57eb54b838
implement CUDA device selection by ID
3 years ago
bb0f1a2cda
inspiration finished
3 years ago
26d1073745
Merge remote-tracking branch 'historytab/master'
3 years ago
f49c08ea56
prevent error spam when processing images without txt files for captions
3 years ago
5e9afa5c8a
Merge branch 'master' into fix/train-preprocess-keep-ratio
3 years ago
ac0aa2b18e
loading SD VAE, see PR #3303
3 years ago
704036ff07
make aspect ratio overlay work regardless of selected localization
3 years ago
02e4d4694d
Change option description of unload_models_when_training
3 years ago
272fa527bb
Remove unused variable.
3 years ago
fccad18a59
Refer to Hypernet's name, sensibly, by its name variable.
3 years ago
19818f023c
Match hypernet name with filename in all cases.
3 years ago
51e3dc9cca
Sanitize hypernet name input.
3 years ago
306e2ff6ab
Update image_embedding.py
3 years ago
d0ea471b0c
Use opts in textual_inversion image_embedding.py for dynamic fonts
3 years ago
9286fe53de
make aestetic embedding ciompatible with prompts longer than 75 tokens
3 years ago
df57064093
do not load aesthetic clip model until it's needed
...
add refresh button for aesthetic embeddings
add aesthetic params to images' infotext
3 years ago
7d6b388d71
Merge branch 'ae'
3 years ago
bf30673f51
Fix Hypernet infotext string split bug for PR #3283
3 years ago
03a1e288c4
turns out LayerNorm also has weight and bias and needs to be pre-multiplied and trained for hypernets
3 years ago
0c5522ea21
Merge branch 'master' into training-help-text
3 years ago
2273e752fb
Remove redundant try/except.
3 years ago
4ff274e1e3
Revise comments.
3 years ago
6014fb8afb
Do nothing if image file already exists.
3 years ago
5245c7a493
Issue #2921-Give PNG info to Hypernet previews.
3 years ago
b69c37d25e
Allow datasets with only 1 image in TI
3 years ago
7157e5d064
interrogate: Fix CLIP-interrogation on CPU
...
Currently, trying to perform CLIP interrogation on a CPU fails, saying:
```
RuntimeError: "slow_conv2d_cpu" not implemented for 'Half'
```
This merge request fixes this issue by detecting whether the target
device is CPU and, if so, force-enabling `--no-half` and passing
`device="cpu"` to `clip.load()` (which then does some extra tricks to
ensure it works correctly on CPU).
3 years ago
d6bd6a425d
Merge branch 'master' into master
3 years ago
c23f666dba
a more strict check for activation type and a more reasonable check for type of layer in hypernets
3 years ago
a26fc2834c
Merge pull request #3199 from discus0434/master
...
Add features to insert activation functions to hypernetworks
3 years ago
9d71eef02e
sort file list in alphabetical ordering in extras
3 years ago
a13c3bed3c
Fixed path issue while extras batch processing
3 years ago
fb5a8cf0d9
Added try except to extras batch from directory
3 years ago
60872c5b40
Fixed path issue while extras batch processing
3 years ago
74088c2a06
allow float sizes for hypernet's layer_structure
3 years ago
4587218190
updated readme and some small stylistic changes to code
3 years ago
49533eed9e
XY grid correctly re-assignes model when config changes
3 years ago
d23a46ceaa
Different approach to skip/interrupt with highres fix
3 years ago
708c3a7bd8
Added PLMS hijack and made sure to always replace methods
3 years ago
d1cb08bfb2
fix skip and interrupt for highres. fix option
3 years ago
92a17a7a4a
Made dummy latents smaller. Minor code cleanups
3 years ago
f89829ec3a
Revert "fix bugs and optimizations"
...
This reverts commit 108be15500
3 years ago
dc66540629
Merge branch 'AUTOMATIC1111:master' into Inspiron
3 years ago
108be15500
fix bugs and optimizations
3 years ago
d07cb46f34
inspiration pull request
3 years ago
a71e021236
only linear
3 years ago
d8acd34f66
generalized some functions and option for ignoring first layer
3 years ago
85dd62c4c7
train: ui: added `Split image threshold` and `Split image overlap ratio` to preprocess
3 years ago
9681419e42
train: fixed preprocess image ratio
3 years ago
f8733ad08b
add linear as a act func (option for doin nothing)
3 years ago
6b38c2c19c
Merge branch 'AUTOMATIC1111:master' into master
3 years ago
7f8ab1ee8f
Merge remote-tracking branch 'origin/master'
3 years ago
930b4c64f7
allow float sizes for hypernet's layer_structure
3 years ago
aa7ff2a197
Fixed non-square highres fix generation
3 years ago
858462f719
do caption copy for both flips
3 years ago
ba469343e6
align ui.py imports with upstream
3 years ago
ecb433b220
update
3 years ago
6f98e89486
update
3 years ago
55d8c6cce6
default to ignore existing captions
3 years ago
9b65c4ecf4
pass preprocess_txt_action param
3 years ago
ab353b141d
link existing txt option
3 years ago
fbcce66601
add existing caption file handling
3 years ago
4d6b9f76a5
reorder create_hypernetwork params
3 years ago
c3835ec85c
pass overwrite old flag
3 years ago
632e8d6602
split learn rates
3 years ago
0087079c2d
allow overwrite old embedding
3 years ago
166be3919b
allow overwrite old hn
3 years ago
d6ea584137
change html output
3 years ago
c418467c03
Don't compute latent mask if were not using it. Also added support for fixed highres_fix generation.
3 years ago
dde9f96072
added support for ddim img2img
3 years ago
0719c10bf1
Fixed copying mistake
3 years ago
8e7097d06a
Added support for RunwayML inpainting model
3 years ago
4d663055de
update ui with extra training options
3 years ago
eb7ba4b713
update training header text
3 years ago
b748b583c0
generation_parameters_copypaste.py: fix indent
3 years ago
46122c4ff6
Send empty prompts as valid generation parameter
3 years ago
5e012e4dfa
Infotext saves more specific hypernet name.
3 years ago
1e4809b251
Added a bit of padding to the left
3 years ago
57eb1a64c8
Update ui.py
3 years ago
2ce52d32e4
fix for #3086 failing to load any previous hypernet
3 years ago
c6e9fed500
fix for #3086 failing to load any previous hypernet
3 years ago
019a3a88f0
Update ui.py
3 years ago
365d4b1650
Merge branch 'AUTOMATIC1111:master' into master
3 years ago
f510a2277e
Merge pull request #3086 from discus0434/master
...
Add settings for multi-layer structure hypernetworks
3 years ago
3770b8d2fa
enable to write layer structure of hn himself
3 years ago
42fbda83bb
layer options moves into create hnet ui
3 years ago
f894dd552f
fix for broken checkpoint merger
3 years ago
bb0e7232b3
Fix typo in prompt_parser.py
...
assoicated -> associated
3 years ago
2362d5f00e
Merge branch 'master' into test_resolve_conflicts
3 years ago
05315d8a23
Merge branch 'master' into hot-reload-javascript
3 years ago
1d4aa376e6
Predictable long operation check for time estimation
3 years ago
442dbedc15
Estimated time displayed if jobs take more 60 sec
3 years ago
bcfbb33e50
Added time left estimation
3 years ago
2fd7935ef4
Remove wrong self reference in CUDA support for invokeai
3 years ago
7f8670c4ef
Merge branch 'master' into master
3 years ago
da72becb13
Use training width/height when training hypernetworks.
3 years ago
5d16f59794
Merge branch 'master' into master
3 years ago
5daf9cbb98
Merge remote-tracking branch 'origin/api'
3 years ago
10aca1ca3e
more careful loading of model weights (eliminates some issues with checkpoints that have weird cond_stage_model layer names)
3 years ago
0f0d6ab8e0
call sampler by name
3 years ago
538bc89c26
Image browser improved
3 years ago
Extraltodeus
yfszzx
AUTOMATIC
Rcmcpe
timntorres
DepFA
guaneec
Patryk Wychowaniec
winterspringsummer
random_thoughtss
Vladimir Repin
aria1th
Milly
discus0434
discus0434
Mackerel
Alexandre Simard
Ikko Ashimine
MalumaDev
Anastasius
Cheka
Silent
arcticfaded