Difference between revisions of "Experiments/GenAI"

From Open Food Facts wiki
Jump to navigation Jump to search
 
(8 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
[https://en.wikipedia.org/wiki/Generative_artificial_intelligence Generative AI] is the field of Artificial intelligence that generates contents (text, images) from prompts or contents.
 +
[[Category:GenAI]]
 +
[[Category:Artificial Intelligence]]
 +
The resulting technologies might have interesting usage for Open Food Facts.
 +
 +
This page list useful resources.
 +
 
=== From Open Food Facts ===
 
=== From Open Food Facts ===
* https://github.com/openfoodfacts/openfoodfacts-ai/issues/289
+
* Describe me here: https://github.com/openfoodfacts/openfoodfacts-ai/issues/289
*https://openfoodfacts.slack.com/archives/CF0286BRQ/p1705611584634089
+
* Describe me here: https://openfoodfacts.slack.com/archives/CF0286BRQ/p1705611584634089
*https://openfoodfacts.slack.com/archives/CF0286BRQ/p1702632877529589
+
* Describe me here: https://openfoodfacts.slack.com/archives/CF0286BRQ/p1702632877529589
*https://openfoodfacts.slack.com/archives/CF0286BRQ/p1702322101115109
+
* Describe me here: https://openfoodfacts.slack.com/archives/CF0286BRQ/p1702322101115109
*https://openfoodfacts.slack.com/archives/C1UNFJVQV/p1712323105541929 - Henadzi Rabkin I have started fine tuning with LLaVA back then in February, then postponed the project. Here's a link on my colab <nowiki>https://colab.research.google.com/drive/1REw1PdZLg1Wy3MAGORIGVe1qIhxn_V2x?usp=sharing</nowiki> - TIMG: https://openfoodfacts.slack.com/archives/C1UNFJVQV/p1709093097394419
+
* Describe me here: https://openfoodfacts.slack.com/archives/C1UNFJVQV/p1712323105541929 - Henadzi Rabkin I have started fine tuning with LLaVA back then in February, then postponed the project. Here's a link on my colab <nowiki>https://colab.research.google.com/drive/1REw1PdZLg1Wy3MAGORIGVe1qIhxn_V2x?usp=sharing</nowiki> - TIMG: https://openfoodfacts.slack.com/archives/C1UNFJVQV/p1709093097394419
*Segmentation: Vince I had to use [[/huggingface.co/briaai/RMBG-1.4|https://huggingface.co/briaai/RMBG-1.4]] in another context, and I was quite impressed with the results. It works quite well on our own use-case too. Here are a few examples:
 
*[[File:Segmentation test by Vince.png|thumb]][[File:Test by Vince of Segmentation.png|thumb]]
 
  
 
===From elsewhere===
 
===From elsewhere===
*[https://artsandculture.google.com/experiment/food-mood/HwHnGalZ3up0EA Google Experiment - Mix 2 cuisines together and create wacky recipes]
+
==== Food ====
*interesting concept, quite disruptive of OFF: https://foodanalyze.com/ - no permanent storage, no barcode orientation and the associated HackerNews thread: https://news.ycombinator.com/item?id=41259425
+
* [https://artsandculture.google.com/experiment/food-mood/HwHnGalZ3up0EA Google Experiment - Mix 2 cuisines together and create wacky recipes]
*https://openfoodfacts.slack.com/archives/C02VDSWHT/p1703301573620989
+
* interesting concept, quite disruptive of Open Food Facts: https://foodanalyze.com/ - no permanent storage, no barcode orientation and the associated HackerNews thread: https://news.ycombinator.com/item?id=41259425
 +
* https://openfoodfacts.slack.com/archives/C02VDSWHT/p1703301573620989
 
* https://engineering.hellofresh.com/recipes-and-generative-ai-6d74a107860c
 
* https://engineering.hellofresh.com/recipes-and-generative-ai-6d74a107860c
* FYI , OpenAI has deployed image input, and interesting use cases are emerging, https://twitter.com/jackloganjohns/status/1712600735366393912?t=R8PYuJuEPSPi9-s6JYj2mQ&s=19
+
* OpenAI has deployed image input, and interesting use cases are emerging, https://twitter.com/jackloganjohns/status/1712600735366393912?t=R8PYuJuEPSPi9-s6JYj2mQ&s=19
* Accidentally found today this dataset, might be helpful, see "Intended use " section https://agdatacommons.nal.usda.gov/articles/dataset/SNAPMe_A_Benchmark_Dataset_of_Food_Photos_with_Food_Records_for_Evaluation_of_Computer_Vision_Algorithms_in_the_Context_of_Dietary_Assessment/24856449  
+
* This dataset might be helpful (see "Intended use " section): https://agdatacommons.nal.usda.gov/articles/dataset/SNAPMe_A_Benchmark_Dataset_of_Food_Photos_with_Food_Records_for_Evaluation_of_Computer_Vision_Algorithms_in_the_Context_of_Dietary_Assessment/24856449  
 
