Update app.py

app.py

@@ -1,41 +1,27 @@
 import streamlit as st
-from transformers import AutoTokenizer, AutoModelForSequenceClassification
-import torch
+from transformers import pipeline
+from PIL import Image
+import req
 
-# Load the pre-trained model and tokenizer
-model_name = "j-hartmann/emotion-english-distilroberta-base"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSequenceClassification.from_pretrained(model_name)
+classifier = pipeline("image-classification", model="google/vit-base-patch16-224")
+def get_img_from_url(url):
+    return Image.open(requests.get(url, stream=True).raw)
 
-# Streamlit UI
-st.title("Text Classification App")
+def main():
+    st.title("image-classification")
 
-# Input text area for user input
-text = st.text_area("Enter text for classification:")
+    with st.form("image"):
+        
+        url = st.text_input("URL to some image", "https://images.livemint.com/img/2022/08/01/600x338/Cat-andriyko-podilnyk-RCfi7vgJjUY-unsplash_1659328989095_1659328998370_1659328998370.jpg")
+        img = get_img_from_url(url)
+        # clicked==True only when the button is clicked
+        clicked = st.form_submit_button("Submit")
+        if clicked:
+          results = classifier(img)
+          st.json(results)
 
-if text:
-    # Tokenize the input text
-    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
 
-    # Perform text classification
-    with torch.no_grad():
-        outputs = model(**inputs)
-
-    # Get the predicted label and score
-    label_id = torch.argmax(outputs.logits, dim=1).item()
-    label = model.config.id2label[label_id]
-    score = torch.softmax(outputs.logits, dim=1)[0, label_id].item()
-
-    # Display the classification result
-    st.subheader("Classification Result:")
-    st.write(f"Predicted Label: {label}")
-    st.write(f"Confidence Score: {score:.4f}")
-
-# Sidebar with model information
-st.sidebar.subheader("Model Information")
-st.sidebar.write(f"Model: {model_name}")
-st.sidebar.write(f"Number of Labels: {model.config.num_labels}")
-
-# Powered by Hugging Face
-st.write("Powered by Hugging Face's Transformers library.")
 
+    
+if _name_ == "_main_":
+    main()