(self, optimizer, warmup_steps, t_total, last_epoch=-1):
def __init__(self, optimizer, warmup_steps, t_total, cycles=1., last_epoch=-1):
def lr_lambda(step):
if step < warmup_steps:
return float(step) / float(max(1, warmup_steps))
return float(t_total - step) / float(max(1.0, t_total - warmup_steps))
else:
progress = float(step - warmup_steps) / float(max(1, t_total - warmup_steps)) # progress after warmup
if progress >= 1.0:
return 0.0
return max(0.0, 0.5 * (1. + math.cos(math.pi * ((float(cycles) * progress) % 1.0))))
super(WarmupLinearSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
super(WarmupCosineWithHardRestartsSchedule, self).__init__(optimizer, lr_lambda, last_epoch=last_epoch)
class AdamW(Optimizer):
""" Implements Adam algorithm with weight decay fix.
Parameters:
lr: learning rate
warmup: portion of t_total for the warmup, -1 means no warmup. Default: -1
t_total: total number of training steps for the learning
rate schedule, -1 means constant learning rate of 1. (no warmup regardless of warmup setting). Default: -1
schedule: schedule to use for the warmup (see above).
Can be `'warmup_linear'`, `'warmup_constant'`, `'warmup_cosine'`, `'none'`, `None` or a `_LRSchedule` object (see below).
If `None` or `'none'`, learning rate is always kept constant.
Default : `'warmup_linear'`
b1: Adams b1. Default: 0.9
b2: Adams b2. Default: 0.999
e: Adams epsilon. Default: 1e-6
weight_decay: Weight decay. Default: 0.01
max_grad_norm: Maximum norm for the gradients (-1 means no clipping). Default: 1.0
correct_bias: can be set to False to avoid correcting bias in Adam (e.g. like in Bert repository)
lr (float): learning rate. Default 1e-3.
betas (tuple of 2 floats): Adams beta parameters (b1, b2). Default: (0.9, 0.999)
eps (float): Adams epsilon. Default: 1e-6
weight_decay (float): Weight decay. Default: 0.0
correct_bias (bool): can be set to False to avoid correcting bias in Adam (e.g. like in Bert TF repository). Default True.
"""
def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0.01, correct_bias=True):
def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0.0, correct_bias=True):
if lr < 0.0:
raise ValueError("Invalid learning rate: {} - should be >= 0.0".format(lr))
if not 0.0 <= betas[0] < 1.0:
raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[0]))
if not 0.0 <= betas[1] < 1.0:
raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[1] ))
raise ValueError("Invalid beta parameter: {} - should be in [0.0, 1.0[".format(betas[1]))
if not 0.0 <= eps:
raise ValueError("Invalid epsilon value: {} - should be >= 0.0".format(e))
raise ValueError("Invalid epsilon value: {} - should be >= 0.0".format(eps))
defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay,
correct_bias=correct_bias)
super(AdamW, self).__init__(params, defaults)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment