app.py

import gradio as gr
from fastai.learner import load_learner
import pandas as pd
import numpy as np

learn = load_learner('movie_recommendation.pkl')

ratings = pd.read_csv('ratings_test.csv')
ids = ratings['userId'].unique()
ids_list = list(map(str, ids.tolist()))

def top_predictions(userId, recommendations = 5):
    userId = int(userId)
    items = pd.Series(learn.dls.classes['title']).unique()    
    clas_items = ratings.loc[(ratings['userId'] == userId) & (ratings['rating'] > 0), 'title']
    no_clas_items = np.setdiff1d(items, clas_items)
    df = pd.DataFrame({'userId': [userId]*len(no_clas_items), 'title': no_clas_items})
    preds,_ = learn.get_preds(dl=learn.dls.test_dl(df))
    df['prediction'] = preds.numpy()    
    top_5 = df.nlargest(recommendations, 'prediction')
    return '\n'.join(top_5['title'].tolist())

iface = gr.Interface(
    description="This model is a film recommender based on our user's ratings of other films .",
    fn=top_predictions,
    inputs=gr.Dropdown(choices=ids_list),
    outputs="text"
    )

iface.launch(share=True)