def rating_pred(user_id,item_list,algo):
pred = {}
for i in item_list:
prediction = algo.predict(user_id, i)
pred[i] = prediction.est
return pred
from data_loader import data, user_ids, movie_ids
from recommender import algo
# Load the training data
trainingSet = data.build_full_trainset()
print("Please have patience computing takes a minute.")
# Train the NN
algo.fit(trainingSet)
while True:
print("\n")
user = int(input("Please enter user: "******"Please enter movie: "))
if user <= len(user_ids) and movie <= len(movie_ids):
# Predict and display the results
prediction = algo.predict(user,movie)
result = prediction.est
print("Prediction for this movie: " + str(result))
else:
print("User or movie doesn't exist!")
test_user_tag_dict = {}
for user in user_list:
spe_user_df = test_df.loc[test_df['user_id'] == user]
spe_user_tag_list = list(
set(spe_user_df['hashtag'].explode('hashtag').tolist()))
test_user_tag_dict[user] = spe_user_tag_list
trainingSet = data.build_full_trainset()
algo.fit(trainingSet)
testF2 = open(
'./' + dataPath + classifierPath + '/test' + classifierPath + '2.dat', "a")
preF = open(
'./' + dataPath + classifierPath + '/pre' + classifierPath + '.txt', 'a')
for user_num, user in enumerate(user_list):
print(user_num)
testF2.write(f"# query {user_num + 1}\n")
positive_tag_list = list(set(test_user_tag_dict[user]))
for tag in positive_tag_list:
testF2.write(f"1 qid:{user_num+1}\n")
precision = algo.predict(user, tag)
preF.write(f"{precision.est}\n")
negative_tag_list = list(set(test_tag_list) - set(positive_tag_list))
for tag in negative_tag_list:
testF2.write(f"0 qid:{user_num + 1}\n")
precision = algo.predict(user, tag)
preF.write(f"{precision.est}\n")
testF2.close()
preF.close()
# -*- coding: utf-8 -*-
"""
Created on Sun Jul 28 14:15:51 2019
@author: PAVAN
"""
from recommender import data, algo
trainingSet = data.build_full_trainset()
algo.fit(trainingSet)
prediction = algo.predict('E', 2)
prediction.est
from load_data import data
from recommender import algo
import pandas as pd
import csv
import math
import functions
trainingSet = data.build_full_trainset()
algo.fit(trainingSet)
test_data = pd.read_csv('test.csv')
errors = []
with open('test2.csv', mode='w', newline='') as thefile:
writer = csv.writer(thefile,
delimiter=',',
quotechar='"',
quoting=csv.QUOTE_MINIMAL)
writer.writerow(
['item', 'user', 'actual rating', 'predicted rating', 'error'])
for i in test_data.values:
predicted_value = algo.predict(i[1], i[0]).est
error = math.fabs(i[2] - predicted_value)
errors.append(error)
rms_error = functions.rms(errors, len(errors))
print('Mean error:', rms_error)