-
Notifications
You must be signed in to change notification settings - Fork 225
/
main.py
197 lines (179 loc) · 8.41 KB
/
main.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
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
import torch
import tqdm
from sklearn.metrics import roc_auc_score
from torch.utils.data import DataLoader
from torchfm.dataset.avazu import AvazuDataset
from torchfm.dataset.criteo import CriteoDataset
from torchfm.dataset.movielens import MovieLens1MDataset, MovieLens20MDataset
from torchfm.model.afi import AutomaticFeatureInteractionModel
from torchfm.model.afm import AttentionalFactorizationMachineModel
from torchfm.model.dcn import DeepCrossNetworkModel
from torchfm.model.dfm import DeepFactorizationMachineModel
from torchfm.model.ffm import FieldAwareFactorizationMachineModel
from torchfm.model.fm import FactorizationMachineModel
from torchfm.model.fnfm import FieldAwareNeuralFactorizationMachineModel
from torchfm.model.fnn import FactorizationSupportedNeuralNetworkModel
from torchfm.model.hofm import HighOrderFactorizationMachineModel
from torchfm.model.lr import LogisticRegressionModel
from torchfm.model.ncf import NeuralCollaborativeFiltering
from torchfm.model.nfm import NeuralFactorizationMachineModel
from torchfm.model.pnn import ProductNeuralNetworkModel
from torchfm.model.wd import WideAndDeepModel
from torchfm.model.xdfm import ExtremeDeepFactorizationMachineModel
from torchfm.model.afn import AdaptiveFactorizationNetwork
def get_dataset(name, path):
if name == 'movielens1M':
return MovieLens1MDataset(path)
elif name == 'movielens20M':
return MovieLens20MDataset(path)
elif name == 'criteo':
return CriteoDataset(path)
elif name == 'avazu':
return AvazuDataset(path)
else:
raise ValueError('unknown dataset name: ' + name)
def get_model(name, dataset):
"""
Hyperparameters are empirically determined, not opitmized.
"""
field_dims = dataset.field_dims
if name == 'lr':
return LogisticRegressionModel(field_dims)
elif name == 'fm':
return FactorizationMachineModel(field_dims, embed_dim=16)
elif name == 'hofm':
return HighOrderFactorizationMachineModel(field_dims, order=3, embed_dim=16)
elif name == 'ffm':
return FieldAwareFactorizationMachineModel(field_dims, embed_dim=4)
elif name == 'fnn':
return FactorizationSupportedNeuralNetworkModel(field_dims, embed_dim=16, mlp_dims=(16, 16), dropout=0.2)
elif name == 'wd':
return WideAndDeepModel(field_dims, embed_dim=16, mlp_dims=(16, 16), dropout=0.2)
elif name == 'ipnn':
return ProductNeuralNetworkModel(field_dims, embed_dim=16, mlp_dims=(16,), method='inner', dropout=0.2)
elif name == 'opnn':
return ProductNeuralNetworkModel(field_dims, embed_dim=16, mlp_dims=(16,), method='outer', dropout=0.2)
elif name == 'dcn':
return DeepCrossNetworkModel(field_dims, embed_dim=16, num_layers=3, mlp_dims=(16, 16), dropout=0.2)
elif name == 'nfm':
return NeuralFactorizationMachineModel(field_dims, embed_dim=64, mlp_dims=(64,), dropouts=(0.2, 0.2))
elif name == 'ncf':
# only supports MovieLens dataset because for other datasets user/item colums are indistinguishable
assert isinstance(dataset, MovieLens20MDataset) or isinstance(dataset, MovieLens1MDataset)
return NeuralCollaborativeFiltering(field_dims, embed_dim=16, mlp_dims=(16, 16), dropout=0.2,
user_field_idx=dataset.user_field_idx,
item_field_idx=dataset.item_field_idx)
elif name == 'fnfm':
return FieldAwareNeuralFactorizationMachineModel(field_dims, embed_dim=4, mlp_dims=(64,), dropouts=(0.2, 0.2))
elif name == 'dfm':
return DeepFactorizationMachineModel(field_dims, embed_dim=16, mlp_dims=(16, 16), dropout=0.2)
elif name == 'xdfm':
return ExtremeDeepFactorizationMachineModel(
field_dims, embed_dim=16, cross_layer_sizes=(16, 16), split_half=False, mlp_dims=(16, 16), dropout=0.2)
elif name == 'afm':
return AttentionalFactorizationMachineModel(field_dims, embed_dim=16, attn_size=16, dropouts=(0.2, 0.2))
