put back make_save_path and fix error in plugin runner

This commit is contained in:
Damian Stewart 2023-09-10 21:37:47 +02:00
parent fa5b38e26b
commit 3fddef3698
3 changed files with 315 additions and 121 deletions

148
plugins/interruptible.py Normal file
View File

@ -0,0 +1,148 @@
import math
import os
import shutil
from plugins.plugins import BasePlugin
from train import save_model
EVERY_N_EPOCHS = 0.3 # how often to save. integers >= 1 save at the end of every nth epoch. floats < 1 subdivide the epoch evenly (eg 0.33 = 3 subdivisions)
class InterruptiblePlugin(BasePlugin):
def __init__(self):
print("Interruptible plugin instantiated")
self.previous_save_path = None
self.every_n_epochs = EVERY_N_EPOCHS
def on_epoch_start(self, **kwargs):
epoch = kwargs['epoch']
epoch_length = kwargs['epoch_length']
self.steps_to_save_this_epoch = self._get_save_step_indices(epoch, epoch_length)
def on_step_end(self, **kwargs):
local_step = kwargs['local_step']
if local_step in self.steps_to_save_this_epoch:
global_step = kwargs['global_step']
epoch = kwargs['epoch']
project_name = kwargs['project_name']
log_folder = kwargs['log_folder']
ckpt_name = f"rolling-{project_name}-ep{epoch:02}-gs{global_step:05}"
save_path = os.path.join(log_folder, "ckpts", ckpt_name)
print(f"{type(self)} saving model to {save_path}")
save_model(save_path, global_step=global_step, ed_state=kwargs['ed_state'], save_ckpt_dir=None, yaml_name=None, save_ckpt=False, save_full_precision=True, save_optimizer_flag=True)
self._remove_previous()
self.previous_save_path = save_path
def on_training_end(self, **kwargs):
self._remove_previous()
def _remove_previous(self):
if self.previous_save_path is not None:
shutil.rmtree(self.previous_save_path, ignore_errors=True)
self.previous_save_path = None
def _get_save_step_indices(self, epoch, epoch_length_steps: int) -> list[int]:
if self.every_n_epochs >= 1:
if ((epoch+1) % self.every_n_epochs) == 0:
# last step only
return [epoch_length_steps-1]
else:
return []
else:
# subdivide the epoch evenly, by rounding self.every_n_epochs to the nearest clean division of steps
num_divisions = max(1, min(epoch_length_steps, round(1/self.every_n_epochs)))
# validation happens after training:
# if an epoch has eg 100 steps and num_divisions is 2, then validation should occur after steps 49 and 99
validate_every_n_steps = epoch_length_steps / num_divisions
return [math.ceil((i+1)*validate_every_n_steps) - 1 for i in range(num_divisions)]
"""
class InterruptiblePlugin_(BasePlugin):
def __init__(self, log_folder, args):
self.log_folder = log_folder
self.project_name = args.project_name
self.max_epochs = args.max_epochs
self.every_n_epochs = 1
@classmethod
def make_resume_path(cls, resume_ckpt_folder):
return os.path.join(resume_ckpt_folder, 'resumable_data.pt')
def load_resume_state(self, resume_ckpt_path: str, ed_state: EveryDreamTrainingState):
resume_path = self.make_resume_path(resume_ckpt_path)
try:
with open(resume_path, 'rb') as f:
resumable_data = torch.load(f)
ed_state.optimizer.load_state_dict(resumable_data['ed_optimizer'])
ed_state.train_batch.load_state_dict(resumable_data['ed_batch'])
except Exception as e:
print(f"InterruptiblePlugin unable to load resume state from {resume_path}: {e}")
return
def on_epoch_start(self, ed_state: EveryDreamTrainingState, **kwargs):
epoch = kwargs['epoch']
