weaviate/agents
Viewer
โข
Updated
โข
2.7k
โข
244
โข
3
Tuana Celik
Tuana
AI & ML interests
๐ฅ Haystack
Gen AI, LLMs, NLP
Recent Activity
updated
a dataset
2 days ago
weaviate/agents
liked
a model
8 months ago
mixedbread-ai/deepset-mxbai-embed-de-large-v1
upvoted
an
article
8 months ago
Mixedbread ๐ค deepset: Announcing our New German/English Embedding Model
Organizations
Tuana's activity
updated
a
dataset
2 days ago
liked
a
model
8 months ago
upvoted
an
article
8 months ago
Article
Mixedbread ๐ค deepset: Announcing our New German/English Embedding Model
By
โข
โข
15
liked
a
Space
9 months ago
reacted to
merve's
post with ๐
9 months ago
Post
4225
I love Depth Anything V2 ๐
Itโs Depth Anything, but scaled with both larger teacher model and a gigantic dataset!
Here's a small TLDR of paper with a lot of findings, experiments and more.
I have also created a collection that has the models, the dataset, the demo and CoreML converted model ๐ merve/depth-anything-v2-release-6671902e798cd404513ffbf5
The authors have analyzed Marigold, a diffusion based model against Depth Anything and found out whatโs up with using synthetic images vs real images for MDE:
๐ Real data has a lot of label noise, inaccurate depth maps (caused by depth sensors missing transparent objects etc) and there are many details overlooked
๐ Synthetic data have more precise and detailed depth labels and they are truly ground-truth, but thereโs a distribution shift between real and synthetic images, and they have restricted scene coverage
The authors train different image encoders only on synthetic images and find out unless the encoder is very large the model canโt generalize well (but large models generalize inherently anyway) ๐ง
But they still fail encountering real images that have wide distribution in labels (e.g. diverse instances of objects) ๐ฅฒ
Depth Anything v2 framework is to..
๐ฆ Train a teacher model based on DINOv2-G based on 595K synthetic images
๐ท๏ธ Label 62M real images using teacher model
๐ฆ Train a student model using the real images labelled by teacher
Result: 10x faster and more accurate than Marigold!
The authors also construct a new benchmark called DA-2K that is less noisy, highly detailed and more diverse!
Itโs Depth Anything, but scaled with both larger teacher model and a gigantic dataset!
Here's a small TLDR of paper with a lot of findings, experiments and more.
I have also created a collection that has the models, the dataset, the demo and CoreML converted model ๐ merve/depth-anything-v2-release-6671902e798cd404513ffbf5
The authors have analyzed Marigold, a diffusion based model against Depth Anything and found out whatโs up with using synthetic images vs real images for MDE:
๐ Real data has a lot of label noise, inaccurate depth maps (caused by depth sensors missing transparent objects etc) and there are many details overlooked
๐ Synthetic data have more precise and detailed depth labels and they are truly ground-truth, but thereโs a distribution shift between real and synthetic images, and they have restricted scene coverage
The authors train different image encoders only on synthetic images and find out unless the encoder is very large the model canโt generalize well (but large models generalize inherently anyway) ๐ง
But they still fail encountering real images that have wide distribution in labels (e.g. diverse instances of objects) ๐ฅฒ
Depth Anything v2 framework is to..
๐ฆ Train a teacher model based on DINOv2-G based on 595K synthetic images
๐ท๏ธ Label 62M real images using teacher model
๐ฆ Train a student model using the real images labelled by teacher
Result: 10x faster and more accurate than Marigold!
The authors also construct a new benchmark called DA-2K that is less noisy, highly detailed and more diverse!
liked
a
dataset
9 months ago
reacted to
anakin87's
post with ๐ฅ
10 months ago
Post
1278
Do you want to play a game against Llama 3? ๐ฆ๐ฆ๐ฆ
Meet ๐งโ๐ซ ๐๐ฎ๐ญ๐จ๐๐ฎ๐ข๐ณ๐ณ๐๐ซ, a new LLM application that you can use for learning or just for fun.
Try it out on Hugging Face Spaces ๐ค deepset/autoquizzer
๐๐จ๐ฐ ๐ข๐ญ ๐ฐ๐จ๐ซ๐ค๐ฌ
You provide an URL -> A multiple-choice quiz is instantly generated.
