MovieMitra Recommender
Suggests films similar to one you already liked, from a 96 film catalogue.
How it works
Each film becomes a bag of words from its genres, tags, director and language. TF-IDF turns those into numbers and cosine similarity compares them.
How to load it
from huggingface_hub import hf_hub_download
import joblib
path = hf_hub_download("YOUR_USERNAME/moviemitra-recommender", "recommender.joblib")
model = joblib.load(path)
model.similar_to("Kaala Paani")
recommender.py must be in your folder for the load to work.
Trained on
96 films and 1,348 ratings from 80 viewers. Made up data, built for teaching.
What it can't do
It only reads labels, never the film itself. Two films with the same genre, director and tags score 1.000 and it can't tell them apart. It also can't suggest anything outside these 96 films.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support