epoch_length = kwargs['epoch_length']
if epoch == 0:
resume_ckpt_path = kwargs['resume_ckpt_path']
self.load_resume_state(resume_ckpt_path, ed_state)
self.steps_to_save_this_epoch = self._get_save_step_indices(epoch, epoch_length)
def _get_save_step_indices(self, epoch, epoch_length_steps: int) -> list[int]:
if self.every_n_epochs >= 1:
if ((epoch+1) % self.every_n_epochs) == 0:
# last step only
return [epoch_length_steps-1]
else:
return []
else:
# subdivide the epoch evenly, by rounding self.every_n_epochs to the nearest clean division of steps
num_divisions = max(1, min(epoch_length_steps, round(1/self.every_n_epochs)))
# validation happens after training:
# if an epoch has eg 100 steps and num_divisions is 2, then validation should occur after steps 49 and 99
validate_every_n_steps = epoch_length_steps / num_divisions
return [math.ceil((i+1)*validate_every_n_steps) - 1 for i in range(num_divisions)]
def on_step_end(self, epoch: int, local_step: int, global_step: int, ed_state: EveryDreamTrainingState):
if local_step in self.steps_to_save_this_epoch:
self.save_and_remove_prior(epoch, global_step, ed_state)
def _save_and_remove_prior(self, epoch: int, global_step: int, ed_state: EveryDreamTrainingState):
rolling_save_path = self.make_save_path(epoch, global_step, prepend="rolling-")
ed_optimizer: EveryDreamOptimizer = ed_state.optimizer
save_model(rolling_save_path,
ed_state=ed_state, save_ckpt_dir=None, yaml_name=None, save_ckpt=False, save_optimizer_flag=True)
kwargs['unet'], kwargs['text_encoder'], kwargs['tokenizer'],
kwargs['noise_scheduler'], kwargs['vae'], ed_optimizer,
save_ckpt_dir=None, yaml_name=None, save_optimizer_flag=True, save_ckpt=False)
train_batch: EveryDreamBatch = kwargs['train_batch']
resumable_data = {
'grad_scaler': ed_optimizer.scaler.state_dict(),
'epoch': epoch,
'global_step': global_step,
'train_batch': train_batch.state_dict()
}
if ed_optimizer.lr_scheduler_te is not None:
resumable_data['lr_scheduler_te'] = ed_optimizer.lr_scheduler_te.state_dict()
if ed_optimizer.lr_scheduler_unet is not None:
resumable_data['lr_scheduler_unet'] = ed_optimizer.lr_scheduler_unet.state_dict()
torch.save(resumable_data, os.path.join(rolling_save_path, 'resumable_data.pt'))
self.prev_epoch = epoch
self.prev_global_step = global_step
if epoch > 0:
prev_rolling_save_path = self.make_save_path(epoch, self.prev_global_step, prepend="rolling-")
shutil.rmtree(prev_rolling_save_path, ignore_errors=True)
pass
def make_save_path(self, epoch, global_step, prepend: str="") -> str:
basename = f"{prepend}{self.project_name}-ep{epoch:02}"
if global_step is not None:
basename += f"-gs{global_step:05}"
return os.path.join(self.log_folder, "ckpts", basename)
"""

View File

@ -44,7 +44,7 @@ class Timer:
def __exit__(self, type, value, traceback):
elapsed_time = time.time() - self.start
if elapsed_time > self.warn_seconds:
logging.warning(f'Execution of {self.label} took {elapsed_time} seconds which is longer than the limit of {self.limit} seconds')
logging.warning(f'Execution of {self.label} took {elapsed_time} seconds which is longer than the limit of {self.warn_seconds} seconds')
class PluginRunner:

276
train.py
View File

@ -27,6 +27,7 @@ import gc
import random
import traceback
import shutil
from typing import Optional
import torch.nn.functional as F
from torch.cuda.amp import autocast
@ -102,6 +103,109 @@ def convert_to_hf(ckpt_path):
is_sd1attn, yaml = get_attn_yaml(ckpt_path)
return ckpt_path, is_sd1attn, yaml
class EveryDreamTrainingState:
def __init__(self,
optimizer: EveryDreamOptimizer,
train_batch: EveryDreamBatch,
unet: UNet2DConditionModel,
text_encoder: CLIPTextModel,
tokenizer: CLIPTokenizer,
scheduler,
vae: AutoencoderKL,
unet_ema: Optional[UNet2DConditionModel],
text_encoder_ema: Optional[CLIPTextModel]
):
self.optimizer = optimizer
self.train_batch = train_batch
self.unet = unet
self.text_encoder = text_encoder
self.tokenizer = tokenizer
self.scheduler = scheduler
self.vae = vae
self.unet_ema = unet_ema,
self.text_encoder = text_encoder_ema
@torch.no_grad()
def save_model(save_path, ed_state: EveryDreamTrainingState, global_step: int, save_ckpt_dir, yaml_name,
save_full_precision=False, save_optimizer_flag=False, save_ckpt=True):
"""
Save the model to disk
"""
def save_ckpt_file(diffusers_model_path, sd_ckpt_path):
nonlocal save_ckpt_dir
nonlocal save_full_precision
nonlocal yaml_name
if save_ckpt_dir is not None:
sd_ckpt_full = os.path.join(save_ckpt_dir, sd_ckpt_path)
else:
sd_ckpt_full = os.path.join(os.curdir, sd_ckpt_path)
save_ckpt_dir = os.curdir
half = not save_full_precision
logging.info(f" * Saving SD model to {sd_ckpt_full}")
converter(model_path=diffusers_model_path, checkpoint_path=sd_ckpt_full, half=half)
if yaml_name and yaml_name != "v1-inference.yaml":
yaml_save_path = f"{os.path.join(save_ckpt_dir, os.path.basename(diffusers_model_path))}.yaml"
logging.info(f" * Saving yaml to {yaml_save_path}")
shutil.copyfile(yaml_name, yaml_save_path)
if global_step is None or global_step == 0:
logging.warning(" No model to save, something likely blew up on startup, not saving")
return
if args.ema_decay_rate != None:
pipeline_ema = StableDiffusionPipeline(
vae=ed_state.vae,
text_encoder=ed_state.text_encoder_ema,
tokenizer=ed_state.tokenizer,
unet=ed_state.unet_ema,
scheduler=ed_state.scheduler,
safety_checker=None, # save vram
requires_safety_checker=None, # avoid nag
feature_extractor=None, # must be none of no safety checker
)
diffusers_model_path = save_path + "_ema"
logging.info(f" * Saving diffusers EMA model to {diffusers_model_path}")
pipeline_ema.save_pretrained(diffusers_model_path)
if save_ckpt:
sd_ckpt_path_ema = f"{os.path.basename(save_path)}_ema.ckpt"
save_ckpt_file(diffusers_model_path, sd_ckpt_path_ema)
pipeline = StableDiffusionPipeline(
vae=ed_state.vae,
text_encoder=ed_state.text_encoder,
tokenizer=ed_state.tokenizer,
unet=ed_state.unet,
scheduler=ed_state.scheduler,
safety_checker=None, # save vram
requires_safety_checker=None, # avoid nag
feature_extractor=None, # must be none of no safety checker
)
diffusers_model_path = save_path
logging.info(f" * Saving diffusers model to {diffusers_model_path}")
pipeline.save_pretrained(diffusers_model_path)
if save_ckpt:
sd_ckpt_path = f"{os.path.basename(save_path)}.ckpt"
save_ckpt_file(diffusers_model_path, sd_ckpt_path)
if save_optimizer_flag:
logging.info(f" Saving optimizer state to {save_path}")
ed_state.optimizer.save(save_path)
def setup_local_logger(args):
"""
configures logger with file and console logging, logs args, and returns the datestamp
@ -478,95 +582,6 @@ def main(args):
if 'cuda' in original_device.type:
torch.cuda.empty_cache()
@torch.no_grad()
