p
/
gallery
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

deeptagger: fix Caformer 

By using the smaller resolution, it starts noticing 2girls,
otherwise the output appears similar.
This commit is contained in:
Přemysl Eric Janouch 2024-01-21 10:38:46 +01:00
parent d37e9e821a
commit 1e3800cc16
Signed by: p
GPG Key ID: A0420B94F92B9493
2 changed files with 18 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
deeptagger/README.adoc
View File

@ -62,16 +62,17 @@ GPU inference
[cols="<,>,>", options=header] [cols="<,>,>", options=header]
|=== |===
|Model|Batch size|Time |Model|Batch size|Time
|ML-Danbooru Caformer dec-5-97527|16|OOM
|WD v1.4 ViT v2 (batch)|16|19 s |WD v1.4 ViT v2 (batch)|16|19 s
|DeepDanbooru|16|21 s |DeepDanbooru|16|21 s
|WD v1.4 SwinV2 v2 (batch)|16|21 s |WD v1.4 SwinV2 v2 (batch)|16|21 s
|ML-Danbooru Caformer dec-5-97527|16|25 s
|WD v1.4 ViT v2 (batch)|4|27 s |WD v1.4 ViT v2 (batch)|4|27 s
|WD v1.4 SwinV2 v2 (batch)|4|30 s |WD v1.4 SwinV2 v2 (batch)|4|30 s
|DeepDanbooru|4|31 s |DeepDanbooru|4|31 s
|ML-Danbooru TResNet-D 6-30000|16|31 s |ML-Danbooru TResNet-D 6-30000|16|31 s
|WD v1.4 MOAT v2 (batch)|16|31 s |WD v1.4 MOAT v2 (batch)|16|31 s
|WD v1.4 ConvNeXT v2 (batch)|16|32 s |WD v1.4 ConvNeXT v2 (batch)|16|32 s
|ML-Danbooru Caformer dec-5-97527|4|32 s
|WD v1.4 ConvNeXTV2 v2 (batch)|16|36 s |WD v1.4 ConvNeXTV2 v2 (batch)|16|36 s
|ML-Danbooru TResNet-D 6-30000|4|39 s |ML-Danbooru TResNet-D 6-30000|4|39 s
|WD v1.4 ConvNeXT v2 (batch)|4|39 s |WD v1.4 ConvNeXT v2 (batch)|4|39 s
@ -79,7 +80,7 @@ GPU inference
|WD v1.4 ConvNeXTV2 v2 (batch)|4|43 s |WD v1.4 ConvNeXTV2 v2 (batch)|4|43 s
|WD v1.4 ViT v2|1|43 s |WD v1.4 ViT v2|1|43 s
|WD v1.4 ViT v2 (batch)|1|43 s |WD v1.4 ViT v2 (batch)|1|43 s
|ML-Danbooru Caformer dec-5-97527|4|48 s |ML-Danbooru Caformer dec-5-97527|1|52 s
|DeepDanbooru|1|53 s |DeepDanbooru|1|53 s
|WD v1.4 MOAT v2|1|53 s |WD v1.4 MOAT v2|1|53 s
|WD v1.4 ConvNeXT v2|1|54 s |WD v1.4 ConvNeXT v2|1|54 s
@ -90,7 +91,6 @@ GPU inference
|WD v1.4 ConvNeXTV2 v2|1|56 s |WD v1.4 ConvNeXTV2 v2|1|56 s
|ML-Danbooru TResNet-D 6-30000|1|58 s |ML-Danbooru TResNet-D 6-30000|1|58 s
|WD v1.4 ConvNeXTV2 v2 (batch)|1|58 s |WD v1.4 ConvNeXTV2 v2 (batch)|1|58 s
|ML-Danbooru Caformer dec-5-97527|1|73 s
|=== |===
CPU inference CPU inference
@ -110,6 +110,7 @@ CPU inference
|WD v1.4 ConvNeXTV2 v2|1|245 s |WD v1.4 ConvNeXTV2 v2|1|245 s
|WD v1.4 ConvNeXTV2 v2 (batch)|4|268 s |WD v1.4 ConvNeXTV2 v2 (batch)|4|268 s
|WD v1.4 ViT v2 (batch)|16|270 s |WD v1.4 ViT v2 (batch)|16|270 s
|ML-Danbooru Caformer dec-5-97527|4|270 s
|WD v1.4 ConvNeXT v2 (batch)|1|272 s |WD v1.4 ConvNeXT v2 (batch)|1|272 s
|WD v1.4 SwinV2 v2 (batch)|4|277 s |WD v1.4 SwinV2 v2 (batch)|4|277 s
|WD v1.4 ViT v2 (batch)|4|277 s |WD v1.4 ViT v2 (batch)|4|277 s
@ -117,6 +118,7 @@ CPU inference
|WD v1.4 SwinV2 v2 (batch)|1|300 s |WD v1.4 SwinV2 v2 (batch)|1|300 s
