Update app.py

app.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from transformers import CLIPModel, CLIPProcessor
+from transformers import CLIPModel, CLIPProcessor, pipeline
 import torch
 from PIL import Image
 
@@ -35,20 +35,31 @@ def inference_clip(options, image, processor, model):
 #################################
 #### LAYOUT
 
-CLIP_large = load_clip(model_size='large')
+#CLIP_large = load_clip(model_size='large')
+model_name = "openai/clip-vit-large-patch14-336"
+classifier = pipeline("zero-shot-image-classification", model = model_name)
 
+#### Loading picture
 picture_file = st.file_uploader("Picture :", type=["jpg", "jpeg", "png"])
 if picture_file is not None:
     image = Image.open(picture_file)
     st.image(image, caption='Please upload an image of the damage', use_column_width=True)
 
+col_l, col_r = st.columns(2)
+
 #image
-default_options = ['black', 'white', 'gray', 'red', 'blue', 'silver', 'red', 'brown', 'green', 'orange', 'beige', 'pruple', 'gold', 'yellow']
-options = st.text_input(label="Please enter the classes", value=default_options)
-#options = list(options)
-
-# button to launch compute
-if st.button("Compute"):
-    clip_processor, clip_model = load_clip(model_size='large')
-    result = inference_clip(options = options, image = image, processor=clip_processor, model=clip_model)
-    st.write(result)
+with col_l:
+    default_options = ['black', 'white', 'gray', 'red', 'blue', 'silver', 'red', 'brown', 'green', 'orange', 'beige', 'pruple', 'gold', 'yellow']
+    options = st.text_input(label="Please enter the classes", value=default_options)
+    #options = list(options)
+
+    # button to launch compute
+    if st.button("Compute"):
+        #clip_processor, clip_model = load_clip(model_size='large')
+        #result = inference_clip(options = options, image = image, processor=clip_processor, model=clip_model)
+        scores = classifier(image, 
+                    candidate_labels = options)
+
+with col_r:
+    #st.write(result)
+    st.dataframe(scores)