import copy
from collections import defaultdict
from distutils.version import LooseVersion
from itertools import chain
from torch.nn.utils import clip_grad
from math import inf
from mmcv.utils import TORCH_VERSION
from ..dist_utils import allreduce_grads
from ..fp16_utils import LossScaler, wrap_fp16_model
from .hook import HOOKS, Hook
import torch
if torch.__version__ >= '1.8':
import torch_npu
from apex import amp
try:
from torch.cuda.amp import GradScaler
except ImportError:
pass
def clip_grad_norm_(parameters, max_norm, norm_type=2):
r"""Clips gradient norm of an iterable of parameters.
The norm is computed over all gradients together, as if they were
concatenated into a single vector. Gradients are modified in-place.
Arguments:
parameters (Iterable[Tensor] or Tensor): an iterable of Tensors or a
single Tensor that will have gradients normalized
max_norm (float or int): max norm of the gradients
norm_type (float or int): type of the used p-norm. Can be ``'inf'`` for
infinity norm.
Returns:
Total norm of the parameters (viewed as a single vector).
"""
if isinstance(parameters, torch.Tensor):
parameters = [parameters]
parameters = list(filter(lambda p: p.grad is not None, parameters))
max_norm = float(max_norm)
norm_type = float(norm_type)
if norm_type == inf:
total_norm = max(p.grad.detach().abs().max() for p in parameters)
else:
total_norm = torch.norm(torch.stack([torch.norm(p.grad.detach().float(), norm_type) for p in parameters]), norm_type)
clip_coef = max_norm / (total_norm + 1e-6)
clip_coef = torch_npu.npu_format_cast(clip_coef, 2)
if clip_coef < 1:
for p in parameters:
p.grad.detach().mul_(clip_coef)
return total_norm
@HOOKS.register_module()
class OptimizerHook(Hook):
def __init__(self, grad_clip=None):
self.grad_clip = grad_clip
def clip_grads(self, params):
params = list(
filter(lambda p: p.requires_grad and p.grad is not None, params))
if len(params) > 0:
return clip_grad_norm_(params, **self.grad_clip)
def after_train_iter(self, runner):
runner.optimizer.zero_grad()
with amp.scale_loss(runner.outputs['loss'], runner.optimizer) as scaled_loss:
scaled_loss.backward()
runner.optimizer.step()
if (TORCH_VERSION != 'parrots'
and LooseVersion(TORCH_VERSION) >= LooseVersion('1.6.0')):
@HOOKS.register_module()
class Fp16OptimizerHook(OptimizerHook):
"""FP16 optimizer hook (using PyTorch's implementation).
If you are using PyTorch >= 1.6, torch.cuda.amp is used as the backend,
to take care of the optimization procedure.
Args:
loss_scale (float | str | dict): Scale factor configuration.
If loss_scale is a float, static loss scaling will be used with
the specified scale. If loss_scale is a string, it must be
'dynamic', then dynamic loss scaling will be used.
It can also be a dict containing arguments of GradScalar.
Defaults to 512. For Pytorch >= 1.6, mmcv uses official
implementation of GradScaler. If you use a dict version of
loss_scale to create GradScaler, please refer to:
https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.GradScaler
for the parameters.
Examples:
>>> loss_scale = dict(
... init_scale=65536.0,
... growth_factor=2.0,
... backoff_factor=0.5,
... growth_interval=2000
... )
>>> optimizer_hook = Fp16OptimizerHook(loss_scale=loss_scale)
"""
def __init__(self,
grad_clip=None,
coalesce=True,
bucket_size_mb=-1,
loss_scale=512.,
distributed=True):
self.grad_clip = grad_clip
self.coalesce = coalesce
self.bucket_size_mb = bucket_size_mb
self.distributed = distributed
self._scale_update_param = None
if loss_scale == 'dynamic':
self.loss_scaler = GradScaler()
elif isinstance(loss_scale, float):
self._scale_update_param = loss_scale
self.loss_scaler = GradScaler(init_scale=loss_scale)
elif isinstance(loss_scale, dict):
self.loss_scaler = GradScaler(**loss_scale)
else:
raise ValueError('loss_scale must be of type float, dict, or '
f'"dynamic", got {loss_scale}')
def before_run(self, runner):
"""Preparing steps before Mixed Precision Training."""
wrap_fp16_model(runner.model)
if 'fp16' in runner.meta and 'loss_scaler' in runner.meta['fp16']:
scaler_state_dict = runner.meta['fp16']['loss_scaler']
self.loss_scaler.load_state_dict(scaler_state_dict)
def copy_grads_to_fp32(self, fp16_net, fp32_weights):
"""Copy gradients from fp16 model to fp32 weight copy."""
for fp32_param, fp16_param in zip(fp32_weights,
fp16_net.parameters()):
if fp16_param.grad is not None:
if fp32_param.grad is None:
fp32_param.grad = fp32_param.data.new(
fp32_param.size())
fp32_param.grad.copy_(fp16_param.grad)
def copy_params_to_fp16(self, fp16_net, fp32_weights):
"""Copy updated params from fp32 weight copy to fp16 model."""
for fp16_param, fp32_param in zip(fp16_net.parameters(),
fp32_weights):
fp16_param.data.copy_(fp32_param.data)
def after_train_iter(self, runner):
"""Backward optimization steps for Mixed Precision Training. For
dynamic loss scaling, please refer to
https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.GradScaler.
