forked from aianaconda/pytorch-GNN-2nd-
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcode_29_train.py
161 lines (116 loc) · 4.86 KB
/
code_29_train.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
# -*- coding: utf-8 -*-
"""
Created on Thu Jul 9 15:09:03 2020
@author: 代码医生工作室
@公众号：xiangyuejiqiren （内有更多优秀文章及学习资料）
@来源: <PyTorch深度学习和图神经网络(卷2）——开发应用>配套代码
@配套代码技术支持：bbs.aianaconda.com
"""
import pickle
import argparse
import numpy as np
import torch
import torch.nn as nn
from torch.utils.data import DataLoader
import torchtext
import dgl
import tqdm
import code_31_sampler as sampler_module
from code_30_model import *
#指定设备
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
#读取数据集
dataset_path = './data.pkl'
with open(dataset_path, 'rb') as f:
dataset = pickle.load(f)
g = dataset['train-graph']
item_texts = dataset['item-texts']
# 设置节点属性值
g.nodes['user'].data['id'] = torch.arange(g.number_of_nodes('user'))
g.nodes['movie'].data['id'] = torch.arange(g.number_of_nodes('movie'))
#加载词向量，用于解析电影标题
fields = {}
examples = []
titlefield = torchtext.data.Field(include_lengths=True, lower=True, batch_first=True)
fields = [('title', titlefield)]
for i in range(g.number_of_nodes('movie')):
example = torchtext.data.Example.fromlist( [item_texts['title'][i] ], fields)
examples.append(example)
textset = torchtext.data.Dataset(examples, fields)
titlefield.build_vocab(getattr(textset, 'title'), vectors = "glove.6B.100d")#将样本数据转为词向量
# titlefield.build_vocab(getattr(textset, 'title'))#将样本数据转为词向量
num_layers =2
hidden_dims = 32
#构建带采样器的输入数据集
neighbor_sampler = sampler_module.NeighborSampler( g, 'user', 'movie',num_layers)
collator = sampler_module.PinSAGECollator(neighbor_sampler, g, 'movie', textset)
batch_size = 32
#正负样本对采样器
batch_sampler = sampler_module.ItemToItemBatchSampler(
g, 'user', 'movie', batch_size)
#训练集数据加载器
dataloader = DataLoader( batch_sampler,collate_fn=collator.collate_train)
#测试集数据加载器
dataloader_test = DataLoader( torch.arange(g.number_of_nodes('movie')),
batch_size=batch_size, collate_fn=collator.collate_test)
dataloader_it = iter(dataloader)
#################################################################
#######################################################################
# Model
model = PinSAGEModel(g, 'movie', textset, hidden_dims, num_layers).to(device)
# Optimizer
opt = torch.optim.Adam(model.parameters(), lr=0.001, weight_decay=5e-4)
# For each batch of head-tail-negative triplets...
num_epochs = 2
batches_per_epoch = 5000
for epoch_id in range(num_epochs):
model.train()
for batch_id in tqdm.trange(batches_per_epoch):
pos_graph, neg_graph, blocks = next(dataloader_it)
# Copy to GPU
for i in range(len(blocks)):
blocks[i] = blocks[i].to(device)
pos_graph = pos_graph.to(device)
neg_graph = neg_graph.to(device)
loss = model(pos_graph, neg_graph, blocks).mean()
opt.zero_grad()
loss.backward()
opt.step()
print(epoch_id,loss)
model.eval()
with torch.no_grad():
h_item_batches = []
for blocks in dataloader_test:
for i in range(len(blocks)):
blocks[i] = blocks[i].to(device)
h_item_batches.append(model.get_repr(blocks))
h_item = torch.cat(h_item_batches, 0)
# len(h_item) h_item.shape #[3706, 32]
# len(item_texts['title'])
# test_matrix= dataset['test-matrix'].tocsr()
# val_matrix = dataset['val-matrix'].tocsr()
# n_users = g.number_of_nodes('user')#6040
graph_slice = g.edge_type_subgraph(['watched'])
latest_interactions = dgl.sampling.select_topk(graph_slice, 1, 'timestamp', edge_dir='out')
user, latest_items = latest_interactions.all_edges(form='uv', order='srcdst')
# each user should have at least one "latest" interaction
# assert torch.equal(user, torch.arange(n_users))
user_batch = user[:batch_size]
latest_item_batch = latest_items[user_batch].to(device=h_item.device)
dist = h_item[latest_item_batch] @ h_item.t()
# dist.shape #[32, 3706]
#删除已经看过的电影
for i, u in enumerate(user_batch.tolist()):
interacted_items = g.successors(u, etype='watched')
dist[i, interacted_items] = -np.inf
scores, re_index = dist.cpu().topk(5, 1)#推荐个数5
# scores = scores.numpy()
for i in range(batch_size):
uid = user_batch[i].numpy()
movieid = latest_item_batch[i].cpu().numpy()
moviestr = item_texts['title']
print("用户:",uid," 最后一次观看的电影:",movieid,moviestr[ movieid ])
print("推荐的电影标题:",moviestr[re_index[i].numpy()])
print("推荐的电影分数:",scores[i].numpy())
print("推荐的电影 id:",re_index[i].numpy())