def __save_model(save_path, tokenizer, scheduler, vae, ed_optimizer, save_ckpt_dir, yaml_name,
save_full_precision=False, save_optimizer_flag=False, save_ckpt=True):
nonlocal unet
nonlocal text_encoder
nonlocal unet_ema
nonlocal text_encoder_ema
"""
Save the model to disk
"""
def save_ckpt_file(diffusers_model_path, sd_ckpt_path):
nonlocal save_ckpt_dir
nonlocal save_full_precision
nonlocal yaml_name
if save_ckpt_dir is not None:
sd_ckpt_full = os.path.join(save_ckpt_dir, sd_ckpt_path)
else:
sd_ckpt_full = os.path.join(os.curdir, sd_ckpt_path)
save_ckpt_dir = os.curdir
half = not save_full_precision
logging.info(f" * Saving SD model to {sd_ckpt_full}")
converter(model_path=diffusers_model_path, checkpoint_path=sd_ckpt_full, half=half)
if yaml_name and yaml_name != "v1-inference.yaml":
yaml_save_path = f"{os.path.join(save_ckpt_dir, os.path.basename(diffusers_model_path))}.yaml"
logging.info(f" * Saving yaml to {yaml_save_path}")
shutil.copyfile(yaml_name, yaml_save_path)
global global_step
if global_step is None or global_step == 0:
logging.warning(" No model to save, something likely blew up on startup, not saving")
return
if args.ema_decay_rate != None:
pipeline_ema = StableDiffusionPipeline(
vae=vae,
text_encoder=text_encoder_ema,
tokenizer=tokenizer,
unet=unet_ema,
scheduler=scheduler,
safety_checker=None, # save vram
requires_safety_checker=None, # avoid nag
feature_extractor=None, # must be none of no safety checker
)
diffusers_model_path = save_path + "_ema"
logging.info(f" * Saving diffusers EMA model to {diffusers_model_path}")
pipeline_ema.save_pretrained(diffusers_model_path)
if save_ckpt:
sd_ckpt_path_ema = f"{os.path.basename(save_path)}_ema.ckpt"
save_ckpt_file(diffusers_model_path, sd_ckpt_path_ema)
pipeline = StableDiffusionPipeline(
vae=vae,
text_encoder=text_encoder,
tokenizer=tokenizer,
unet=unet,
scheduler=scheduler,
safety_checker=None, # save vram
requires_safety_checker=None, # avoid nag
feature_extractor=None, # must be none of no safety checker
)
diffusers_model_path = save_path
logging.info(f" * Saving diffusers model to {diffusers_model_path}")
pipeline.save_pretrained(diffusers_model_path)
if save_ckpt:
sd_ckpt_path = f"{os.path.basename(save_path)}.ckpt"
save_ckpt_file(diffusers_model_path, sd_ckpt_path)
if save_optimizer_flag:
logging.info(f" Saving optimizer state to {save_path}")
ed_optimizer.save(save_path)
use_ema_dacay_training = (args.ema_decay_rate != None) or (args.ema_strength_target != None)
ema_model_loaded_from_file = False
@ -575,6 +590,7 @@ def main(args):
ema_device = torch.device(args.ema_device)
optimizer_state_path = None
try:
# check for a local file
hf_cache_path = get_hf_ckpt_cache_path(args.resume_ckpt)
@ -583,10 +599,6 @@ def main(args):
text_encoder = CLIPTextModel.from_pretrained(model_root_folder, subfolder="text_encoder")
vae = AutoencoderKL.from_pretrained(model_root_folder, subfolder="vae")
unet = UNet2DConditionModel.from_pretrained(model_root_folder, subfolder="unet")
optimizer_state_path = os.path.join(args.resume_ckpt, "optimizer.pt")
if not os.path.exists(optimizer_state_path):
optimizer_state_path = None
else:
# try to download from HF using resume_ckpt as a repo id
downloaded = try_download_model_from_hf(repo_id=args.resume_ckpt)
@ -701,7 +713,9 @@ def main(args):
# Make sure correct types are used for models
unet_ema = unet_ema.to(ema_device, dtype=unet.dtype)
