-
Notifications
You must be signed in to change notification settings - Fork 58
/
ScheduleTypes.py
229 lines (149 loc) · 9.99 KB
/
ScheduleTypes.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
import comfy
import numexpr
import torch
import numpy as np
import pandas as pd
import re
import json
from .ScheduleFuncs import *
from .BatchFuncs import *
def prompt_schedule(settings:ScheduleSettings,clip):
settings.start_frame = 0
# modulus rollover when current frame exceeds max frames
settings.current_frame = settings.current_frame % settings.max_frames
# clear whitespace and newlines from json
animation_prompts = process_input_text(settings.text_g)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
pos, neg = batch_split_weighted_subprompts(animation_prompts, settings.pre_text_G, settings.app_text_G)
# Interpolate the positive prompt weights over frames
pos_cur_prompt, pos_nxt_prompt, weight = interpolate_prompt_seriesA(pos, settings)
neg_cur_prompt, neg_nxt_prompt, weight = interpolate_prompt_seriesA(neg, settings)
# Apply composable diffusion across the batch
p = PoolAnimConditioning(pos_cur_prompt[settings.current_frame], pos_nxt_prompt[settings.current_frame],
weight[settings.current_frame], clip)
n = PoolAnimConditioning(neg_cur_prompt[settings.current_frame], neg_nxt_prompt[settings.current_frame],
weight[settings.current_frame], clip)
# return the positive and negative conditioning at the current frame
return (p, n,)
def batch_prompt_schedule(settings:ScheduleSettings,clip):
# Clear whitespace and newlines from json
animation_prompts = process_input_text(settings.text_g)
# Add pre_text and app_text then split the combined prompt into positive and negative prompts
pos, neg = batch_split_weighted_subprompts(animation_prompts, settings.pre_text_G, settings.app_text_G)
# Interpolate the positive prompt weights over frames
pos_cur_prompt, pos_nxt_prompt, weight = interpolate_prompt_seriesA(pos, settings)
neg_cur_prompt, neg_nxt_prompt, weight = interpolate_prompt_seriesA(neg, settings)
# Apply composable diffusion across the batch
p = BatchPoolAnimConditioning(pos_cur_prompt, pos_nxt_prompt, weight, clip, settings)
n = BatchPoolAnimConditioning(neg_cur_prompt, neg_nxt_prompt, weight, clip, settings)
# return positive and negative conditioning as well as the current and next prompts for each
return (p, n,)
def batch_prompt_schedule_latentInput(settings:ScheduleSettings,clip, latents):
# Clear whitespace and newlines from json
animation_prompts = process_input_text(settings.text_g)
# Add pre_text and app_text then split the combined prompt into positive and negative prompts
pos, neg = batch_split_weighted_subprompts(animation_prompts, settings.pre_text_G, settings.app_text_G)
# Interpolate the positive prompt weights over frames
pos_cur_prompt, pos_nxt_prompt, weight = interpolate_prompt_seriesA(pos, settings)
# Apply composable diffusion across the batch
p = BatchPoolAnimConditioning(pos_cur_prompt, pos_nxt_prompt, weight, clip, settings)
# Interpolate the negative prompt weights over frames
neg_cur_prompt, neg_nxt_prompt, weight = interpolate_prompt_seriesA(neg, settings)
# Apply composable diffusion across the batch
n = BatchPoolAnimConditioning(neg_cur_prompt, neg_nxt_prompt, weight, clip, settings)
return (p, n, latents,)
def string_schedule(settings:ScheduleSettings):
settings.start_frame = 0
# modulus rollover when current frame exceeds max frames
settings.current_frame = settings.current_frame % settings.max_frames
# Clear whitespace and newlines from json
animation_prompts = process_input_text(settings.text_g)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
pos, neg = batch_split_weighted_subprompts(animation_prompts, settings.pre_text_G, settings.app_text_G)
# Interpolate the positive prompt weights over frames
pos_cur_prompt, pos_nxt_prompt, weight = interpolate_prompt_seriesA(pos, settings)
# Interpolate the negative prompt weights over frames
neg_cur_prompt, neg_nxt_prompt, weight = interpolate_prompt_seriesA(neg, settings)
return(pos_cur_prompt[settings.current_frame], neg_cur_prompt[settings.current_frame], )
def batch_string_schedule(settings:ScheduleSettings):
settings.start_frame = 0
# Clear whitespace and newlines from json
animation_prompts = process_input_text(settings.text_g)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
pos, neg = batch_split_weighted_subprompts(animation_prompts, settings.pre_text_G, settings.app_text_G)
# Interpolate the positive prompt weights over frames
pos_cur_prompt, pos_nxt_prompt, weight = interpolate_prompt_seriesA(pos, settings)
