Sunday, February 5, 2023
Show HN: Visualize ResNet Embeddings with UMAP https://ift.tt/k0npbFf
Show HN: Visualize ResNet Embeddings with UMAP I took my personal backup (mostly family photos, some random scans, etc), extracted the embeddings from ResNet (pen-ultimate layer activations), and then plotted them with the UMAP algorithm and bokeh library This was a one-afternoon learning exercise while doing the awesome Fast AI course. The results are quite fun. ResNet returns 512 feature and UMAP maps those features into 2D plane, preserving distances as much as possible. The outliers show that the embeddings "make sense", clustering the images that are similar to other ones in the cluster, and different from those outside the cluster. https://ift.tt/D9drLu7 February 6, 2023 at 04:15AM
Subscribe to:
Post Comments (Atom)
Show HN: Training a model to identify AI web content from structure alone https://ift.tt/OpNIVfk
Show HN: Training a model to identify AI web content from structure alone Hey HN! We’re Vincent and Jochen from Sitefire ( https://sitefire....
-
Show HN: WebApp that blows water out of your phone's speakers https://ift.tt/2VmXkOf March 16, 2021 at 08:22AM
-
Show HN: Music player for big local collections with mpd support mpz is a C++/Qt music player focused on UX, with derectory tree and playlis...
-
Show HN: Stickerbox, a kid-safe, AI-powered voice to sticker printer Bob and Arun here, creators of Stickerbox. If AI were built for kids, w...
No comments:
Post a Comment