未验证 提交 3984c24b 编写于 作者: B Bai Yifan 提交者: GitHub

Revert reader (#1216)

* revert reader
上级 a0a381fc
"""
This code is based on https://github.com/fchollet/keras/blob/master/keras/utils/data_utils.py
"""
import time
import numpy as np
import threading
import multiprocessing
try:
import queue
except ImportError:
import Queue as queue
class GeneratorEnqueuer(object):
"""
Builds a queue out of a data generator.
Args:
generator: a generator function which endlessly yields data
use_multiprocessing (bool): use multiprocessing if True,
otherwise use threading.
wait_time (float): time to sleep in-between calls to `put()`.
random_seed (int): Initial seed for workers,
will be incremented by one for each workers.
"""
def __init__(self,
generator,
use_multiprocessing=False,
wait_time=0.05,
random_seed=None):
self.wait_time = wait_time
self._generator = generator
self._use_multiprocessing = use_multiprocessing
self._threads = []
self._stop_event = None
self.queue = None
self._manager = None
self.seed = random_seed
def start(self, workers=1, max_queue_size=10):
"""
Start worker threads which add data from the generator into the queue.
Args:
workers (int): number of worker threads
max_queue_size (int): queue size
(when full, threads could block on `put()`)
"""
def data_generator_task():
"""
Data generator task.
"""
def task():
if (self.queue is not None and
self.queue.qsize() < max_queue_size):
generator_output = next(self._generator)
self.queue.put((generator_output))
else:
time.sleep(self.wait_time)
if not self._use_multiprocessing:
while not self._stop_event.is_set():
with self.genlock:
try:
task()
except Exception:
self._stop_event.set()
break
else:
while not self._stop_event.is_set():
try:
task()
except Exception:
self._stop_event.set()
break
try:
if self._use_multiprocessing:
self._manager = multiprocessing.Manager()
self.queue = self._manager.Queue(maxsize=max_queue_size)
self._stop_event = multiprocessing.Event()
else:
self.genlock = threading.Lock()
self.queue = queue.Queue()
self._stop_event = threading.Event()
for _ in range(workers):
if self._use_multiprocessing:
# Reset random seed else all children processes
# share the same seed
np.random.seed(self.seed)
thread = multiprocessing.Process(target=data_generator_task)
thread.daemon = True
if self.seed is not None:
self.seed += 1
else:
thread = threading.Thread(target=data_generator_task)
self._threads.append(thread)
thread.start()
except:
self.stop()
raise
def is_running(self):
"""
Returns:
bool: Whether the worker theads are running.
"""
return self._stop_event is not None and not self._stop_event.is_set()
def stop(self, timeout=None):
"""
Stops running threads and wait for them to exit, if necessary.
Should be called by the same thread which called `start()`.
Args:
timeout(int|None): maximum time to wait on `thread.join()`.
"""
if self.is_running():
self._stop_event.set()
for thread in self._threads:
if self._use_multiprocessing:
if thread.is_alive():
thread.terminate()
else:
thread.join(timeout)
if self._manager:
self._manager.shutdown()
self._threads = []
self._stop_event = None
self.queue = None
def get(self):
"""
Creates a generator to extract data from the queue.
Skip the data if it is `None`.
# Yields
tuple of data in the queue.