text_encoder_ema = text_encoder_ema.to(ema_device, dtype=text_encoder.dtype)
else:
unet_ema = None
text_encoder_ema = None
try:
#unet = torch.compile(unet)
@ -835,9 +849,9 @@ def main(args):
logging.error(f"{Fore.LIGHTRED_EX} CTRL-C received, attempting to save model to {interrupted_checkpoint_path}{Style.RESET_ALL}")
logging.error(f"{Fore.LIGHTRED_EX} ************************************************************************{Style.RESET_ALL}")
time.sleep(2) # give opportunity to ctrl-C again to cancel save
__save_model(interrupted_checkpoint_path, tokenizer, noise_scheduler, vae,
ed_optimizer, args.save_ckpt_dir, args.save_full_precision, args.save_optimizer,
save_ckpt=not args.no_save_ckpt)
save_model(interrupted_checkpoint_path, global_step=global_step, ed_state=make_current_ed_state(),
save_ckpt_dir=args.save_ckpt_dir, yaml_name=yaml, save_full_precision=args.save_full_precision,
save_optimizer_flag=args.save_optimizer, save_ckpt=not args.no_save_ckpt)
exit(_SIGTERM_EXIT_CODE)
else:
# non-main threads (i.e. dataloader workers) should exit cleanly
@ -1032,7 +1046,12 @@ def main(args):
torch.cuda.empty_cache()
def make_save_path(epoch, global_step, prepend=""):
return os.path.join(f"{log_folder}/ckpts/{prepend}{args.project_name}-ep{epoch:02}-gs{global_step:05}")
basename = f"{prepend}{args.project_name}"
if epoch is not None:
basename += f"-ep{epoch:02}"
if global_step is not None:
basename += f"-gs{global_step:05}"
return os.path.join(log_folder, "ckpts", basename)
@ -1057,26 +1076,42 @@ def main(args):
from plugins.plugins import PluginRunner
plugin_runner = PluginRunner(plugins=plugins)
def make_current_ed_state() -> EveryDreamTrainingState:
return EveryDreamTrainingState(optimizer=ed_optimizer,
train_batch=train_batch,
unet=unet,
text_encoder=text_encoder,
tokenizer=tokenizer,
scheduler=noise_scheduler,
vae=vae,
unet_ema=unet_ema,
text_encoder_ema=text_encoder_ema)
epoch = None
try:
write_batch_schedule(args, log_folder, train_batch, epoch = 0)
plugin_runner.run_on_training_start(log_folder=log_folder, project_name=args.project_name)
for epoch in range(args.max_epochs):
if args.load_settings_every_epoch:
load_train_json_from_file(args)
epoch_len = math.ceil(len(train_batch) / args.batch_size)
plugin_runner.run_on_epoch_start(epoch=epoch,
plugin_runner.run_on_epoch_start(
epoch=epoch,
global_step=global_step,
epoch_length=epoch_len,
project_name=args.project_name,
log_folder=log_folder,
data_root=args.data_root)
data_root=args.data_root
)
loss_epoch = []
epoch_start_time = time.time()
images_per_sec_log_step = []
epoch_len = math.ceil(len(train_batch) / args.batch_size)
steps_pbar = tqdm(range(epoch_len), position=1, leave=False, dynamic_ncols=True)
steps_pbar.set_description(f"{Fore.LIGHTCYAN_EX}Steps{Style.RESET_ALL}")
@ -1086,12 +1121,15 @@ def main(args):
)
for step, batch in enumerate(train_dataloader):
step_start_time = time.time()
plugin_runner.run_on_step_start(epoch=epoch,
local_step=step,
global_step=global_step,
project_name=args.project_name,
log_folder=log_folder,
batch=batch)
batch=batch,
ed_state=make_current_ed_state())
model_pred, target, loss = get_model_prediction_and_target(batch["image"], batch["tokens"], args.zero_frequency_noise_ratio, return_loss=True)
@ -1158,27 +1196,29 @@ def main(args):
min_since_last_ckpt = (time.time() - last_epoch_saved_time) / 60