elif name == 'afi':
return AutomaticFeatureInteractionModel(
field_dims, embed_dim=16, atten_embed_dim=64, num_heads=2, num_layers=3, mlp_dims=(400, 400), dropouts=(0, 0, 0))
elif name == 'afn':
print("Model:AFN")
return AdaptiveFactorizationNetwork(
field_dims, embed_dim=16, LNN_dim=1500, mlp_dims=(400, 400, 400), dropouts=(0, 0, 0))
else:
raise ValueError('unknown model name: ' + name)
class EarlyStopper(object):
def __init__(self, num_trials, save_path):
self.num_trials = num_trials
self.trial_counter = 0
self.best_accuracy = 0
self.save_path = save_path
def is_continuable(self, model, accuracy):
if accuracy > self.best_accuracy:
self.best_accuracy = accuracy
self.trial_counter = 0
torch.save(model, self.save_path)
return True
elif self.trial_counter + 1 < self.num_trials:
self.trial_counter += 1
return True
else:
return False
def train(model, optimizer, data_loader, criterion, device, log_interval=100):
model.train()
total_loss = 0
tk0 = tqdm.tqdm(data_loader, smoothing=0, mininterval=1.0)
for i, (fields, target) in enumerate(tk0):
fields, target = fields.to(device), target.to(device)
y = model(fields)
loss = criterion(y, target.float())
model.zero_grad()
loss.backward()
optimizer.step()
total_loss += loss.item()
if (i + 1) % log_interval == 0:
tk0.set_postfix(loss=total_loss / log_interval)
total_loss = 0
def test(model, data_loader, device):
model.eval()
targets, predicts = list(), list()
with torch.no_grad():
for fields, target in tqdm.tqdm(data_loader, smoothing=0, mininterval=1.0):
fields, target = fields.to(device), target.to(device)
y = model(fields)
targets.extend(target.tolist())
predicts.extend(y.tolist())
return roc_auc_score(targets, predicts)
def main(dataset_name,
dataset_path,
model_name,
epoch,
learning_rate,
batch_size,
weight_decay,
device,
save_dir):
device = torch.device(device)
dataset = get_dataset(dataset_name, dataset_path)
train_length = int(len(dataset) * 0.8)
valid_length = int(len(dataset) * 0.1)
test_length = len(dataset) - train_length - valid_length
train_dataset, valid_dataset, test_dataset = torch.utils.data.random_split(
dataset, (train_length, valid_length, test_length))
train_data_loader = DataLoader(train_dataset, batch_size=batch_size, num_workers=8)
valid_data_loader = DataLoader(valid_dataset, batch_size=batch_size, num_workers=8)
test_data_loader = DataLoader(test_dataset, batch_size=batch_size, num_workers=8)
model = get_model(model_name, dataset).to(device)
criterion = torch.nn.BCELoss()
optimizer = torch.optim.Adam(params=model.parameters(), lr=learning_rate, weight_decay=weight_decay)
early_stopper = EarlyStopper(num_trials=2, save_path=f'{save_dir}/{model_name}.pt')
for epoch_i in range(epoch):
train(model, optimizer, train_data_loader, criterion, device)
auc = test(model, valid_data_loader, device)
print('epoch:', epoch_i, 'validation: auc:', auc)
if not early_stopper.is_continuable(model, auc):
print(f'validation: best auc: {early_stopper.best_accuracy}')
break
auc = test(model, test_data_loader, device)
print(f'test auc: {auc}')
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--dataset_name', default='criteo')
parser.add_argument('--dataset_path', help='criteo/train.txt, avazu/train, or ml-1m/ratings.dat')
parser.add_argument('--model_name', default='afi')
parser.add_argument('--epoch', type=int, default=100)
parser.add_argument('--learning_rate', type=float, default=0.001)
parser.add_argument('--batch_size', type=int, default=2048)
parser.add_argument('--weight_decay', type=float, default=1e-6)
parser.add_argument('--device', default='cuda:0')
parser.add_argument('--save_dir', default='chkpt')
args = parser.parse_args()
main(args.dataset_name,
args.dataset_path,
args.model_name,
args.epoch,
args.learning_rate,
args.batch_size,
args.weight_decay,
args.device,
args.save_dir)