Update app.py

app.py

@@ -2,33 +2,62 @@ import streamlit as st
 import tensorflow as tf
 from PIL import Image
 import numpy as np
+from tensorflow.keras.applications.vgg16 import VGG16,preprocess_input
+from tensorflow.keras.preprocessing.image import load_img,img_to_array
+from tensorflow.keras.preprocessing.text import Tokenizer
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+from tensorflow.keras.models import Model
+from tensorflow.keras.utils import to_categorical,plot_model
+from tensorflow.keras.layers import Input,Dense,LSTM,Embedding, Dropout, add
+from keras.models import load_model
 
+model = load_model('image_caption.h5')
+tokenizer = Tokenizer()
+max_length=35
 # Load pre-trained model
-model = tf.keras.applications.ResNet50(weights='imagenet')
+vgg_model = VGG16()
+vgg_model = Model(inputs=vgg_model.inputs, outputs=vgg_model.layers[-2].output)
 
 # Set Streamlit configurations
 st.set_page_config(page_title="Image Classifier App", layout="wide")
 
-# Define labels for ImageNet classes
-LABELS_PATH = 'https://raw.githubusercontent.com/anishathalye/imagenet-simple-labels/master/imagenet-simple-labels.json'
-labels = tf.keras.utils.get_file('ImageNetLabels.json', LABELS_PATH)
-with open(labels) as f:
-    classes = eval(f.read())
 
 # Function to preprocess the input image
 def preprocess_image(image):
-    image = image.resize((224, 224))  # Resize image to match model input size
-    image = tf.keras.preprocessing.image.img_to_array(image)
-    image = np.expand_dims(image, axis=0)
-    image = tf.keras.applications.resnet50.preprocess_input(image)
+    image = load_img(image, target_size=(224, 224))
+    image = img_to_array(image)
+    image = image.reshape((1, image.shape[0], image.shape[1], image.shape[2]))
+    image = preprocess_input(image)
     return image
 
 # Function to make predictions on the input image
 def predict(image):
     image = preprocess_image(image)
-    preds = model.predict(image)
-    top_predictions = tf.keras.applications.resnet50.decode_predictions(preds, top=5)[0]
-    return top_predictions
+    feature = vgg_model.predict(image, verbose=0)
+    preds = predict_caption(model, feature, tokenizer, max_length)
+    preds=preds[8:-7]
+    return preds
+
+def idx_word(integer,tok):
+    for word,index in tok.word_index.items():
+        if index== integer:
+            return word
+    return None
+
+def predict_caption(model,image,tok,max_len):
+    in_text="startseq"
+    for i in range(max_len):
+        seq=tok.texts_to_sequences([in_text])[0]
+        seq=pad_sequences([seq],max_len)
+        yhat = model.predict([image, seq], verbose=0)
+        yhat = np.argmax(yhat)
+        word = idx_word(yhat, tok)
+        if word is None:
+            break
+        in_text += " " + word
+        if word == 'endseq':
+            break
+    return in_text
 
 # Streamlit app
 def main():