needs_save = False
if args.ckpt_every_n_minutes is not None and (min_since_last_ckpt > args.ckpt_every_n_minutes):
last_epoch_saved_time = time.time()
logging.info(f"Saving model, {args.ckpt_every_n_minutes} mins at step {global_step}")
save_path = make_save_path(epoch, global_step)
__save_model(save_path, tokenizer, noise_scheduler, vae, ed_optimizer,
args.save_ckpt_dir, yaml, args.save_full_precision, args.save_optimizer,
save_ckpt=not args.no_save_ckpt)
needs_save = True
if epoch > 0 and epoch % args.save_every_n_epochs == 0 and step == 0 and epoch < args.max_epochs - 1 and epoch >= args.save_ckpts_from_n_epochs:
logging.info(f" Saving model, {args.save_every_n_epochs} epochs at step {global_step}")
needs_save = True
if needs_save:
save_path = make_save_path(epoch, global_step)
__save_model(save_path, tokenizer, noise_scheduler, vae, ed_optimizer,
args.save_ckpt_dir, yaml, args.save_full_precision, args.save_optimizer,
save_ckpt=not args.no_save_ckpt)
save_model(save_path, global_step=global_step, ed_state=make_current_ed_state(),
save_ckpt_dir=None, yaml_name=None,
save_full_precision=args.save_full_precision,
save_optimizer_flag=args.save_optimizer, save_ckpt=False)
plugin_runner.run_on_step_end(epoch=epoch,
global_step=global_step,
local_step=step,
project_name=args.project_name,
log_folder=log_folder,
data_root=args.data_root,
batch=batch)
batch=batch,
ed_state=make_current_ed_state())
del batch
global_step += 1
@ -1195,6 +1235,7 @@ def main(args):
train_batch.shuffle(epoch_n=epoch, max_epochs = args.max_epochs)
write_batch_schedule(args, log_folder, train_batch, epoch + 1)
if len(loss_epoch) > 0:
loss_epoch = sum(loss_epoch) / len(loss_epoch)
log_writer.add_scalar(tag="loss/epoch", scalar_value=loss_epoch, global_step=global_step)
@ -1209,9 +1250,13 @@ def main(args):
# end of training
epoch = args.max_epochs
plugin_runner.run_on_training_end()
save_path = make_save_path(epoch, global_step, prepend=("" if args.no_prepend_last else "last-"))
__save_model(save_path, tokenizer, noise_scheduler, vae, ed_optimizer, args.save_ckpt_dir,
yaml, args.save_full_precision, args.save_optimizer, save_ckpt=not args.no_save_ckpt)
save_model(save_path, global_step=global_step, ed_state=make_current_ed_state(),
save_ckpt_dir=args.save_ckpt_dir, yaml_name=yaml, save_full_precision=args.save_full_precision,
save_optimizer_flag=args.save_optimizer, save_ckpt=not args.no_save_ckpt)
total_elapsed_time = time.time() - training_start_time
logging.info(f"{Fore.CYAN}Training complete{Style.RESET_ALL}")
@ -1221,8 +1266,9 @@ def main(args):
except Exception as ex:
logging.error(f"{Fore.LIGHTYELLOW_EX}Something went wrong, attempting to save model{Style.RESET_ALL}")
save_path = make_save_path(epoch, global_step, prepend="errored-")
__save_model(save_path, tokenizer, noise_scheduler, vae, ed_optimizer, args.save_ckpt_dir,
yaml, args.save_full_precision, args.save_optimizer, save_ckpt=not args.no_save_ckpt)
save_model(save_path, global_step=global_step, ed_state=make_current_ed_state(),
save_ckpt_dir=args.save_ckpt_dir, yaml_name=yaml, save_full_precision=args.save_full_precision,
save_optimizer_flag=args.save_optimizer, save_ckpt=not args.no_save_ckpt)
logging.info(f"{Fore.LIGHTYELLOW_EX}Model saved, re-raising exception and exiting. Exception was:{Style.RESET_ALL}{Fore.LIGHTRED_EX} {ex} {Style.RESET_ALL}")
raise ex