ir.py

from gensim.models import KeyedVectors
from gensim.utils import simple_preprocess
import gensim.downloader as api
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np


class SearchEngine:
	def __init__(self, model_api_path='glove-wiki-gigaword-100'):
		self.model = api.load(model_api_path)
		self.prompts = []
		self.vectors = []

	def vectorize(self, prompt):
		words = simple_preprocess(prompt)
		vectors = [self.model[word] for word in words if word in self.model]
		return np.mean(vectors, axis=0)

	def add(self, prompt):
		vector = self.vectorize(prompt)
		self.prompts.append(prompt)
		self.vectors.append(vector)

	def search(self, input_prompt):
		vectorized_input = self.vectorize(input_prompt)
		similarity_scores = cosine_similarity([vectorized_input], self.vectors)[0]
		most_similar_idx = np.argmax(similarity_scores)

		return self.prompts[most_similar_idx]


if __name__ == "__main__":
	storage = SearchEngine()

	storage.add("I love programming")
	storage.add("You need to graduate")
	storage.add("Library is open")

	input_prompt = "I enjoy coding"
	most_similar_prompt = storage.search(input_prompt)

	print(f"Input Prompt: {input_prompt}")
	print(f"Most Similar Prompt: {most_similar_prompt}")