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
|
import logging
import os
import sys
import importlib
from argparse import ArgumentParser
import csv
import numpy
import theano
from theano import printing
from theano import tensor
from theano.ifelse import ifelse
from blocks.filter import VariableFilter
from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity
from blocks.bricks.lookup import LookupTable
from blocks.initialization import IsotropicGaussian, Constant
from blocks.model import Model
from fuel.datasets.hdf5 import H5PYDataset
from fuel.transformers import Batch
from fuel.streams import DataStream
from fuel.schemes import ConstantScheme, SequentialExampleScheme
from blocks.algorithms import GradientDescent, Scale, AdaDelta, Momentum
from blocks.graph import ComputationGraph
from blocks.main_loop import MainLoop
from blocks.extensions import Printing
from blocks.extensions.saveload import Dump, LoadFromDump
from blocks.extensions.monitoring import DataStreamMonitoring
import data
import transformers
import hdist
import apply_model
if __name__ == "__main__":
if len(sys.argv) != 2:
print >> sys.stderr, 'Usage: %s config' % sys.argv[0]
sys.exit(1)
config = importlib.import_module(sys.argv[1])
def setup_stream():
# Load the training and test data
train = H5PYDataset('/data/lisatmp3/simonet/taxi/data.hdf5', which_set='train', subset=slice(0, config.train_size - config.n_valid), load_in_memory=True)
train = DataStream(train, iteration_scheme=SequentialExampleScheme(config.train_size - config.n_valid))
train = transformers.add_first_k(config.n_begin_end_pts, train)
train = transformers.add_random_k(config.n_begin_end_pts, train)
train = transformers.add_destination(train)
train = transformers.Select(train, ('origin_stand', 'origin_call', 'first_k_latitude', 'last_k_latitude', 'first_k_longitude', 'last_k_longitude', 'destination_latitude', 'destination_longitude'))
train_stream = Batch(train, iteration_scheme=ConstantScheme(config.batch_size))
valid = H5PYDataset('/data/lisatmp3/simonet/taxi/data.hdf5', which_set='train', subset=slice(config.train_size - config.n_valid, config.train_size), load_in_memory=True)
valid = DataStream(valid, iteration_scheme=SequentialExampleScheme(config.n_valid))
valid = transformers.add_first_k(config.n_begin_end_pts, valid)
valid = transformers.add_last_k(config.n_begin_end_pts, valid)
valid = transformers.add_destination(valid)
valid = transformers.Select(valid, ('origin_stand', 'origin_call', 'first_k_latitude', 'last_k_latitude', 'first_k_longitude', 'last_k_longitude', 'destination_latitude', 'destination_longitude'))
valid_stream = Batch(valid, iteration_scheme=ConstantScheme(1000))
return (train_stream, valid_stream)
def main():
# The input and the targets
x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.porto_center[0]) / data.data_std[0]
x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.porto_center[1]) / data.data_std[1]
x_firstk = tensor.concatenate((x_firstk_latitude, x_firstk_longitude), axis=1)
x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.porto_center[0]) / data.data_std[0]
x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.porto_center[1]) / data.data_std[1]
x_lastk = tensor.concatenate((x_lastk_latitude, x_lastk_longitude), axis=1)
x_client = tensor.lvector('origin_call')
x_stand = tensor.lvector('origin_stand')
y = tensor.concatenate((tensor.vector('destination_latitude')[:, None], tensor.vector('destination_longitude')[:, None]), axis=1)
# Define the model
client_embed_table = LookupTable(length=config.n_clients+1, dim=config.dim_embed, name='client_lookup')
stand_embed_table = LookupTable(length=config.n_stands+1, dim=config.dim_embed, name='stand_lookup')
mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + [None],
dims=[config.dim_input] + config.dim_hidden + [config.dim_output])
# Create the Theano variables
client_embed = client_embed_table.apply(x_client).flatten(ndim=2)
stand_embed = stand_embed_table.apply(x_stand).flatten(ndim=2)
inputs = tensor.concatenate([x_firstk, x_lastk, client_embed, stand_embed],
axis=1)
# inputs = theano.printing.Print("inputs")(inputs)
outputs = mlp.apply(inputs)
# Normalize & Center
outputs = data.data_std * outputs + data.porto_center
outputs.name = 'outputs'
# Calculate the cost
cost = (outputs - y).norm(2, axis=1).mean()
cost.name = 'cost'
hcost = hdist.hdist(outputs, y).mean()
hcost.name = 'hcost'
# Initialization
client_embed_table.weights_init = IsotropicGaussian(0.001)
stand_embed_table.weights_init = IsotropicGaussian(0.001)
mlp.weights_init = IsotropicGaussian(0.01)
mlp.biases_init = Constant(0.001)
client_embed_table.initialize()
stand_embed_table.initialize()
mlp.initialize()
(train_stream, valid_stream) = setup_stream()
# Training
cg = ComputationGraph(cost)
algorithm = GradientDescent(
cost=cost,
# step_rule=AdaDelta(decay_rate=0.5),
step_rule=Momentum(learning_rate=config.learning_rate, momentum=config.momentum),
params=cg.parameters)
extensions=[DataStreamMonitoring([cost, hcost], valid_stream,
prefix='valid',
every_n_batches=1000),
Printing(every_n_batches=1000),
# Dump('taxi_model', every_n_batches=100),
# LoadFromDump('taxi_model'),
]
main_loop = MainLoop(
model=Model([cost]),
data_stream=train_stream,
algorithm=algorithm,
extensions=extensions)
main_loop.run()
# Produce an output on the test data
'''
test = data.test_data
test = DataStream(test)
test = transformers.add_first_k(conifg.n_begin_end_pts, test)
test = transformers.add_last_k(config.n_begin_end_pts, test)
test = transformers.Select(test, ('trip_id', 'origin_stand', 'origin_call', 'first_k', 'last_k'))
test_stream = Batch(test, iteration_scheme=ConstantScheme(1000))
outfile = open("test-output.csv", "w")
outcsv = csv.writer(outfile)
outcsv.writerow(["TRIP_ID", "LATITUDE", "LONGITUDE"])
for out in apply_model.Apply(outputs=outputs, stream=test_stream, return_vars=['trip_id', 'outputs']):
dest = out['outputs']
for i, trip in enumerate(out['trip_id']):
outcsv.writerow([trip, repr(dest[i, 1]), repr(dest[i, 0])])
outfile.close()
'''
if __name__ == "__main__":
logging.basicConfig(level=logging.INFO)
main()
|