"""
while self.is_running():
if not self.queue.empty():
inputs = self.queue.get()
if inputs is not None:
yield inputs
else:
time.sleep(self.wait_time)
...@@ -22,7 +22,6 @@ import os ...@@ -22,7 +22,6 @@ import os
import time import time
import copy import copy
import six import six
from data_util import GeneratorEnqueuer
class Settings(object): class Settings(object):
...@@ -168,7 +167,7 @@ def preprocess(img, bbox_labels, mode, settings): ...@@ -168,7 +167,7 @@ def preprocess(img, bbox_labels, mode, settings):
return img, sampled_labels return img, sampled_labels
def coco(settings, file_list, mode, batch_size, shuffle): def coco(settings, file_list, mode, shuffle):
# cocoapi # cocoapi
from pycocotools.coco import COCO from pycocotools.coco import COCO
from pycocotools.cocoeval import COCOeval from pycocotools.cocoeval import COCOeval
...@@ -183,10 +182,9 @@ def coco(settings, file_list, mode, batch_size, shuffle): ...@@ -183,10 +182,9 @@ def coco(settings, file_list, mode, batch_size, shuffle):
images = images[:settings.toy] if len(images) > settings.toy else images images = images[:settings.toy] if len(images) > settings.toy else images
print("{} on {} with {} images".format(mode, settings.dataset, len(images))) print("{} on {} with {} images".format(mode, settings.dataset, len(images)))
while True: def reader():
if mode == "train" and shuffle: if mode == 'train' and shuffle:
np.random.shuffle(images) np.random.shuffle(images)
batch_out = []
for image in images: for image in images:
image_name = image['file_name'] image_name = image['file_name']
image_path = os.path.join(settings.data_dir, image_name) image_path = os.path.join(settings.data_dir, image_name)
...@@ -223,28 +221,25 @@ def coco(settings, file_list, mode, batch_size, shuffle): ...@@ -223,28 +221,25 @@ def coco(settings, file_list, mode, batch_size, shuffle):
boxes = sample_labels[:, 1:5] boxes = sample_labels[:, 1:5]
lbls = sample_labels[:, 0].astype('int32') lbls = sample_labels[:, 0].astype('int32')
iscrowd = sample_labels[:, -1].astype('int32') iscrowd = sample_labels[:, -1].astype('int32')
if 'cocoMAP' in settings.ap_version: if 'cocoMAP' in settings.ap_version:
batch_out.append((im, boxes, lbls, iscrowd, yield im, boxes, lbls, iscrowd, \
[im_id, im_width, im_height])) [im_id, im_width, im_height]
else: else:
batch_out.append((im, boxes, lbls, iscrowd)) yield im, boxes, lbls, iscrowd
if len(batch_out) == batch_size:
yield batch_out return reader
batch_out = []
def pascalvoc(settings, file_list, mode, batch_size, shuffle): def pascalvoc(settings, file_list, mode, shuffle):
flist = open(file_list) flist = open(file_list)
images = [line.strip() for line in flist] images = [line.strip() for line in flist]
if not settings.toy == 0: if not settings.toy == 0:
images = images[:settings.toy] if len(images) > settings.toy else images images = images[:settings.toy] if len(images) > settings.toy else images
print("{} on {} with {} images".format(mode, settings.dataset, len(images))) print("{} on {} with {} images".format(mode, settings.dataset, len(images)))
while True: def reader():
if mode == "train" and shuffle: if mode == 'train' and shuffle:
np.random.shuffle(images) np.random.shuffle(images)
batch_out = []
for image in images: for image in images:
image_path, label_path = image.split() image_path, label_path = image.split()
image_path = os.path.join(settings.data_dir, image_path) image_path = os.path.join(settings.data_dir, image_path)
...@@ -278,51 +273,7 @@ def pascalvoc(settings, file_list, mode, batch_size, shuffle): ...@@ -278,51 +273,7 @@ def pascalvoc(settings, file_list, mode, batch_size, shuffle):
boxes = sample_labels[:, 1:5] boxes = sample_labels[:, 1:5]
lbls = sample_labels[:, 0].astype('int32') lbls = sample_labels[:, 0].astype('int32')
difficults = sample_labels[:, -1].astype('int32') difficults = sample_labels[:, -1].astype('int32')
batch_out.append((im, boxes, lbls, difficults)) yield im, boxes, lbls, difficults
if len(batch_out) == batch_size:
yield batch_out
batch_out = []
def batch_reader(settings,
file_list,
batch_size,
mode,
shuffle=True,
num_workers=8):
file_list = os.path.join(settings.data_dir, file_list)
if 'coco' in settings.dataset:
train_settings = copy.copy(settings)
if '2014' in file_list:
sub_dir = "train2014"
elif '2017' in file_list:
sub_dir = "train2017"
train_settings.data_dir = os.path.join(settings.data_dir, sub_dir)
def reader():
try:
if 'coco' in settings.dataset:
enqueuer = GeneratorEnqueuer(
coco(train_settings, file_list, mode, batch_size, shuffle),
use_multiprocessing=False)
else:
enqueuer = GeneratorEnqueuer(
pascalvoc(settings, file_list, mode, batch_size, shuffle),
use_multiprocessing=False)
enqueuer.start(max_queue_size=24, workers=num_workers)
generator_output = None
while True:
while enqueuer.is_running():
if not enqueuer.queue.empty():
generator_output = enqueuer.queue.get()
break
else:
time.sleep(0.01)
yield generator_output
generator_output = None
finally:
if enqueuer is not None:
enqueuer.stop()
return reader return reader
...@@ -341,7 +292,7 @@ def train(settings, file_list, shuffle=True): ...@@ -341,7 +292,7 @@ def train(settings, file_list, shuffle=True):
return pascalvoc(settings, file_list, 'train', shuffle) return pascalvoc(settings, file_list, 'train', shuffle)
def test(settings, file_list, batch_size): def test(settings, file_list):
file_list = os.path.join(settings.data_dir, file_list) file_list = os.path.join(settings.data_dir, file_list)
if 'coco' in settings.dataset: if 'coco' in settings.dataset:
test_settings = copy.copy(settings) test_settings = copy.copy(settings)
...@@ -350,9 +301,9 @@ def test(settings, file_list, batch_size): ...@@ -350,9 +301,9 @@ def test(settings, file_list, batch_size):
elif '2017' in file_list: elif '2017' in file_list:
sub_dir = "val2017" sub_dir = "val2017"
test_settings.data_dir = os.path.join(settings.data_dir, sub_dir) test_settings.data_dir = os.path.join(settings.data_dir, sub_dir)
return coco(test_settings, file_list, 'test', batch_size, False) return coco(test_settings, file_list, 'test', False)
else: else:
return pascalvoc(settings, file_list, 'test', batch_size, False) return pascalvoc(settings, file_list, 'test', False)
def infer(settings, image_path): def infer(settings, image_path):
......
...@@ -15,7 +15,7 @@ parser = argparse.ArgumentParser(description=__doc__) ...@@ -15,7 +15,7 @@ parser = argparse.ArgumentParser(description=__doc__)
add_arg = functools.partial(add_arguments, argparser=parser) add_arg = functools.partial(add_arguments, argparser=parser)
# yapf: disable # yapf: disable
add_arg('learning_rate', float, 0.001, "Learning rate.") add_arg('learning_rate', float, 0.001, "Learning rate.")
add_arg('batch_size', int, 16, "Minibatch size.") add_arg('batch_size', int, 64, "Minibatch size.")
add_arg('num_passes', int, 120, "Epoch number.") add_arg('num_passes', int, 120, "Epoch number.")
add_arg('use_gpu', bool, True, "Whether use GPU.") add_arg('use_gpu', bool, True, "Whether use GPU.")
add_arg('parallel', bool, True, "Parallel.") add_arg('parallel', bool, True, "Parallel.")