# Interpolate the negative prompt weights over frames
neg_cur_prompt, neg_nxt_prompt, weight = interpolate_prompt_seriesA(neg, settings)
return (pos_cur_prompt, neg_cur_prompt,)
def prompt_schedule_SDXL(settings:ScheduleSettings,clip):
# modulus rollover when current frame exceeds max frames
settings.current_frame = settings.current_frame % settings.max_frames
# Clear whitespace and newlines from json
animation_prompts_G = process_input_text(settings.text_g)
animation_prompts_L = process_input_text(settings.text_l)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
posG, negG = batch_split_weighted_subprompts(animation_prompts_G, settings.pre_text_G, settings.app_text_G)
posL, negL = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
pc, pn, pw = BatchInterpolatePromptsSDXL(posG, posL, clip, settings, )
nc, nn, nw = BatchInterpolatePromptsSDXL(negG, negL, clip, settings, )
#apply composable diffusion to the current frame
p = addWeighted(pc[settings.current_frame], pn[settings.current_frame], pw[settings.current_frame])
n = addWeighted(nc[settings.current_frame], nn[settings.current_frame], nw[settings.current_frame])
return (p, n,)
def batch_prompt_schedule_SDXL(settings:ScheduleSettings,clip):
# Clear whitespace and newlines from json
animation_prompts_G = process_input_text(settings.text_g)
animation_prompts_L = process_input_text(settings.text_l)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
posG, negG = batch_split_weighted_subprompts(animation_prompts_G, settings.pre_text_G, settings.app_text_G)
posL, negL = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
pc, pn, pw = BatchInterpolatePromptsSDXL(posG, posL, clip, settings,)
nc, nn, nw = BatchInterpolatePromptsSDXL(negG, negL, clip, settings,)
p = BatchPoolAnimConditioningSDXL(pc, pn, pw, clip, settings)
n = BatchPoolAnimConditioningSDXL(nc, nn, nw, clip, settings)
return (p, n,)
def batch_prompt_schedule_SDXL_latentInput(settings:ScheduleSettings,clip, latents):
settings.start_frame = 0
# Clear whitespace and newlines from json
animation_prompts_G = process_input_text(settings.text_g)
animation_prompts_L = process_input_text(settings.text_l)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
posG, negG = batch_split_weighted_subprompts(animation_prompts_G, settings.pre_text_G, settings.app_text_G)
posL, negL = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
pc, pn, pw = BatchInterpolatePromptsSDXL(posG, posL, clip, settings)
nc, nn, nw = BatchInterpolatePromptsSDXL(negG, negL, clip, settings)
p = BatchPoolAnimConditioningSDXL(pc, pn, pw, clip, settings)
n = BatchPoolAnimConditioningSDXL(nc, nn, nw, clip, settings)
return (p, n, latents,)
def prompt_schedule_SD3(settings:ScheduleSettings,clip):
# modulus rollover when current frame exceeds max frames
settings.current_frame = settings.current_frame % settings.max_frames
# Clear whitespace and newlines from json
animation_prompts_G = process_input_text(settings.text_g)
animation_prompts_L = process_input_text(settings.text_l)
animation_prompts_T = process_input_text(settings.text_l)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
posG, negG = batch_split_weighted_subprompts(animation_prompts_G, settings.pre_text_G, settings.app_text_G)
posL, negL = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
posT, negT = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
pc, pn, pw = BatchInterpolatePromptsSD3(posG, posL, clip, settings, )
nc, nn, nw = BatchInterpolatePromptsSD3(negG, negL, clip, settings, )
#apply composable diffusion to the current frame
p = addWeighted(pc[settings.current_frame], pn[settings.current_frame], pw[settings.current_frame])
n = addWeighted(nc[settings.current_frame], nn[settings.current_frame], nw[settings.current_frame])
return (p, n,)
def batch_prompt_schedule_SD3(settings:ScheduleSettings,clip):
# Clear whitespace and newlines from json
animation_prompts_G = process_input_text(settings.text_g)
animation_prompts_L = process_input_text(settings.text_l)
animation_prompts_T = process_input_text(settings.text_l)
# add pre_text and app_text then split the combined prompt into positive and negative prompts
posG, negG = batch_split_weighted_subprompts(animation_prompts_G, settings.pre_text_G, settings.app_text_G)
posL, negL = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
posT, negT = batch_split_weighted_subprompts(animation_prompts_L, settings.pre_text_L, settings.app_text_L)
#pc, pn, pw = BatchInterpolatePromptsSD3(posG, posL, clip, settings,)
#nc, nn, nw = BatchInterpolatePromptsSD3(negG, negL, clip, settings,)
#p = BatchPoolAnimConditioningSD3(pc, pn, pw)
#n = BatchPoolAnimConditioningSD3(nc, nn, nw)
return (p, n,)