prompt_parser: allow spaces in schedules, add test, log/ignore errors

Only build the parser once (at import time) instead of for each step.

doctest is run by simply executing modules/prompt_parser.py
This commit is contained in:
Rae Fu 2022-10-04 09:49:51 -06:00 committed by AUTOMATIC1111
parent 1eb588cbf1
commit 90e911fd54
2 changed files with 95 additions and 54 deletions

View File

@ -84,7 +84,7 @@ class StableDiffusionProcessing:
self.s_tmin = opts.s_tmin self.s_tmin = opts.s_tmin
self.s_tmax = float('inf') # not representable as a standard ui option self.s_tmax = float('inf') # not representable as a standard ui option
self.s_noise = opts.s_noise self.s_noise = opts.s_noise
if not seed_enable_extras: if not seed_enable_extras:
self.subseed = -1 self.subseed = -1
self.subseed_strength = 0 self.subseed_strength = 0
@ -296,7 +296,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
assert(len(p.prompt) > 0) assert(len(p.prompt) > 0)
else: else:
assert p.prompt is not None assert p.prompt is not None
devices.torch_gc() devices.torch_gc()
seed = get_fixed_seed(p.seed) seed = get_fixed_seed(p.seed)
@ -359,8 +359,8 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
#uc = p.sd_model.get_learned_conditioning(len(prompts) * [p.negative_prompt]) #uc = p.sd_model.get_learned_conditioning(len(prompts) * [p.negative_prompt])
#c = p.sd_model.get_learned_conditioning(prompts) #c = p.sd_model.get_learned_conditioning(prompts)
with devices.autocast(): with devices.autocast():
uc = prompt_parser.get_learned_conditioning(len(prompts) * [p.negative_prompt], p.steps) uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps)
c = prompt_parser.get_learned_conditioning(prompts, p.steps) c = prompt_parser.get_learned_conditioning(shared.sd_model, prompts, p.steps)
if len(model_hijack.comments) > 0: if len(model_hijack.comments) > 0:
for comment in model_hijack.comments: for comment in model_hijack.comments:
@ -527,7 +527,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
# GC now before running the next img2img to prevent running out of memory # GC now before running the next img2img to prevent running out of memory
x = None x = None
devices.torch_gc() devices.torch_gc()
samples = self.sampler.sample_img2img(self, samples, noise, conditioning, unconditional_conditioning, steps=self.steps) samples = self.sampler.sample_img2img(self, samples, noise, conditioning, unconditional_conditioning, steps=self.steps)
return samples return samples

