AbstractPhil commited on
Commit
2a57df1
·
verified ·
1 Parent(s): 46e4939

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -0
README.md CHANGED
@@ -25,6 +25,11 @@ I used the 32 dim; as it seemed to be the weakest with flow-match euler-discreet
25
 
26
  The output model is much larger than I wanted; which defeats the purpose of the overall structure - but it's paired directly at the knee with clip-vit-base-patch32, so I'll prepare a decoupled version here in a bit.
27
 
 
 
 
 
 
28
  # Notebook-6 · Crystal-CLIP CIFAR-100
29
 
30
  One-vector image embeddings (HF CLIP) + pentachora vocabulary anchors → cosine-similarity classifier for CIFAR-100.
 
25
 
26
  The output model is much larger than I wanted; which defeats the purpose of the overall structure - but it's paired directly at the knee with clip-vit-base-patch32, so I'll prepare a decoupled version here in a bit.
27
 
28
+
29
+ # Why clip-vit instead of just vit?
30
+
31
+ I believe the clip-vit variations have more utility overall so I wanted to ensure a fair target was assessed.
32
+
33
  # Notebook-6 · Crystal-CLIP CIFAR-100
34
 
35
  One-vector image embeddings (HF CLIP) + pentachora vocabulary anchors → cosine-similarity classifier for CIFAR-100.