deviandice/DL-Art-School
1
0
Fork 0
forked from mrq/DL-Art-School
Code Issues Pull Requests Packages Projects Releases Wiki Activity

gaussian_diffusion: support fp16

Browse Source
This commit is contained in:
James Betker 2021-12-12 19:52:21 -07:00
parent aa7cfd1edf
commit 76f86c0e47
1 changed files with 25 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

70
codes/trainer/injectors/gaussian_diffusion_injector.py
View File

@ -2,6 +2,7 @@ import random
import time
import torch
from torch.cuda.amp import autocast
from models.diffusion.gaussian_diffusion import GaussianDiffusion, get_named_beta_schedule
from models.diffusion.resample import create_named_schedule_sampler, LossAwareSampler
@ -35,20 +36,21 @@ class GaussianDiffusionInjector(Injector):
torch.manual_seed(0)
random.seed(0)
model_inputs = {k: state[v] for k, v in self.model_input_keys.items()}
t, weights = self.schedule_sampler.sample(hq.shape[0], hq.device)
diffusion_outputs = self.diffusion.training_losses(gen, hq, t, model_kwargs=model_inputs)
if isinstance(self.schedule_sampler, LossAwareSampler):
self.schedule_sampler.update_with_local_losses(t, diffusion_outputs['losses'])
with autocast(enabled=self.env['opt']['fp16']):
model_inputs = {k: state[v] for k, v in self.model_input_keys.items()}
t, weights = self.schedule_sampler.sample(hq.shape[0], hq.device)
diffusion_outputs = self.diffusion.training_losses(gen, hq, t, model_kwargs=model_inputs)
if isinstance(self.schedule_sampler, LossAwareSampler):
self.schedule_sampler.update_with_local_losses(t, diffusion_outputs['losses'])
if len(self.extra_model_output_keys) > 0:
assert(len(self.extra_model_output_keys) == len(diffusion_outputs['extra_outputs']))
out = {k: v for k, v in zip(self.extra_model_output_keys, diffusion_outputs['extra_outputs'])}
else:
out = {}
out.update({self.output: diffusion_outputs['mse'],
self.output_variational_bounds_key: diffusion_outputs['vb'],
self.output_x_start_key: diffusion_outputs['x_start_predicted']})
if len(self.extra_model_output_keys) > 0:
assert(len(self.extra_model_output_keys) == len(diffusion_outputs['extra_outputs']))
out = {k: v for k, v in zip(self.extra_model_output_keys, diffusion_outputs['extra_outputs'])}
else:
out = {}
out.update({self.output: diffusion_outputs['mse'],
self.output_variational_bounds_key: diffusion_outputs['vb'],
self.output_x_start_key: diffusion_outputs['x_start_predicted']})
# Absolutely critical to undo the above seed.
if not gen.training:
@ -58,36 +60,13 @@ class GaussianDiffusionInjector(Injector):
return out
class AutoregressiveGaussianDiffusionInjector(Injector):
def __init__(self, opt, env):
super().__init__(opt, env)
self.generator = opt['generator']
self.output_variational_bounds_key = opt['out_key_vb_loss']
self.output_x_start_key = opt['out_key_x_start']
opt['diffusion_args']['betas'] = get_named_beta_schedule(**opt['beta_schedule'])
opt['diffusion_args']['use_timesteps'] = space_timesteps(opt['beta_schedule']['num_diffusion_timesteps'],
[opt['beta_schedule']['num_diffusion_timesteps']])
self.diffusion = SpacedDiffusion(**opt['diffusion_args'])
self.schedule_sampler = create_named_schedule_sampler(opt['sampler_type'], self.diffusion)
self.model_input_keys = opt_get(opt, ['model_input_keys'], [])
self.model_output_keys = opt['model_output_keys']
self.model_eps_pred_key = opt['prediction_key']
def forward(self, state):
gen = self.env['generators'][self.opt['generator']]
hq = state[self.input]
model_inputs = {k: state[v] for k, v in self.model_input_keys.items()}
t, weights = self.schedule_sampler.sample(hq.shape[0], hq.device)
diffusion_outputs = self.diffusion.autoregressive_training_losses(gen, hq, t, self.model_output_keys,
self.model_eps_pred_key,
model_kwargs=model_inputs)
if isinstance(self.schedule_sampler, LossAwareSampler):
self.schedule_sampler.update_with_local_losses(t, diffusion_outputs['losses'])
outputs = {k: diffusion_outputs[k] for k in self.model_output_keys}
outputs.update({self.output: diffusion_outputs['mse'],
self.output_variational_bounds_key: diffusion_outputs['vb'],
self.output_x_start_key: diffusion_outputs['x_start_predicted']})
return outputs
def closest_multiple(inp, multiple):
div = inp / multiple
mod = inp % multiple
if mod == 0:
return inp
else:
return (div+1)*multiple
# Performs inference using a network trained to predict a reverse diffusion process, which nets a image.
@ -110,6 +89,7 @@ class GaussianDiffusionInferenceInjector(Injector):
self.model_input_keys = opt_get(opt, ['model_input_keys'], [])
self.use_ema_model = opt_get(opt, ['use_ema'], False)
self.noise_style = opt_get(opt, ['noise_type'], 'random') # 'zero', 'fixed' or 'random'
self.multiple_requirement = opt_get(opt, ['multiple_requirement'], 4096)
def forward(self, state):
if self.use_ema_model:
@ -124,10 +104,10 @@ class GaussianDiffusionInferenceInjector(Injector):
model_inputs['low_res'].shape[-1] * self.output_scale_factor)
dev = model_inputs['low_res'].device
elif 'spectrogram' in model_inputs.keys():
output_shape = (self.output_batch_size, 1, model_inputs['spectrogram'].shape[-1] * self.output_scale_factor)
output_shape = (self.output_batch_size, 1, closest_multiple(model_inputs['spectrogram'].shape[-1] * self.output_scale_factor, self.multiple_requirement))
dev = model_inputs['spectrogram'].device
elif 'discrete_spectrogram' in model_inputs.keys():
output_shape = (self.output_batch_size, 1, model_inputs['discrete_spectrogram'].shape[-1]*1024)
output_shape = (self.output_batch_size, 1, closest_multiple(model_inputs['discrete_spectrogram'].shape[-1]*1024, self.multiple_requirement))
dev = model_inputs['discrete_spectrogram'].device
else:
raise NotImplementedError