1. Scale the loss by a scale factor.
2. Backward the loss to obtain the gradients.
3. Unscale the optimizer’s gradient tensors.
4. Call optimizer.step() and update scale factor.
5. Save loss_scaler state_dict for resume purpose.
"""
runner.model.zero_grad()
runner.optimizer.zero_grad()
self.loss_scaler.scale(runner.outputs['loss']).backward()
self.loss_scaler.unscale_(runner.optimizer)
if self.grad_clip is not None:
grad_norm = self.clip_grads(runner.model.parameters())
if grad_norm is not None:
runner.log_buffer.update({'grad_norm': float(grad_norm)},
runner.outputs['num_samples'])
self.loss_scaler.step(runner.optimizer)
self.loss_scaler.update(self._scale_update_param)
runner.meta.setdefault(
'fp16', {})['loss_scaler'] = self.loss_scaler.state_dict()
else:
@HOOKS.register_module()
class Fp16OptimizerHook(OptimizerHook):
"""FP16 optimizer hook (mmcv's implementation).
The steps of fp16 optimizer is as follows.
1. Scale the loss value.
2. BP in the fp16 model.
2. Copy gradients from fp16 model to fp32 weights.
3. Update fp32 weights.
4. Copy updated parameters from fp32 weights to fp16 model.
Refer to https://arxiv.org/abs/1710.03740 for more details.
Args:
loss_scale (float | str | dict): Scale factor configuration.
If loss_scale is a float, static loss scaling will be used with
the specified scale. If loss_scale is a string, it must be
'dynamic', then dynamic loss scaling will be used.
It can also be a dict containing arguments of LossScaler.
Defaults to 512.
"""
def __init__(self,
grad_clip=None,
coalesce=True,
bucket_size_mb=-1,
loss_scale=512.,
distributed=True):
self.grad_clip = grad_clip
self.coalesce = coalesce
self.bucket_size_mb = bucket_size_mb
self.distributed = distributed
if loss_scale == 'dynamic':
self.loss_scaler = LossScaler(mode='dynamic')
elif isinstance(loss_scale, float):
self.loss_scaler = LossScaler(
init_scale=loss_scale, mode='static')
elif isinstance(loss_scale, dict):
self.loss_scaler = LossScaler(**loss_scale)
else:
raise ValueError('loss_scale must be of type float, dict, or '
f'"dynamic", got {loss_scale}')
def before_run(self, runner):
"""Preparing steps before Mixed Precision Training.
1. Make a master copy of fp32 weights for optimization.
2. Convert the main model from fp32 to fp16.
"""
old_groups = runner.optimizer.param_groups
runner.optimizer.param_groups = copy.deepcopy(
runner.optimizer.param_groups)
state = defaultdict(dict)
p_map = {
old_p: p
for old_p, p in zip(
chain(*(g['params'] for g in old_groups)),
chain(*(g['params']
for g in runner.optimizer.param_groups)))
}
for k, v in runner.optimizer.state.items():
state[p_map[k]] = v
runner.optimizer.state = state
wrap_fp16_model(runner.model)
if 'fp16' in runner.meta and 'loss_scaler' in runner.meta['fp16']:
scaler_state_dict = runner.meta['fp16']['loss_scaler']
self.loss_scaler.load_state_dict(scaler_state_dict)
def copy_grads_to_fp32(self, fp16_net, fp32_weights):
"""Copy gradients from fp16 model to fp32 weight copy."""
for fp32_param, fp16_param in zip(fp32_weights,
fp16_net.parameters()):
if fp16_param.grad is not None:
if fp32_param.grad is None:
fp32_param.grad = fp32_param.data.new(
fp32_param.size())
fp32_param.grad.copy_(fp16_param.grad)
def copy_params_to_fp16(self, fp16_net, fp32_weights):
"""Copy updated params from fp32 weight copy to fp16 model."""
for fp16_param, fp32_param in zip(fp16_net.parameters(),
fp32_weights):
fp16_param.data.copy_(fp32_param.data)
def after_train_iter(self, runner):
"""Backward optimization steps for Mixed Precision Training. For
dynamic loss scaling, please refer `loss_scalar.py`
1. Scale the loss by a scale factor.
2. Backward the loss to obtain the gradients (fp16).
3. Copy gradients from the model to the fp32 weight copy.
4. Scale the gradients back and update the fp32 weight copy.
5. Copy back the params from fp32 weight copy to the fp16 model.
6. Save loss_scaler state_dict for resume purpose.
"""
runner.model.zero_grad()
runner.optimizer.zero_grad()
scaled_loss = runner.outputs['loss'] * self.loss_scaler.loss_scale
scaled_loss.backward()
fp32_weights = []
for param_group in runner.optimizer.param_groups:
fp32_weights += param_group['params']
self.copy_grads_to_fp32(runner.model, fp32_weights)
if self.distributed:
allreduce_grads(fp32_weights, self.coalesce,
self.bucket_size_mb)
has_overflow = self.loss_scaler.has_overflow(fp32_weights)
if not has_overflow:
for param in fp32_weights:
if param.grad is not None:
param.grad.div_(self.loss_scaler.loss_scale)
if self.grad_clip is not None:
grad_norm = self.clip_grads(fp32_weights)
if grad_norm is not None:
runner.log_buffer.update(
{'grad_norm': float(grad_norm)},
runner.outputs['num_samples'])
runner.optimizer.step()
self.copy_params_to_fp16(runner.model, fp32_weights)
self.loss_scaler.update_scale(has_overflow)
if has_overflow:
runner.logger.warning('Check overflow, downscale loss scale '
f'to {self.loss_scaler.cur_scale}')
runner.meta.setdefault(
'fp16', {})['loss_scaler'] = self.loss_scaler.state_dict()