Exploring Personalized Shopping Experiences with Qdrant's Discovery API

app.py

import streamlit as st
from sentence_transformers import SentenceTransformer
from qdrant_client import models, QdrantClient


#instantiate qdrant client
print("[INFO] Client created...")
qdrant = QdrantClient("localhost", port=6333)

#importing the GTE model
print("[INFO] Loading encoder model...")
encoder = SentenceTransformer('thenlper/gte-small')

print("[INFO] Running Streamlit Application!! ")

Gadget = ["Apple Macbook", "USB Drive", "Hard Drive", 
          "Intel Processor", "Memory Card", "Graphics Card"] 


def main():
    st.title("Discovery API for Personalized Shopping Experience")
    st.subheader("By Niranjan Akella")

    # multi-select checkbox for gadgets
    choice_of_gadget = st.multiselect(
        "Pick gadgets you like",
        Gadget,
        default = ["Intel Processor", "Graphics Card"],
        key="category1"
    )
    disliked_gadgets = list(set(Gadget) - set(choice_of_gadget))
                     
    # "Personalize" button to trigger a function
    if st.button("Personalize"):
        personalize_function(choice_of_gadget, disliked_gadgets)

#personalize fn to perform discovery search
def personalize_function(choice_of_gadget, disliked_gadgets):

    #encode choices and group context pairs
    contexts = [models.ContextExamplePair(positive=encoder.encode(l).tolist(), negative=encoder.encode(d).tolist()) for (l,d) in list(zip(choice_of_gadget, disliked_gadgets))]

    discovered_products = qdrant.discover(collection_name='e-shopping', context=contexts, limit=3)
    st.write("Top Recommended Products:")
    for product in discovered_products:
        st.write(f"Title: {product.payload['title_left']}")
        st.write(product.payload['description_left'])
        st.write("\n")

    

if __name__ == "__main__":
    main()

database.py

from datasets import load_dataset
from sentence_transformers import SentenceTransformer
from qdrant_client import models, QdrantClient
from tqdm import tqdm

'''
If you face any SSL issues while downloading the dataset or model
uncomment the follwoing to by-pass curl certificate verification
'''
# import os
# os.environ['CURL_CA_BUNDLE'] = ''

#instantiate qdrant client
print("[INFO] Client created...")
qdrant = QdrantClient("localhost", port=6333)

#download dataset from hugging face
print("[INFO] Loading dataset...")
dataset = load_dataset("wdc/products-2017", split='train')

#process dataset
print("[INFO] Processing dataset...")
data = []
fields = ['title_left', 'description_left']
for i in tqdm(range(len(dataset)), total=len(dataset)):
  if dataset[i]['description_left']:
    data.append({field:dataset[i][field] for field in fields})

#importing the GTE model
print("[INFO] Loading encoder model...")
encoder = SentenceTransformer('thenlper/gte-small')

#creating data collection in qdrant
print("[INFO] Creating a data collection...")
qdrant.recreate_collection(
    collection_name="e-shopping",
    vectors_config=models.VectorParams(
        size=encoder.get_sentence_embedding_dimension(),  # Vector size is defined by used model
        distance=models.Distance.COSINE,
    ),
)

#uploading vectors to data collection
print("[INFO] Uploading data to data collection...")
records = []
for idx, sample in tqdm(enumerate(data), total=len(data)):
    if sample['description_left']:
        records.append(models.Record(
                id=idx, vector=encoder.encode(sample["description_left"]).tolist(), payload=sample
            ) ) 
        
qdrant.upload_records(
    collection_name="e-shopping",
    records=records,
)
print("[INFO] Successfully uploaded data to datacollection!")

requirements.txt

datasets==2.11.0
qdrant-client==1.7.0
sentence_transformers==2.2.2
tqdm==4.65.0
sentencepiece==0.1.99
transformers==4.36.2
streamlit==1.29.0