mirror of
https://github.com/openai/CLIP.git
synced 2026-11-05 01:37:26 +00:00
[GH-ISSUE #174] Emoji in Text #153
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
openai/CLIP#153
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @steveguang on GitHub (Nov 5, 2021).
Original GitHub issue: https://github.com/openai/CLIP/issues/174
Is clip model able to process emoji? or text and images only? I tried to encode two emojis '😀' and '😃'. It generates two embeddings that are different. they are like,
array([[5.50842285e-02, 2.57812500e-01, 4.94384766e-02, .....
array([[ 7.32421875e-02, 3.23242188e-01, 1.26586914e-01,......
Any reason why? It doesnt seem to be random
@jongwook commented on GitHub (Nov 16, 2021):
The BPE encoding supports arbitrary UTF-8 sequence, and it can encode and decode emoji characters accordingly. It'd be interesting project to plot the t-SNE of emoji embeddings.