|WD v1.4 SwinV2 v2|1|302 s |WD v1.4 SwinV2 v2|1|302 s
|WD v1.4 SwinV2 v2 (batch)|16|305 s |WD v1.4 SwinV2 v2 (batch)|16|305 s
|ML-Danbooru Caformer dec-5-97527|16|305 s
|WD v1.4 MOAT v2 (batch)|4|307 s |WD v1.4 MOAT v2 (batch)|4|307 s
|WD v1.4 ViT v2|1|308 s |WD v1.4 ViT v2|1|308 s
|WD v1.4 ViT v2 (batch)|1|311 s |WD v1.4 ViT v2 (batch)|1|311 s
@ -124,9 +126,7 @@ CPU inference
|WD v1.4 MOAT v2|1|332 s |WD v1.4 MOAT v2|1|332 s
|WD v1.4 MOAT v2 (batch)|16|335 s |WD v1.4 MOAT v2 (batch)|16|335 s
|WD v1.4 MOAT v2 (batch)|1|339 s |WD v1.4 MOAT v2 (batch)|1|339 s
|ML-Danbooru Caformer dec-5-97527|4|637 s |ML-Danbooru Caformer dec-5-97527|1|352 s
|ML-Danbooru Caformer dec-5-97527|16|689 s
|ML-Danbooru Caformer dec-5-97527|1|829 s
|=== |===
Model benchmarks (macOS) Model benchmarks (macOS)
@ -166,12 +166,12 @@ GPU inference
|WD v1.4 ConvNeXTV2 v2 (batch)|1|160 s |WD v1.4 ConvNeXTV2 v2 (batch)|1|160 s
|WD v1.4 MOAT v2 (batch)|1|165 s |WD v1.4 MOAT v2 (batch)|1|165 s
|WD v1.4 SwinV2 v2|1|166 s |WD v1.4 SwinV2 v2|1|166 s
|ML-Danbooru Caformer dec-5-97527|1|263 s
|WD v1.4 ConvNeXT v2|1|273 s |WD v1.4 ConvNeXT v2|1|273 s
|WD v1.4 MOAT v2|1|273 s |WD v1.4 MOAT v2|1|273 s
|WD v1.4 ConvNeXTV2 v2|1|340 s |WD v1.4 ConvNeXTV2 v2|1|340 s
|ML-Danbooru Caformer dec-5-97527|1|551 s |ML-Danbooru Caformer dec-5-97527|4|445 s
|ML-Danbooru Caformer dec-5-97527|4|swap hell |ML-Danbooru Caformer dec-5-97527|8|1790 s
|ML-Danbooru Caformer dec-5-97527|8|swap hell
|WD v1.4 MOAT v2 (batch)|4|kernel panic |WD v1.4 MOAT v2 (batch)|4|kernel panic
|=== |===
@ -189,11 +189,14 @@ CPU inference
|WD v1.4 SwinV2 v2 (batch)|1|98 s |WD v1.4 SwinV2 v2 (batch)|1|98 s
|ML-Danbooru TResNet-D 6-30000|4|99 s |ML-Danbooru TResNet-D 6-30000|4|99 s
|WD v1.4 SwinV2 v2|1|99 s |WD v1.4 SwinV2 v2|1|99 s
|ML-Danbooru Caformer dec-5-97527|4|110 s
|ML-Danbooru Caformer dec-5-97527|8|110 s
|WD v1.4 ViT v2 (batch)|4|111 s |WD v1.4 ViT v2 (batch)|4|111 s
|WD v1.4 ViT v2 (batch)|8|111 s |WD v1.4 ViT v2 (batch)|8|111 s
|WD v1.4 ViT v2 (batch)|1|113 s |WD v1.4 ViT v2 (batch)|1|113 s
|WD v1.4 ViT v2|1|113 s |WD v1.4 ViT v2|1|113 s
|ML-Danbooru TResNet-D 6-30000|1|118 s |ML-Danbooru TResNet-D 6-30000|1|118 s
|ML-Danbooru Caformer dec-5-97527|1|122 s
|WD v1.4 ConvNeXT v2 (batch)|8|124 s |WD v1.4 ConvNeXT v2 (batch)|8|124 s
|WD v1.4 ConvNeXT v2 (batch)|4|125 s |WD v1.4 ConvNeXT v2 (batch)|4|125 s
|WD v1.4 ConvNeXTV2 v2 (batch)|8|129 s |WD v1.4 ConvNeXTV2 v2 (batch)|8|129 s
@ -206,9 +209,6 @@ CPU inference
|WD v1.4 MOAT v2 (batch)|1|156 s |WD v1.4 MOAT v2 (batch)|1|156 s
|WD v1.4 MOAT v2|1|156 s |WD v1.4 MOAT v2|1|156 s
|WD v1.4 ConvNeXTV2 v2 (batch)|1|157 s |WD v1.4 ConvNeXTV2 v2 (batch)|1|157 s
|ML-Danbooru Caformer dec-5-97527|4|241 s
|ML-Danbooru Caformer dec-5-97527|8|241 s
|ML-Danbooru Caformer dec-5-97527|1|262 s
|=== |===
Comparison with WDMassTagger Comparison with WDMassTagger