View File

@ -1,10 +1,7 @@
import re import re
from collections import namedtuple from collections import namedtuple
import torch
from lark import Lark, Transformer, Visitor
import functools
import modules.shared as shared import lark
# a prompt like this: "fantasy landscape with a [mountain:lake:0.25] and [an oak:a christmas tree:0.75][ in foreground::0.6][ in background:0.25] [shoddy:masterful:0.5]" # a prompt like this: "fantasy landscape with a [mountain:lake:0.25] and [an oak:a christmas tree:0.75][ in foreground::0.6][ in background:0.25] [shoddy:masterful:0.5]"
# will be represented with prompt_schedule like this (assuming steps=100): # will be represented with prompt_schedule like this (assuming steps=100):
@ -14,25 +11,48 @@ import modules.shared as shared
# [75, 'fantasy landscape with a lake and an oak in background masterful'] # [75, 'fantasy landscape with a lake and an oak in background masterful']
# [100, 'fantasy landscape with a lake and a christmas tree in background masterful'] # [100, 'fantasy landscape with a lake and a christmas tree in background masterful']
schedule_parser = lark.Lark(r"""
!start: (prompt | /[][():]/+)*
prompt: (emphasized | scheduled | plain | WHITESPACE)*
!emphasized: "(" prompt ")"
| "(" prompt ":" prompt ")"
| "[" prompt "]"
scheduled: "[" [prompt ":"] prompt ":" [WHITESPACE] NUMBER "]"
WHITESPACE: /\s+/
plain: /([^\\\[\]():]|\\.)+/
%import common.SIGNED_NUMBER -> NUMBER
""")
def get_learned_conditioning_prompt_schedules(prompts, steps): def get_learned_conditioning_prompt_schedules(prompts, steps):
grammar = r"""
start: prompt
prompt: (emphasized | scheduled | weighted | plain)*
!emphasized: "(" prompt ")"
| "(" prompt ":" prompt ")"
| "[" prompt "]"
scheduled: "[" (prompt ":")? prompt ":" NUMBER "]"
!weighted: "{" weighted_item ("|" weighted_item)* "}"
!weighted_item: prompt (":" prompt)?
plain: /([^\\\[\](){}:|]|\\.)+/
%import common.SIGNED_NUMBER -> NUMBER
""" """
parser = Lark(grammar, parser='lalr') >>> g = lambda p: get_learned_conditioning_prompt_schedules([p], 10)[0]
>>> g("test")
[[10, 'test']]
>>> g("a [b:3]")
[[3, 'a '], [10, 'a b']]
>>> g("a [b: 3]")
[[3, 'a '], [10, 'a b']]
>>> g("a [[[b]]:2]")
[[2, 'a '], [10, 'a [[b]]']]
>>> g("[(a:2):3]")
[[3, ''], [10, '(a:2)']]
>>> g("a [b : c : 1] d")
[[1, 'a b d'], [10, 'a c d']]
>>> g("a[b:[c:d:2]:1]e")
[[1, 'abe'], [2, 'ace'], [10, 'ade']]
>>> g("a [unbalanced")
[[10, 'a [unbalanced']]
>>> g("a [b:.5] c")
[[5, 'a c'], [10, 'a b c']]
>>> g("a [{b|d{:.5] c") # not handling this right now
[[5, 'a c'], [10, 'a {b|d{ c']]
>>> g("((a][:b:c [d:3]")
[[3, '((a][:b:c '], [10, '((a][:b:c d']]
"""
def collect_steps(steps, tree): def collect_steps(steps, tree):
l = [steps] l = [steps]
class CollectSteps(Visitor): class CollectSteps(lark.Visitor):
def scheduled(self, tree): def scheduled(self, tree):
tree.children[-1] = float(tree.children[-1]) tree.children[-1] = float(tree.children[-1])
if tree.children[-1] < 1: if tree.children[-1] < 1:
@ -43,13 +63,10 @@ def get_learned_conditioning_prompt_schedules(prompts, steps):
return sorted(set(l)) return sorted(set(l))
def at_step(step, tree): def at_step(step, tree):
class AtStep(Transformer): class AtStep(lark.Transformer):
def scheduled(self, args): def scheduled(self, args):
if len(args) == 2: before, after, _, when = args
before, after, when = (), *args yield before or () if step <= when else after
else:
before, after, when = args
yield before if step <= when else after
def start(self, args): def start(self, args):
def flatten(x): def flatten(x):
if type(x) == str: if type(x) == str:
@ -57,16 +74,22 @@ def get_learned_conditioning_prompt_schedules(prompts, steps):
else: else:
for gen in x: for gen in x:
yield from flatten(gen) yield from flatten(gen)
return ''.join(flatten(args[0])) return ''.join(flatten(args))
def plain(self, args): def plain(self, args):
yield args[0].value yield args[0].value
def __default__(self, data, children, meta): def __default__(self, data, children, meta):
