提交 409a5774 编写于 作者: Y Yu Yang

Complete a very simple mnist demo.

上级 eaba2e2e
"""
A very basic example for how to use current Raw SWIG API to train mnist network.
Current implementation uses Raw SWIG, which means the API call is directly \
passed to C++ side of Paddle.
The user api could be simpler and carefully designed.
"""
import py_paddle.swig_paddle as api import py_paddle.swig_paddle as api
from py_paddle import DataProviderConverter from py_paddle import DataProviderConverter
import paddle.trainer.PyDataProvider2 as dp import paddle.trainer.PyDataProvider2 as dp
import paddle.trainer.config_parser import paddle.trainer.config_parser
import numpy as np import numpy as np
import random
from mnist_util import read_from_mnist from mnist_util import read_from_mnist
...@@ -27,6 +36,18 @@ def generator_to_batch(generator, batch_size): ...@@ -27,6 +36,18 @@ def generator_to_batch(generator, batch_size):
yield ret_val yield ret_val
class BatchPool(object):
def __init__(self, generator, batch_size):
self.data = list(generator)
self.batch_size = batch_size
def __call__(self):
random.shuffle(self.data)
for offset in xrange(0, len(self.data), self.batch_size):
limit = min(offset + self.batch_size, len(self.data))
yield self.data[offset:limit]
def input_order_converter(generator): def input_order_converter(generator):
for each_item in generator: for each_item in generator:
yield each_item['pixel'], each_item['label'] yield each_item['pixel'], each_item['label']
...@@ -37,46 +58,115 @@ def main(): ...@@ -37,46 +58,115 @@ def main():
config = paddle.trainer.config_parser.parse_config( config = paddle.trainer.config_parser.parse_config(
'simple_mnist_network.py', '') 'simple_mnist_network.py', '')
# get enable_types for each optimizer.
# enable_types = [value, gradient, momentum, etc]
# For each optimizer(SGD, Adam), GradientMachine should enable different
# buffers.
opt_config = api.OptimizationConfig.createFromProto(config.opt_config) opt_config = api.OptimizationConfig.createFromProto(config.opt_config)
_temp_optimizer_ = api.ParameterOptimizer.create(opt_config) _temp_optimizer_ = api.ParameterOptimizer.create(opt_config)
enable_types = _temp_optimizer_.getParameterTypes() enable_types = _temp_optimizer_.getParameterTypes()
# Create Simple Gradient Machine.
m = api.GradientMachine.createFromConfigProto( m = api.GradientMachine.createFromConfigProto(
config.model_config, api.CREATE_MODE_NORMAL, enable_types) config.model_config, api.CREATE_MODE_NORMAL, enable_types)
# This type check is not useful. Only enable type hint in IDE.
# Such as PyCharm
assert isinstance(m, api.GradientMachine) assert isinstance(m, api.GradientMachine)
# Initialize Parameter by numpy.
init_parameter(network=m) init_parameter(network=m)
# Create Local Updater. Local means not run in cluster.
# For a cluster training, here we can change to createRemoteUpdater
# in future.
updater = api.ParameterUpdater.createLocalUpdater(opt_config) updater = api.ParameterUpdater.createLocalUpdater(opt_config)
assert isinstance(updater, api.ParameterUpdater) assert isinstance(updater, api.ParameterUpdater)
# Initialize ParameterUpdater.
updater.init(m) updater.init(m)
# DataProvider Converter is a utility convert Python Object to Paddle C++
# Input. The input format is as same as Paddle's DataProvider.
converter = DataProviderConverter( converter = DataProviderConverter(
input_types=[dp.dense_vector(784), dp.integer_value(10)]) input_types=[dp.dense_vector(784), dp.integer_value(10)])
train_file = './data/raw_data/train' train_file = './data/raw_data/train'
test_file = './data/raw_data/t10k'
# start gradient machine.
# the gradient machine must be started before invoke forward/backward.
# not just for training, but also for inference.
m.start() m.start()
for _ in xrange(100): # evaluator can print error rate, etc. It is a C++ class.
updater.startPass() batch_evaluator = m.makeEvaluator()
test_evaluator = m.makeEvaluator()
# Get Train Data.
# TrainData will stored in a data pool. Currently implementation is not care
# about memory, speed. Just a very naive implementation.
train_data_generator = input_order_converter(read_from_mnist(train_file))
train_data = BatchPool(train_data_generator, 128)
# outArgs is Neural Network forward result. Here is not useful, just passed
# to gradient_machine.forward
outArgs = api.Arguments.createArguments(0) outArgs = api.Arguments.createArguments(0)
train_data_generator = input_order_converter(
read_from_mnist(train_file))
for batch_id, data_batch in enumerate(
generator_to_batch(train_data_generator, 2048)):
trainRole = updater.startBatch(len(data_batch))
for pass_id in xrange(2): # we train 2 passes.
updater.startPass()
for batch_id, data_batch in enumerate(train_data()):
# data_batch is input images.
# here, for online learning, we could get data_batch from network.
# Start update one batch.
pass_type = updater.startBatch(len(data_batch))
# Start BatchEvaluator.
# batch_evaluator can be used between start/finish.
batch_evaluator.start()
# A callback when backward.
# It is used for updating weight values vy calculated Gradient.
def updater_callback(param): def updater_callback(param):
updater.update(param) updater.update(param)
# forwardBackward is a shortcut for forward and backward.
# It is sometimes faster than invoke forward/backward separately,
# because in GradientMachine, it may be async.
m.forwardBackward( m.forwardBackward(
converter(data_batch), outArgs, trainRole, updater_callback) converter(data_batch), outArgs, pass_type, updater_callback)
# Get cost. We use numpy to calculate total cost for this batch.
cost_vec = outArgs.getSlotValue(0) cost_vec = outArgs.getSlotValue(0)
cost_vec = cost_vec.copyToNumpyMat() cost_vec = cost_vec.copyToNumpyMat()
cost = cost_vec.sum() / len(data_batch) cost = cost_vec.sum() / len(data_batch)
print 'Batch id', batch_id, 'with cost=', cost
# Make evaluator works.
m.eval(batch_evaluator)
# Print logs.
print 'Pass id', pass_id, 'Batch id', batch_id, 'with cost=', \
cost, batch_evaluator
batch_evaluator.finish()
# Finish batch.
# * will clear gradient.
# * ensure all values should be updated.
updater.finishBatch(cost) updater.finishBatch(cost)
# testing stage. use test data set to test current network.
test_evaluator.start()
test_data_generator = input_order_converter(read_from_mnist(test_file))
for data_batch in generator_to_batch(test_data_generator, 128):
# in testing stage, only forward is needed.
m.forward(converter(data_batch), outArgs, api.PASS_TEST)
m.eval(test_evaluator)
# print error rate for test data set
print 'Pass', pass_id, ' test evaluator: ', test_evaluator
test_evaluator.finish()
updater.finishPass() updater.finishPass()
m.finish() m.finish()
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册