From 666f8c27935514a41ac854afc743a6d27840253f Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:32:54 +0900 Subject: [PATCH 1/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 007bded..14d31bc 100644 --- a/README.md +++ b/README.md @@ -18,7 +18,7 @@ - $ python make-tags-with-wd-tagger.py --dir "IMAGE FILES CONTAINED DIR PATH" - The script searches directory structure recursively :) - This takes quite a while... - - About 1 file/s at middle spec desktop PC (GPU is not used) + - About 0.5 sec/file at middle spec desktop PC (GPU is not used) - AMD Ryzen 7 5700X 8-Core Processor 4.50 GHz - You may speed up with editing the script to use CUDAExecutionProvider, CoreMLExecutionProvider and etc :) - Plese see [here](https://onnxruntime.ai/docs/execution-providers/) From b5577d92644c4f4eb41ad839d803e165a6783622 Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:36:45 +0900 Subject: [PATCH 2/7] Update README.md --- README.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 14d31bc..780655b 100644 --- a/README.md +++ b/README.md @@ -6,10 +6,10 @@ - So, I wrote simple scripts ## Method -- Search Images matching with Query Texts on Latent Representation Vectors +- Search Images matching with Query Texts on Latent Semantic Representation Vector Space - Vectors are generated with embedding model: Visual Transformar (ViT) Tagger x Latent Semantic Indexing (LSI) -- LSI is Ssed for Covering Tagging Presision - - You can use tags to search which are difficult for tagging because search index is applyed LSI +- LSI is used for Covering Tagging Presision + - You can use tags to search which are difficult for tagging because the index data which is composed of vectors is applyed LSI - implemented with Gensim lib - ( Web UI is implemented with StreamLit ) From 4b068f1da3d970eccd782cd686015896ce4a11b8 Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:38:49 +0900 Subject: [PATCH 3/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 780655b..3c6babb 100644 --- a/README.md +++ b/README.md @@ -55,7 +55,7 @@ ## TODO - [ ] Search on latent representation generated by CLIP model - - This was tried but precition with current public available CLIP models which are not fit for anime style illust was bad :| + - This was alredy tried but precition was not good because current public available CLIP models are not fitting for anime style illust :| - [ ] Weight specifying to keyword like prompt format of Stable Diffusion Web UI - Current implemenataion uses all keywords faialy. But there is many cases that users want to emphasize specific keyword and can't get appropriate results without that! - [ ] Incremental index updating at image files increasing From 03936eb6d8f22fa034ae6a087d42a8a18244d398 Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:52:50 +0900 Subject: [PATCH 4/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 3c6babb..8132349 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,4 @@ -# Anime Style Illustration Specific Image Search App with Vit Tagger x LSI +# Anime Style Illustration Specific Image Search App with ViT Tagger x LSI ## What's This? - Anime Style Illustration Specific Image Search App with ML Technique - can be used for photos. but flexible photo search is offered by Google Photos or etc :) From efc91fedcf6c52b283e8a6e6d9325b247c9d6b1a Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:54:20 +0900 Subject: [PATCH 5/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 8132349..46c5e04 100644 --- a/README.md +++ b/README.md @@ -43,7 +43,7 @@ - Solution - Search words you want to use from taggs-wd-tagger.txt with grep, editor or something for existance checking - If exist, there is no problem. If not, you should think similar words and search it in same manner :) -- Charcter code of file pathes +- Character code of file pathes - If file path contains charactors which can't be contered to Unicode or utf-8, scripts may ouput error message at processing the file - But, it doesn't mean that your script usage is wrong. Though these files is ignored or not displayed at Web UI :| - This is problem of current implentation. When you use scripts on Windows and charactor code of directory/file names isn't utf-8, the problem may occur From 0fbee10fbfac1171a93da4efc38fe006d4758b41 Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:55:46 +0900 Subject: [PATCH 6/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 46c5e04..99f2ee3 100644 --- a/README.md +++ b/README.md @@ -59,7 +59,7 @@ - [ ] Weight specifying to keyword like prompt format of Stable Diffusion Web UI - Current implemenataion uses all keywords faialy. But there is many cases that users want to emphasize specific keyword and can't get appropriate results without that! - [ ] Incremental index updating at image files increasing -- [ ] Similar image search with specifying target image file +- [ ] Similar image search with specifying a image file - [ ] Exporting found files list feature - In text file. Once you get list, many other tools and viewer you like can be used :) - [ ] Making binary package of this app which doesn't need python environment building From c46d3d72de46ca34fda36e6626a7ce3c1cd89c65 Mon Sep 17 00:00:00 2001 From: Ryo Kanbayashi Date: Mon, 14 Oct 2024 13:58:15 +0900 Subject: [PATCH 7/7] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 99f2ee3..df4da1c 100644 --- a/README.md +++ b/README.md @@ -6,7 +6,7 @@ - So, I wrote simple scripts ## Method -- Search Images matching with Query Texts on Latent Semantic Representation Vector Space +- Search Images Matching with Query Texts on Latent Semantic Representation Vector Space - Vectors are generated with embedding model: Visual Transformar (ViT) Tagger x Latent Semantic Indexing (LSI) - LSI is used for Covering Tagging Presision - You can use tags to search which are difficult for tagging because the index data which is composed of vectors is applyed LSI