Adds FiftyOne Art Analysis Recipe to Cookbook #89
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
notebooks/en/_toctree.yml
Outdated
| @@ -29,6 +29,8 @@ | |||
|
|
|||
| - title: Multimodal Recipes | |||
| sections: | |||
| - local: analyzing_art_with_hf_and_fiftyone | |||
| title: Analyzing Artistic Styles with 🤗 Transformers, 🤗 Hub, and FiftyOne | |||
There was a problem hiding this comment.
I think it would be better if you could express something around the modality here, e.g. Analyzing Artworks using Multimodal Embeddings
I'd like @stevhliu to leave an opinion as well
| @@ -0,0 +1,851 @@ | |||
| { | |||
There was a problem hiding this comment.
| @@ -0,0 +1,851 @@ | |||
| { | |||
There was a problem hiding this comment.
I think multimodal is a very broad term here, the word you're looking for is image-text encoder models IMO (SigLIP, EvaCLIP etc) you could say CLIP-like models
Reply via ReviewNB
| @@ -0,0 +1,851 @@ | |||
| { | |||
There was a problem hiding this comment.
Maybe give single link to overall integrations page not to break readability or add that at the end in something like useful resources
Reply via ReviewNB
| @@ -0,0 +1,851 @@ | |||
| { | |||
There was a problem hiding this comment.
| @@ -0,0 +1,851 @@ | |||
| { | |||
There was a problem hiding this comment.
|
fiftyone seems to be a very cool app @jacobmarks, thanks a lot for contributing this 🤗 |
|
Thanks for the review @merveenoyan ! I've made the requested changes :) Let me know what else I can do. Grateful for your guidance. |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
| @@ -0,0 +1,838 @@ | |||
| { | |||
There was a problem hiding this comment.
Small typo here (italicized so you can see it easier) :)
"This will allow us to see how the images cluster based on their style, genre, and artist."
Reply via ReviewNB
Thanks @stevhliu, really appreciate it :) |
This PR demonstrates how you can use a combination of traditional image processing, multimodal embeddings, unsupervised learning, and visualization to bring structure to unstructured visual data. The notebook illustrates these concepts in the context of art, using the WikiArt dataset from the Hugging Face Hub.
Fixes # (issue)
Who can review?
Feel free to tag members/contributors who may be interested in your PR.