[[File:Open AI demo with Pasta - Be My Eyes.png|thumb]]
 
[[File:Open AI demo with Pasta - Be My Eyes.png|thumb]]
 
* Carrefour playing with virtual avatars and ChatGPT for the Nutri-Score.  https://lehub.laposte.fr/repere/chatgpt-carrefour-cree-son-premier-avatar-grace-intelligence-artificielle-generative
 
* Carrefour playing with virtual avatars and ChatGPT for the Nutri-Score.  https://lehub.laposte.fr/repere/chatgpt-carrefour-cree-son-premier-avatar-grace-intelligence-artificielle-generative
*(Open Products Facts) Interesting Demo from Google about troubleshooting a record player: https://youtu.be/XEzRZ35urlk?t=3031
+
* Interesting demo from Google about meal planning: https://youtu.be/XEzRZ35urlk?t=2852
*Interesting demo from Google about meal planning: https://youtu.be/XEzRZ35urlk?t=2852
+
* Circle to search on food: https://youtu.be/XEzRZ35urlk?t=4814
*Circle to search on food: https://youtu.be/XEzRZ35urlk?t=4814
+
* a recap video with a Microwave meal planning on a budget (cc @Raphael Odini) https://youtu.be/XEzRZ35urlk?t=3219 and a Camera troubleshooting (Open Products Facts)
*a recap video with a Microwave meal planning on a budget (cc @Raphael Odini) https://youtu.be/XEzRZ35urlk?t=3219 and a Camera troubleshooting (Open Products Facts)
+
* Climate Q&A chatGPT [[https://huggingface.co/spaces/Ekimetrics/climate-question-answering|https://huggingface.co/spaces/Ekimetrics/climate-question-answering]] [[File:Assistant Memory Peanuts.png|thumb]]Assistants are going to try and remember you are allergic to peanuts.
*Climate Q&A chatGPT [[/huggingface.co/spaces/Ekimetrics/climate-question-answering|https://huggingface.co/spaces/Ekimetrics/climate-question-answering]] [[File:Assistant Memory Peanuts.png|thumb]]Assistants are going to try and remember you are allergic to peanuts.
+
==== Open Products Facts ====
 +
* Interesting Demo from Google about troubleshooting a record player: https://youtu.be/XEzRZ35urlk?t=3031
 +
 
 +
=== Client side LLMs ===
 +
* ​​https://medium.com/@kavinask007/pitfalls-in-trying-to-run-mistral-7b-model-locally-in-browser-using-rust-wasm-4668c129abd0
 +
* upcoming feature of the next version of Chrome: embedded local LLM: https://huggingface.co/blog/Xenova/run-gemini-nano-in-your-browser
 +
* https://developer.chrome.com/docs/ai
 +
* https://news.ycombinator.com/item?id=40834600
 +
* 65% of the userbase worldwide will get it, so it might be worth investigating and see if it’s useful enough for spellcheck
 +
 
 +
=== Opensource multimodal ===
 +
* paligemma (opensource multimodal)
 +
* See https://llava.hliu.cc/ 2023
 +
 
 +
=== Prompt database ===
 +
* Open Food Facts prompt database: https://docs.google.com/spreadsheets/d/1Iq4_vJcPEuCNQa9wa5aX3sES_67q5MJZT9-3E3SX8g4/edit?gid=808139179#gid=808139179

Latest revision as of 12:22, 27 August 2024

Generative AI is the field of Artificial intelligence that generates contents (text, images) from prompts or contents. The resulting technologies might have interesting usage for Open Food Facts.

This page list useful resources.

From Open Food Facts

From elsewhere

Food

Open AI demo with Pasta - Be My Eyes.png

Open Products Facts

Client side LLMs

Opensource multimodal

Prompt database