app.py

import streamlit as st
import tensorflow as tf
from tensorflow import keras
from PIL import Image
import numpy as np


#model
model = tf.keras.models.load_model(r'C:\Users\Souvik Chand\Documents\python_my\apps\stream lits\cats and dogs\model3.h5')
labels = ['Cat', 'Dog']

def preprocess_image(image):
    """resizes the image"""
    image = image.resize((256, 256))      # Adjust based on your model's input size
    image = np.array(image) / 255.0       # Normalize the image
    image = np.expand_dims(image, axis=0)
    return image

def process(image):
    """not in use"""
    image= tf.cast(image/255, tf.float32)
    return image

st.title('Dogs and cats')
st.write('Upload an image and the model will predict its class.')

st.sidebar.title('upload a photo')
uploaded_file = st.sidebar.file_uploader('choose image',accept_multiple_files=False, type=['jpg'])

if uploaded_file is not None:
  image = Image.open(uploaded_file)
  
  st.image(image, caption='Uploaded Image', use_column_width=True,width=100)
  st.write(f'Original Image Shape: {image.size}')
  
  preprocessed_image = preprocess_image(image)
  test_input = preprocessed_image.reshape((1,256,256,3))

  predictions = model.predict(test_input)[0][0]
  confidence= round((abs(predictions-0.5)/0.5)*100)
  st.write(predictions)

  if predictions<0.2:
     st.write(f'Predicted class: Cat')
  elif predictions>0.8:
     st.write('Prediction class: Dog')
  elif (predictions <0.7) or (predictions >0.6):
     st.write('i feel you are trying to trick me!')
  else:
     st.write("looks like neither")

  st.write(f'confidence: {confidence}%')