index
:
textual-inversion-diff
master
textual-inversion-diff
summary
refs
log
tree
commit
diff
stats
log msg
author
committer
range
path:
root
/
data
Commit message (
Collapse
)
Author
Age
Files
Lines
*
Dataset: Repeat data to fill batch to batch_size
Volpeon
2023-02-15
1
-0
/
+3
|
*
Update
Volpeon
2023-02-13
1
-4
/
+4
|
*
Update
Volpeon
2023-01-17
1
-1
/
+1
|
*
If valid set size is 0, re-use one image from train set
Volpeon
2023-01-16
1
-1
/
+1
|
*
Moved multi-TI code from Dreambooth to TI script
Volpeon
2023-01-16
1
-1
/
+1
|
*
More training adjustments
Volpeon
2023-01-16
1
-18
/
+21
|
*
Pad dataset if len(items) < batch_size
Volpeon
2023-01-16
1
-0
/
+3
|
*
Handle empty validation dataset
Volpeon
2023-01-16
1
-20
/
+27
|
*
Extended Dreambooth: Train TI tokens separately
Volpeon
2023-01-16
1
-0
/
+7
|
*
Implemented extended Dreambooth training
Volpeon
2023-01-16
1
-0
/
+19
|
*
Restored functional trainer
Volpeon
2023-01-15
1
-8
/
+0
|
*
Update
Volpeon
2023-01-15
1
-7
/
+4
|
*
Fix
Volpeon
2023-01-14
1
-1
/
+1
|
*
Cleanup
Volpeon
2023-01-14
1
-13
/
+8
|
*
Removed PromptProcessor, modularized training loop
Volpeon
2023-01-13
1
-16
/
+20
|
*
Code deduplication
Volpeon
2023-01-13
1
-5
/
+42
|
*
Added arg to disable tag shuffling
Volpeon
2023-01-10
1
-1
/
+3
|
*
Enable buckets for validation, fixed vaildation repeat arg
Volpeon
2023-01-09
1
-2
/
+3
|
*
Add --valid_set_repeat
Volpeon
2023-01-09
1
-1
/
+5
|
*
No cache after all
Volpeon
2023-01-08
1
-17
/
+7
|
*
Cache token IDs in dataset
Volpeon
2023-01-08
1
-8
/
+20
|
*
Fix
Volpeon
2023-01-08
1
-5
/
+5
|
*
Improved aspect ratio bucketing
Volpeon
2023-01-08
1
-3
/
+19
|
*
Fixed aspect ratio bucketing
Volpeon
2023-01-08
1
-3
/
+5
|
*
Cleanup
Volpeon
2023-01-08
1
-15
/
+25
|
*
Fixed aspect ratio bucketing; allow passing token IDs to pipeline
Volpeon
2023-01-08
1
-35
/
+43
|
*
Improved aspect ratio bucketing
Volpeon
2023-01-08
1
-119
/
+154
|
*
Made aspect ratio bucketing configurable
Volpeon
2023-01-07
1
-11
/
+22
|
*
Added progressive aspect ratio bucketing
Volpeon
2023-01-07
1
-56
/
+88
|
*
Fix
Volpeon
2023-01-06
1
-2
/
+4
|
*
Update
Volpeon
2023-01-05
2
-21
/
+22
|
*
Update
Volpeon
2023-01-04
1
-1
/
+1
|
*
Various updates
Volpeon
2023-01-04
1
-18
/
+27
|
*
Fixed reproducibility, more consistant validation
Volpeon
2023-01-04
1
-3
/
+7
|
*
Fixed accuracy calc, other improvements
Volpeon
2023-01-01
1
-1
/
+1
|
*
Misc improvements
Volpeon
2022-12-30
1
-20
/
+20
|
*
Training script improvements
Volpeon
2022-12-30
1
-11
/
+4
|
*
Set default dimensions to 768; add config inheritance
Volpeon
2022-12-26
1
-2
/
+2
|
*
Update
Volpeon
2022-12-24
1
-8
/
+8
|
*
Fix
Volpeon
2022-12-24
1
-2
/
+2
|
*
Better dataset prompt handling
Volpeon
2022-12-23
1
-9
/
+16
|
*
num_class_images is now class images per train image
Volpeon
2022-12-23
1
-2
/
+2
|
*
Improved class prompt handling
Volpeon
2022-12-23
1
-17
/
+12
|
*
Dependency cleanup/upgrades
Volpeon
2022-12-20
1
-3
/
+2
|
*
Improved dataset prompt handling, fixed
Volpeon
2022-12-19
1
-18
/
+23
|
*
Avoid increased VRAM usage on validation
Volpeon
2022-12-15
1
-3
/
+10
|
*
More generic datset filter
Volpeon
2022-12-14
1
-5
/
+5
|
*
Optimized Textual Inversion training by filtering dataset by existence of ↵
Volpeon
2022-12-13
1
-7
/
+31
|
|
|
|
added tokens
*
Training improvements such as tag dropout
Volpeon
2022-12-11
1
-3
/
+10
|
*
Various updated; shuffle prompt content during training
Volpeon
2022-12-10
1
-9
/
+20
|
[next]