10
deeptagger/download.sh
View File

@ -115,7 +115,7 @@ wd14() {
# These models are an undocumented mess, thus using ONNX preconversions. # These models are an undocumented mess, thus using ONNX preconversions.
mldanbooru() { mldanbooru() {
local name=$1 basename=$2 local name=$1 size=$2 basename=$3
status "$name" status "$name"
if ! [ -d ml-danbooru-onnx ] if ! [ -d ml-danbooru-onnx ]
@ -138,7 +138,7 @@ mldanbooru() {
channels=rgb channels=rgb
normalize=true normalize=true
pad=stretch pad=stretch
size=640 size=$size
interpret=sigmoid interpret=sigmoid
END END
} }
@ -157,5 +157,7 @@ wd14 'WD v1.4 SwinV2 v2' 'SmilingWolf/wd-v1-4-swinv2-tagger-v2'
wd14 'WD v1.4 MOAT v2' 'SmilingWolf/wd-v1-4-moat-tagger-v2' wd14 'WD v1.4 MOAT v2' 'SmilingWolf/wd-v1-4-moat-tagger-v2'
# As suggested by author https://github.com/IrisRainbowNeko/ML-Danbooru-webui # As suggested by author https://github.com/IrisRainbowNeko/ML-Danbooru-webui
mldanbooru 'ML-Danbooru Caformer dec-5-97527' 'ml_caformer_m36_dec-5-97527.onnx' mldanbooru 'ML-Danbooru Caformer dec-5-97527' \
mldanbooru 'ML-Danbooru TResNet-D 6-30000' 'TResnet-D-FLq_ema_6-30000.onnx' 448 'ml_caformer_m36_dec-5-97527.onnx'
mldanbooru 'ML-Danbooru TResNet-D 6-30000' \
640 'TResnet-D-FLq_ema_6-30000.onnx'