victorkohler

import sys
import pandas as pd
import numpy as np
import scipy.sparse as sparse
from scipy.sparse.linalg import spsolve
import random

from sklearn.preprocessing import MinMaxScaler

import implicit # The Cython library

victorkohler

import sys
import pandas as pd
import numpy as np
import scipy.sparse as sparse
from scipy.sparse.linalg import spsolve
import random

from sklearn.preprocessing import MinMaxScaler

import implicit

victorkohler

import tensorflow as tf
import pandas as pd
import numpy as np
import scipy.sparse as sp

from tqdm import tqdm

#---------------------------
# LOAD AND PREPARE THE DATA
#---------------------------

victorkohler

#-------------
# HYPERPARAMS
#-------------

epochs = 50
batches = 30
num_factors = 64 # Number of latent features

# Independent lambda regularization values 

victorkohler

#-------------------------
# TENSORFLOW GRAPH
#-------------------------

# Set up our Tensorflow graph
graph = tf.Graph()

def init_variable(size, dim, name=None):
    '''

victorkohler

with graph.as_default():
    '''
    Loss function: 
    -SUM ln σ(xui - xuj) + λ(w1)**2 + λ(w2)**2 + λ(w3)**2 ...

    ln = the natural log
    σ(xuij) = the sigmoid function of xuij.
    λ = lambda regularization value.
    ||W||**2 = the squared L2 norm of our model parameters.

victorkohler

#------------------
# GRAPH EXECUTION
#------------------

# Run the session. 
session = tf.Session(config=None, graph=graph)
session.run(init)

# This has noting to do with tensorflow but gives

victorkohler

#-----------------------
# FIND SIMILAR ARTISTS
#-----------------------

def find_similar_artists(artist=None, num_items=10):
    """Find artists similar to an artist.

    Args:
        artist (str): The name of the artist we want to find similar artists for

victorkohler

#---------------------
# MAKE RECOMMENDATION
#---------------------

def make_recommendation(user_id=None, num_items=10):
    """Recommend items for a given user given a trained model

    Args:
        user_id (int): The id of the user we want to create recommendations for.

victorkohler

import pandas as pd
import numpy as np
import pickle

def load_dataset():
    """
    Loads the lastfm dataset from a pickle file into a pandas dataframe
    and transforms it into the format we need. 
    
    We then split it into a training and a test set.