app[].py

import numpy as np
import tensorflow

from tensorflow.keras.preprocessing import image
from tensorflow.keras.layers import GlobalMaxPooling2D
from tensorflow.keras.applications.resnet50 import ResNet50,preprocess_input
from numpy.linalg import norm
import os
from tqdm import tqdm
import pickle

model = ResNet50(weights="imagenet", include_top=False,input_shape=(224,224,3))
model.trainable=False

model1 = tensorflow.keras.Sequential([
    model,
    GlobalMaxPooling2D()
])

def  extract_features(img_path,model):
    img=image.load_img(img_path,target_size = (224,224))
    image_array = image.img_to_array(img)
    expanded_image_array = np.expand_dims(image_array,axis=0)
    processed_image = preprocess_input(expanded_image_array)
    result = model.predict(processed_image).flatten()
    normalized_result=result/norm(result)
    return  normalized_result

filenames =[]

for file in os.listdir('set0'):
    filenames.append(os.path.join('set0',file))

for file in os.listdir('set1'):
    filenames.append(os.path.join('set1',file))

for file in os.listdir('set2'):
    filenames.append(os.path.join('set2',file))

for file in os.listdir('set3'):
    filenames.append(os.path.join('set3',file))

for file in os.listdir('set4'):
    filenames.append(os.path.join('set4',file))

feature_list = []

for i in tqdm(filenames):
    feature_list.append(extract_features(i,model1))

print(np.array(feature_list).shape)

import pickle
pickle.dump(feature_list,open('embeddings2.pkl','wb'))
pickle.dump(filenames,open('filenames2.pkl','wb'))


import streamlit as st
import os
from PIL import Image
import pickle
import tensorflow
import numpy as np
from numpy.linalg import norm
from tensorflow.keras.preprocessing import image
from tensorflow.keras.layers import GlobalMaxPooling2D
from tensorflow.keras.applications.resnet50 import ResNet50, preprocess_input
from sklearn.neighbors import NearestNeighbors

feature_list = np.array(pickle.load(open('embeddings2.pkl', 'rb')))
filenames = pickle.load(open('filenames2.pkl', 'rb'))

model = ResNet50(weights='imagenet', include_top=False, input_shape=(224, 224, 3))
model.trainable = False

model = tensorflow.keras.Sequential([
    model,
    GlobalMaxPooling2D()
])

st.title("Fashion Recommender System")


def extract_features(img_path, model):
    img = image.load_img(img_path, target_size=(224, 224))
    image_array = image.img_to_array(img)
    expanded_image_array = np.expand_dims(image_array, axis=0)
    processed_image = preprocess_input(expanded_image_array)
    result = model.predict(processed_image).flatten()
    normalized_result = result / norm(result)
    return normalized_result

def recommend(features,feature_list):
    neighbors = NearestNeighbors(n_neighbors=5, algorithm='brute', metric='euclidean')
    neighbors.fit(feature_list)

    distances, indices = neighbors.kneighbors([features])
    return indices


def save_uploaded_file(uploaded_file):
    try:
        with open(os.path.join('uploads', uploaded_file.name), 'wb') as f:
            f.write(uploaded_file.getbuffer())
        return 1
    except:
        return 0


uploaded_file = st.file_uploader("choose an image")

if uploaded_file is not None:
    if save_uploaded_file(uploaded_file):
        display_image = Image.open(uploaded_file)
        st.image(display_image)
        features = extract_features(os.path.join("uploads",uploaded_file.name),model)
        #st.text(features)
        indices = recommend(features,feature_list)
        col1,col2,col3,col4,col5 = st.columns(5)
        with col1:
            st.image(filenames[indices[0][0]])
        with col2:
            st.image(filenames[indices[0][1]])
        with col3:
            st.image(filenames[indices[0][2]])
        with col4:
            st.image(filenames[indices[0][3]])
        with col5:
            st.image(filenames[indices[0][4]])
    else:
        st.header("Some error has occured while uploading file")