Improve code

app.py

@@ -1,23 +1,10 @@
 import gradio as gr
-import requests
-from numpy import asarray
-import tensorflow as tf
+from PIL import Image
 from transformers import pipeline
 
-inception_net = tf.keras.applications.MobileNetV2()
-answer = requests.get("https://git.io/JJkYN")
-labels =answer.text.split("\n")
-
 transcribe = pipeline("automatic-speech-recognition", model = "facebook/wav2vec2-large-xlsr-53-spanish")
 classifier = pipeline("text-classification", model = "pysentimiento/robertuito-sentiment-analysis")
-
-def classify_image(inp): 
-  inp = asarray(inp.resize((224, 224)))
-  inp = inp.reshape((-1,) + inp.shape) 
-  inp = tf.keras.applications.mobilenet_v2.preprocess_input(inp) 
-  prediction = inception_net.predict(inp).flatten() 
-  confidences = {labels[k]: float(prediction[k]) for k in range(1000)} 
-  return confidences
+image_classifier = pipeline("image-classification", model="microsoft/swin-tiny-patch4-window7-224")
 
 def audio_to_text(audio):
   text = transcribe(audio)["text"]
@@ -25,7 +12,13 @@ def audio_to_text(audio):
 
 def text_to_sentiment(text):
   return classifier(text)[0]["label"]
-
+    
+def classify_image(image): 
+  image = Image.fromarray(image.astype('uint8'), 'RGB')
+  answers = image_classifier(image)
+  labels = {answer["label"]: answer["score"] for answer in answers}
+  return labels
+    
 demo = gr.Blocks()
 
 with demo: