khuangaf

from sklearn.preprocessing import LabelEncoder

df = pd.read_csv('../input/yoochoose-click.dat', header=None)
df.columns=['session_id','timestamp','item_id','category']

buy_df = pd.read_csv('../input/yoochoose-buys.dat', header=None)
buy_df.columns=['session_id','timestamp','item_id','price','quantity']

item_encoder = LabelEncoder()
df['item_id'] = item_encoder.fit_transform(df.item_id)

khuangaf

import torch
from torch_geometric.data import Data


x = torch.tensor([[2,1], [5,6], [3,7], [12,0]], dtype=torch.float)
y = torch.tensor([0, 1, 0, 1], dtype=torch.float)

edge_index = torch.tensor([[0, 2, 1, 0, 3],
                           [3, 1, 0, 1, 2]], dtype=torch.long)

khuangaf

edge_index = torch.tensor([[0, 2, 1, 0, 3],
                           [3, 1, 0, 1, 2]], dtype=torch.long)

khuangaf

edge_index = torch.tensor([[0, 2, 1, 0, 3],
                           [3, 1, 0, 1, 2]], dtype=torch.long)

khuangaf

edge_index = torch.tensor([[0, 1, 2, 0, 3],
                           [1, 0, 1, 3, 2]], dtype=torch.long)

khuangaf

x = torch.tensor([[2,1], [5,6], [3,7], [12,0]], dtype=torch.float)
y = torch.tensor([0, 1, 0, 1], dtype=torch.float)

khuangaf

'''
Author: Kung-hsiang, Huang (Steeve)
Date: 2019/Mar/15
'''

class CategoricalEncoder():
    '''
    This class is for those operating on large data, in which sklearn's LabelEncoder class may take too much time.
    This encoder is only suitable for 1-d array/ list. You may modify it to become n-d compatible.
    '''

khuangaf

def forward(self, user_indices, item_indices, titles):
    user_embedding_mlp = self.embedding_user_mlp(user_indices)
    item_embedding_mlp = self.embedding_item_mlp(item_indices)
    user_embedding_mf = self.embedding_user_mf(user_indices)
    item_embedding_mf = self.embedding_item_mf(item_indices)

    #### mf part
    mf_vector =torch.mul(user_embedding_mf, item_embedding_mf)
    mf_vector = torch.nn.Dropout(self.config.dropout_rate_mf)(mf_vector)

khuangaf

def __init__(self, config):
    super(NeuMF, self).__init__()

    #mf part
    self.embedding_user_mf = torch.nn.Embedding(num_embeddings=self.num_users, embedding_dim=self.latent_dim_mf)
    self.embedding_item_mf = torch.nn.Embedding(num_embeddings=self.num_items, embedding_dim=self.latent_dim_mf)

    #mlp part
    self.embedding_user_mlp = torch.nn.Embedding(num_embeddings=self.num_users, embedding_dim=self.latent_dim_mlp)
    self.embedding_item_mlp = torch.nn.Embedding(num_embeddings=self.num_items, embedding_dim=self.latent_dim_mlp)

khuangaf

class NeuMF(torch.nn.Module):
    def __init__(self, config):
        super(NeuMF, self).__init__()

        #mf part
        self.embedding_user_mf = torch.nn.Embedding(num_embeddings=self.num_users, embedding_dim=self.latent_dim_mf)
        self.embedding_item_mf = torch.nn.Embedding(num_embeddings=self.num_items, embedding_dim=self.latent_dim_mf)
        
        #mlp part
        self.embedding_user_mlp = torch.nn.Embedding(num_embeddings=self.num_users, embedding_dim=self.latent_dim_mlp)