Update app.py

app.py

@@ -6,7 +6,8 @@ from transformers import pipeline
 
 asr = pipeline("automatic-speech-recognition")
 latent = gr.Interface.load("spaces/multimodalart/latentdiffusion")
-zero = gr.Interface.load("spaces/Datatrooper/zero-shot-image-classification")
+zero = pipeline("zero-shot-image-classification", model="openai/clip-vit-base-patch32")
+#zero = gr.Interface.load("spaces/Datatrooper/zero-shot-image-classification")
 #tts = gr.Interface.load("spaces/osanseviero/tortoisse-tts")
 
 def text2image_latent(text, steps, width, height, images, diversity):
@@ -39,8 +40,20 @@ def speech_to_text(mic=None, file=None):
     return transcription
 
 def zero_shot(image, text_input):
-    results = zero(image, text_input)
-    return results
+    PIL_image = Image.fromarray(np.uint8(image)).convert('RGB')
+    labels = labels_text.split(",")
+    res = pipe(images=PIL_image, 
+           candidate_labels=labels,
+           hypothesis_template= "This is a photo of a {}")
+    return {dic["label"]: dic["score"] for dic in res}
+
+def shot(image, labels_text):
+    PIL_image = Image.fromarray(np.uint8(image)).convert('RGB')
+    labels = labels_text.split(",")
+    res = pipe(images=PIL_image, 
+           candidate_labels=labels,
+           hypothesis_template= "This is a photo of a {}")
+    return {dic["label"]: dic["score"] for dic in res}
     
 with gr.Blocks() as demo: 
     gr.Markdown( """