fix

requirements.txt

@@ -1,7 +1,6 @@
 transformers==4.27.4
 sentencepiece==0.1.97
 sacremoses==0.0.53
-clip-interrogator==0.6.0
 torch==2.0.0
 gradio==3.24.1
 pillow>=9.0.0
@@ -10,3 +9,4 @@ onnxruntime>=1.12.0
 protobuf<=3.20.1,>=3.12.2
 opencv-python==4.7.0.72
 huggingface-hub==0.13.2
+clip-interrogator==0.6.0

utils/image2text.py

@@ -30,8 +30,8 @@ class Models(object):
     LABEL_FILENAME = "selected_tags.csv"
 
     # CLIP models
-    VIT_H_14_MODEL_REPO = "laion/CLIP-ViT-H-14-laion2B-s32B-b79K"  # Stable Diffusion 2.X
-    VIT_L_14_MODEL_REPO = "openai/clip-vit-large-patch14"  # Stable Diffusion 1.X
+    VIT_H_14_MODEL_REPO = "ViT-H-14/laion2b_s32b_b79k"  # Stable Diffusion 2.X
+    VIT_L_14_MODEL_REPO = "ViT-L-14/openai"  # Stable Diffusion 1.X
 
     def __init__(self):
         pass