๐น You can play the quiz yourself.
๐น You can let the LLM play in two different ways
๐ Closed book: the LLM responds only by knowing the general topic and using its parametric knowledge and reasoning abilities.
๐๐ Web RAG: for each question, a Google search is done and the top 3 snippets are included in the prompt for the LLM.
๐๐ญ๐๐๐ค
๐๏ธ Haystack LLM framework https://haystack.deepset.ai/
๐ฆ Llama 3 8B Instruct
โก Groq
Original idea: @Tuana
Meet ๐งโ๐ซ ๐๐ฎ๐ญ๐จ๐๐ฎ๐ข๐ณ๐ณ๐๐ซ, a new LLM application that you can use for learning or just for fun.
Try it out on Hugging Face Spaces ๐ค deepset/autoquizzer
๐๐จ๐ฐ ๐ข๐ญ ๐ฐ๐จ๐ซ๐ค๐ฌ
You provide an URL -> A multiple-choice quiz is instantly generated.
๐น You can play the quiz yourself.
๐น You can let the LLM play in two different ways
๐ Closed book: the LLM responds only by knowing the general topic and using its parametric knowledge and reasoning abilities.
๐๐ Web RAG: for each question, a Google search is done and the top 3 snippets are included in the prompt for the LLM.
๐๐ญ๐๐๐ค
๐๏ธ Haystack LLM framework https://haystack.deepset.ai/
๐ฆ Llama 3 8B Instruct
โก Groq
Original idea: @Tuana
-
- 1 reply
Thanks for sharing @anakin87 โค๏ธ Really cool demo!
reacted to
anakin87's
post with โค๏ธ
10 months ago
Post
1278
Do you want to play a game against Llama 3? ๐ฆ๐ฆ๐ฆ
Meet ๐งโ๐ซ ๐๐ฎ๐ญ๐จ๐๐ฎ๐ข๐ณ๐ณ๐๐ซ, a new LLM application that you can use for learning or just for fun.
Try it out on Hugging Face Spaces ๐ค deepset/autoquizzer
๐๐จ๐ฐ ๐ข๐ญ ๐ฐ๐จ๐ซ๐ค๐ฌ
You provide an URL -> A multiple-choice quiz is instantly generated.
๐น You can play the quiz yourself.
๐น You can let the LLM play in two different ways
๐ Closed book: the LLM responds only by knowing the general topic and using its parametric knowledge and reasoning abilities.
๐๐ Web RAG: for each question, a Google search is done and the top 3 snippets are included in the prompt for the LLM.
๐๐ญ๐๐๐ค
๐๏ธ Haystack LLM framework https://haystack.deepset.ai/
๐ฆ Llama 3 8B Instruct
โก Groq
Original idea: @Tuana
Meet ๐งโ๐ซ ๐๐ฎ๐ญ๐จ๐๐ฎ๐ข๐ณ๐ณ๐๐ซ, a new LLM application that you can use for learning or just for fun.
Try it out on Hugging Face Spaces ๐ค deepset/autoquizzer
๐๐จ๐ฐ ๐ข๐ญ ๐ฐ๐จ๐ซ๐ค๐ฌ
You provide an URL -> A multiple-choice quiz is instantly generated.
๐น You can play the quiz yourself.
๐น You can let the LLM play in two different ways
๐ Closed book: the LLM responds only by knowing the general topic and using its parametric knowledge and reasoning abilities.
๐๐ Web RAG: for each question, a Google search is done and the top 3 snippets are included in the prompt for the LLM.
๐๐ญ๐๐๐ค
๐๏ธ Haystack LLM framework https://haystack.deepset.ai/
๐ฆ Llama 3 8B Instruct
โก Groq
Original idea: @Tuana
-
- 1 reply
updated
a
Space
12 months ago
This space has been ported to Haystack 2.0
#8 opened about 1 year ago
by
Tuana
updated
3
Spaces
about 1 year ago
liked
a
Space
about 1 year ago
liked
a
Space
over 1 year ago
updated
a
Space
over 1 year ago
liked
2
Spaces
over 1 year ago
Experiences with complex instructions
1
#50 opened over 1 year ago
by
Tuana