fix: use AutoImageProcessor

app.py

@@ -5,7 +5,7 @@ import numpy as np
 import torch
 
 from transformers import (
-    AutoProcessor,
+    AutoImageProcessor,
 )
 
 import gradio as gr
@@ -20,7 +20,7 @@ EXAMPLES = [["./images/sample.jpg"], ["./images/sample2.webp"]]
 
 model = SiglipForImageClassification.from_pretrained(MODEL_NAME, token=HF_TOKEN)
 # model = torch.compile(model)
-processor = AutoProcessor.from_pretrained(PROCESSOR_NAME, token=HF_TOKEN)
+processor = AutoImageProcessor.from_pretrained(PROCESSOR_NAME, token=HF_TOKEN)
 
 
 def compose_text(results: dict[str, float], threshold: float = 0.3):
@@ -35,7 +35,7 @@ def compose_text(results: dict[str, float], threshold: float = 0.3):
 
 @torch.no_grad()
 def predict_tags(image: Image.Image, threshold: float):
-    inputs = processor(images=image, return_tensors="pt")
+    inputs = processor(image, return_tensors="pt")
 
     logits = model(**inputs.to(model.device, model.dtype)).logits.detach().cpu()