Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,20 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
# Contrastive Booru-Image Embedding Model
|
| 6 |
+
|
| 7 |
+
The ViT-B model was replaced by another [Encoder](https://huggingface.co/nightknocker/u-dae-nll).
|
| 8 |
+
|
| 9 |
+
The images were tagged by the [CL tagger](https://huggingface.co/cella110n/cl_tagger).
|
| 10 |
+
The caption (prompt) was tokenized by [ModernBERT](https://huggingface.co/answerdotai/ModernBERT-base).
|
| 11 |
+
|
| 12 |
+
## Objectives
|
| 13 |
+
|
| 14 |
+
*Image-Caption Similarity Scoring*: The model assigns a score to each image based on how closely it is related to its associated caption.
|
| 15 |
+
|
| 16 |
+
*Image Labeling*: The image embedding is used to label images by character name.
|
| 17 |
+
|
| 18 |
+
## Updates
|
| 19 |
+
|
| 20 |
+
This version contains up to 5000 characters extracted from public and private datasets.
|