inference.py

import pickle
import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np
from typing import List, Dict

# Function to load model components
def load_model_components(model_path):
    with open(model_path, 'rb') as f:
        model_components = pickle.load(f)
    return model_components

# Function to recommend jobs based on input skills and major
def recommend_jobs_for_input_skills(input_hard_skills: str, input_soft_skills: str, input_major: str,
                                    jobs_data: pd.DataFrame, model_path: str) -> List[str]:
    # Load saved model components
    tfidf_vectorizer_skills, tfidf_vectorizer_majors, companies_skills_vec, companies_majors_vec = load_model_components(model_path)

    # Vectorize input skills and major
    input_hard_skills_vec = tfidf_vectorizer_skills.transform([input_hard_skills])
    input_soft_skills_vec = tfidf_vectorizer_skills.transform([input_soft_skills])
    input_major_vec = tfidf_vectorizer_majors.transform([input_major])

    # Average the vectorized hard and soft skills
    input_skills_vec = (input_hard_skills_vec + input_soft_skills_vec) / 2

    # Compute similarities
    skills_similarity = cosine_similarity(input_skills_vec, companies_skills_vec)
    major_similarity = cosine_similarity(input_major_vec, companies_majors_vec)

    # Ensure the number of companies in both similarities is aligned
    if skills_similarity.shape[1] != major_similarity.shape[1]:
        min_dim = min(skills_similarity.shape[1], major_similarity.shape[1])
        skills_similarity = skills_similarity[:, :min_dim]
        major_similarity = major_similarity[:, :min_dim]

    # Combine similarities
    combined_similarity = (skills_similarity + major_similarity) / 2

    # Get top 3 job recommendations
    sorted_company_indices = np.argsort(-combined_similarity[0])
    recommended_jobs = jobs_data.iloc[sorted_company_indices]['Major'].values[:3]

    return recommended_jobs.tolist()

# Example usage for testing
if __name__ == "__main__":
    input_hard_skills = "Python, Java, Finance, Excel"
    input_soft_skills = "Communication, Teamwork"
    input_major = "Computer Science"

    jobs_data = pd.read_csv("jobs_data.csv")
    model_path = "recommendation_model.pkl"

    recommended_jobs = recommend_jobs_for_input_skills(input_hard_skills, input_soft_skills, input_major, jobs_data, model_path)
    print("Recommended Jobs based on input skills and major:")
    print(recommended_jobs)