get_cv.py

import pandas as pd
import openai
from data import data as df
import numpy as np
import os

openai.api_key = os.environ.get("openai")

def cosine_similarity(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))


def get_embedding(text, model="text-embedding-ada-002"):
    try:
        text = text.replace("\n", " ")
    except:
        None
    return openai.embeddings.create(input = [text], model=model).data[0].embedding
    
def get_embedding2(text, model="text-embedding-ada-002"):
    try:
        text = text.replace("\n", " ")
    except:
        None
    try:
        return openai.Embedding.create(input = [text], model=model)['data'][0]['embedding']
    except:
        time.sleep(2)

def search_cv(search, nb=3, pprint=True):
    embedding = get_embedding(search, model='text-embedding-ada-002')
    df_replicate = df.copy()
    
    def wrap_cos(x,y):
        try:
            res = cosine_similarity(x,y)
        except:
            res = 0
        return res
        
    df_replicate['similarities'] = df_replicate.embedding.apply(lambda x: wrap_cos(x, embedding))
    res = df_replicate.sort_values('similarities', ascending=False).head(int(nb))
    return res

def get_cv(text, nb):
    result = search_cv(text,nb).to_dict(orient="records")
    final_str = ""
    for r in result:
        final_str += "#### Candidat avec " + str(round(r["similarities"]*100,2)) + "% de similarité :\n"+ str(r["summary"]).replace("#","")
        final_str += "\n\n[-> Lien vers le CV complet]("+ str(r["url"]) + ')'
        final_str += "\n\n-----------------------------------------------------------------------------------------------------\n\n"
    final_str = final_str.replace("`", "")
    return  final_str