for child in children: for child in children:
yield from child yield from child
return AtStep().transform(tree) return AtStep().transform(tree)
def get_schedule(prompt): def get_schedule(prompt):
tree = parser.parse(prompt) try:
tree = schedule_parser.parse(prompt)
except lark.exceptions.LarkError as e:
if 0:
import traceback
traceback.print_exc()
return [[steps, prompt]]
return [[t, at_step(t, tree)] for t in collect_steps(steps, tree)] return [[t, at_step(t, tree)] for t in collect_steps(steps, tree)]
promptdict = {prompt: get_schedule(prompt) for prompt in set(prompts)} promptdict = {prompt: get_schedule(prompt) for prompt in set(prompts)}
@ -77,8 +100,7 @@ ScheduledPromptConditioning = namedtuple("ScheduledPromptConditioning", ["end_at
ScheduledPromptBatch = namedtuple("ScheduledPromptBatch", ["shape", "schedules"]) ScheduledPromptBatch = namedtuple("ScheduledPromptBatch", ["shape", "schedules"])
def get_learned_conditioning(prompts, steps): def get_learned_conditioning(model, prompts, steps):
res = [] res = []
prompt_schedules = get_learned_conditioning_prompt_schedules(prompts, steps) prompt_schedules = get_learned_conditioning_prompt_schedules(prompts, steps)
@ -92,7 +114,7 @@ def get_learned_conditioning(prompts, steps):
continue continue
texts = [x[1] for x in prompt_schedule] texts = [x[1] for x in prompt_schedule]
conds = shared.sd_model.get_learned_conditioning(texts) conds = model.get_learned_conditioning(texts)
cond_schedule = [] cond_schedule = []
for i, (end_at_step, text) in enumerate(prompt_schedule): for i, (end_at_step, text) in enumerate(prompt_schedule):
@ -105,12 +127,13 @@ def get_learned_conditioning(prompts, steps):
def reconstruct_cond_batch(c: ScheduledPromptBatch, current_step): def reconstruct_cond_batch(c: ScheduledPromptBatch, current_step):
res = torch.zeros(c.shape, device=shared.device, dtype=next(shared.sd_model.parameters()).dtype) param = c.schedules[0][0].cond
res = torch.zeros(c.shape, device=param.device, dtype=param.dtype)
for i, cond_schedule in enumerate(c.schedules): for i, cond_schedule in enumerate(c.schedules):
target_index = 0 target_index = 0
for curret_index, (end_at, cond) in enumerate(cond_schedule): for current, (end_at, cond) in enumerate(cond_schedule):
if current_step <= end_at: if current_step <= end_at:
target_index = curret_index target_index = current
break break
res[i] = cond_schedule[target_index].cond res[i] = cond_schedule[target_index].cond
@ -148,23 +171,26 @@ def parse_prompt_attention(text):
\\ - literal character '\' \\ - literal character '\'
anything else - just text anything else - just text
Example: >>> parse_prompt_attention('normal text')
[['normal text', 1.0]]
'a (((house:1.3)) [on] a (hill:0.5), sun, (((sky))).' >>> parse_prompt_attention('an (important) word')
[['an ', 1.0], ['important', 1.1], [' word', 1.0]]
produces: >>> parse_prompt_attention('(unbalanced')
[['unbalanced', 1.1]]
[ >>> parse_prompt_attention('\(literal\]')
['a ', 1.0], [['(literal]', 1.0]]
['house', 1.5730000000000004], >>> parse_prompt_attention('(unnecessary)(parens)')
[' ', 1.1], [['unnecessaryparens', 1.1]]
['on', 1.0], >>> parse_prompt_attention('a (((house:1.3)) [on] a (hill:0.5), sun, (((sky))).')
[' a ', 1.1], [['a ', 1.0],
['hill', 0.55], ['house', 1.5730000000000004],
[', sun, ', 1.1], [' ', 1.1],
['sky', 1.4641000000000006], ['on', 1.0],
['.', 1.1] [' a ', 1.1],
] ['hill', 0.55],
[', sun, ', 1.1],
['sky', 1.4641000000000006],
['.', 1.1]]
""" """
res = [] res = []
@ -206,4 +232,19 @@ def parse_prompt_attention(text):
if len(res) == 0: if len(res) == 0:
res = [["", 1.0]] res = [["", 1.0]]
# merge runs of identical weights
i = 0
while i + 1 < len(res):
if res[i][1] == res[i + 1][1]:
res[i][0] += res[i + 1][0]
res.pop(i + 1)
else:
i += 1
return res return res
if __name__ == "__main__":
import doctest
doctest.testmod(optionflags=doctest.NORMALIZE_WHITESPACE)
else:
import torch # doctest faster