Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,12 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
pipeline_tag: zero-shot-classification
|
| 5 |
---
|
| 6 |
+
You probably do not need this unless you are training your own IP Adapters.
|
| 7 |
+
|
| 8 |
+
Modified version of the vision encoder of [CLIP-ViT-H-14-laion2B-s32B-b79K](https://huggingface.co/laion/CLIP-ViT-H-14-laion2B-s32B-b79K) to handle 448 x 448 inputs
|
| 9 |
+
vs the original 224 x 224 inputs. It will probbaly not work for classification (as is), but will DIP work for for IP+ adapters that use CLIP-ViT-H, though they will need to be
|
| 10 |
+
fine tuned a little more.
|
| 11 |
+
|
| 12 |
+
Hidden layer outputs go from `(257, 1280)` to `(1025, 1280), which can be digested by the Resampler without modification or weight resizing.
|