Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- mobileclip installation by default - Merge pull request raycast#15 from Andrew-Ayman123/main - Added PowerToys Run Plugin - powertoys run plugin - removing autocast from mobile CLIP - removing custom installation - -fixing torch installation -increase settings height - fixing issue in torch cuda installation - prevent reinstall torch if already installed - updating requirements.txt - -fixing bug in args - fixing bug in settings gui - adding topk and threshold to server, raycast and webui - new setup - remvove text and pdf notebook - support metal - -updating create_db with new embedding providers -updated requirements.txt file - dropping wsl support - multithreaded scan - Settings UI + new config with new embedding providers - text embeddings now support hugging face transformers, Llama.cpp, Ollama and any OpenAI Compatible API - adding threshold and top_k to search functions - debouncing to avoid execssive requests - raising exception of Ollama isn\'t initalized - - fixing issue in text embeddings data base - adding verbose to clean defaulted False - fixing an error in the text embeddings collection - fixing typo - chaning OCR dir using batch OCR in vector db - restoring main.py to improve memory - moving OCR inside dir - batched OCR and parallel IO & post processing - batched vector database - batch processing of images in HF transformers CLIP (both model and io) - batch processing of images in mobileclip (both model and io) - fixing error in deep_scan disabled - fixing error in flow launcher - removing include paths from Everything - using PIL instead of matplotlib fixing error in OCR - updating to the new port - updating HTML for new API - updating config for deep_scan - new scan functionality - Search image can work with any image locally or from web - updating extensions to the new API - using query for all routes - raycast extension - notebook for text files and documents - change default config - reading paths from sql Changing port number to 23107 to avoid conflicts - recovering original everything - commenting include /exclude functionality - scanning given directories - removing extra lines - adding notes - supporting text embeddings from ollama - removing unecessary files - Merge pull request raycast#11 from yousefmrashad/everything-include-exclude - Add include/exclude folders functionality to everything image retrieval. Fix: exclude the recycle bin folder in code. - chaning webui route from /index to / - Merge pull request raycast#10 from yousefmrashad/utf-16-paths - Remove redundant code. - Used "UTF-16" to read image file paths for better compatibility. - updating html readme - better UI - native OS theme - Merge pull request raycast#9 from t0saki/feat/serve-index-webpage-with-server - handling wsl server - updating gif - feat: Add web interface URL to server startup message - restore img list source - add index to server - updating readme - setting default settings - adding configs to everything - adding include / execlude folders to config - adding license - adding future work to issues - adding video link - updating READMEs - fixing title - adding webui - flow launcher readme - updating README - updating docstrings - removing image itself when searching by image - fixing bug in image search functionality - adding search by image functionality - pip package - Everything SDK now downloads by default - removing test notebook - v0.1 config not added yet - main embedding and OCR Files - first commit
- Loading branch information