...@@ -36,56 +36,6 @@ add_arg('data_dir', str, 'data/pascalvoc', "data directory") ...@@ -36,56 +36,6 @@ add_arg('data_dir', str, 'data/pascalvoc', "data directory")
add_arg('enable_ce', bool, False, "Whether use CE to evaluate the model") add_arg('enable_ce', bool, False, "Whether use CE to evaluate the model")
#yapf: enable #yapf: enable
def build_program(is_train, main_prog, startup_prog, args, data_args,
boundaries=None, values=None, train_file_list=None):
image_shape = [3, data_args.resize_h, data_args.resize_w]
if 'coco' in data_args.dataset:
num_classes = 91
elif 'pascalvoc' in data_args.dataset:
num_classes = 21
def get_optimizer():
if not args.enable_ce:
optimizer = fluid.optimizer.RMSProp(
learning_rate=fluid.layers.piecewise_decay(boundaries, values),
regularization=fluid.regularizer.L2Decay(0.00005), )
else:
optimizer = fluid.optimizer.RMSProp(learning_rate=0.001)
return optimizer
with fluid.program_guard(main_prog, startup_prog):
py_reader = fluid.layers.py_reader(
capacity=64,
shapes=[[-1] + image_shape, [-1, 4], [-1, 1], [-1, 1]],
lod_levels=[0, 1, 1, 1],
dtypes=["float32", "float32", "int32", "int32"],
use_double_buffer=True)
with fluid.unique_name.guard():
image, gt_box, gt_label, difficult = fluid.layers.read_file(py_reader)
locs, confs, box, box_var = mobile_net(num_classes, image, image_shape)
if is_train:
loss = fluid.layers.ssd_loss(locs, confs, gt_box, gt_label, box,
box_var)
loss = fluid.layers.reduce_sum(loss)
optimizer = get_optimizer()
optimizer.minimize(loss)
else:
nmsed_out = fluid.layers.detection_output(
locs, confs, box, box_var, nms_threshold=args.nms_threshold)
with fluid.program_guard(main_prog):
loss = fluid.evaluator.DetectionMAP(
nmsed_out,
gt_label,
gt_box,
difficult,
num_classes,
overlap_threshold=0.5,
evaluate_difficult=False,
ap_version=args.ap_version)
if not is_train:
main_prog = main_prog.clone(for_test=True)
return py_reader, loss
def train(args, def train(args,
train_file_list, train_file_list,
val_file_list, val_file_list,
...@@ -95,140 +45,150 @@ def train(args, ...@@ -95,140 +45,150 @@ def train(args,
num_passes, num_passes,
model_save_dir, model_save_dir,
pretrained_model=None): pretrained_model=None):
if args.enable_ce:
fluid.framework.default_startup_program().random_seed = 111
startup_prog = fluid.Program() image_shape = [3, data_args.resize_h, data_args.resize_w]
train_prog = fluid.Program() if 'coco' in data_args.dataset:
test_prog = fluid.Program() num_classes = 91
elif 'pascalvoc' in data_args.dataset:
num_classes = 21
devices = os.getenv("CUDA_VISIBLE_DEVICES") or "" devices = os.getenv("CUDA_VISIBLE_DEVICES") or ""
devices_num = len(devices.split(",")) devices_num = len(devices.split(","))
image = fluid.layers.data(name='image', shape=image_shape, dtype='float32')
gt_box = fluid.layers.data(
name='gt_box', shape=[4], dtype='float32', lod_level=1)
gt_label = fluid.layers.data(
name='gt_label', shape=[1], dtype='int32', lod_level=1)
difficult = fluid.layers.data(
name='gt_difficult', shape=[1], dtype='int32', lod_level=1)
locs, confs, box, box_var = mobile_net(num_classes, image, image_shape)
nmsed_out = fluid.layers.detection_output(
locs, confs, box, box_var, nms_threshold=args.nms_threshold)
loss = fluid.layers.ssd_loss(locs, confs, gt_box, gt_label, box,
box_var)
loss = fluid.layers.reduce_sum(loss)
test_program = fluid.default_main_program().clone(for_test=True)
with fluid.program_guard(test_program):
map_eval = fluid.evaluator.DetectionMAP(
nmsed_out,
gt_label,
gt_box,
difficult,
num_classes,
overlap_threshold=0.5,
evaluate_difficult=False,
ap_version=args.ap_version)
if 'coco' in data_args.dataset: if 'coco' in data_args.dataset:
# learning rate decay in 12, 19 pass, respectively # learning rate decay in 12, 19 pass, respectively
if '2014' in train_file_list: if '2014' in train_file_list:
epocs = 82783 // batch_size // devices_num epocs = 82783 // batch_size
test_epocs = 40504 // batch_size
boundaries = [epocs * 12, epocs * 19] boundaries = [epocs * 12, epocs * 19]
elif '2017' in train_file_list: elif '2017' in train_file_list:
epocs = 118287 // batch_size // devices_num epocs = 118287 // batch_size
test_epocs = 5000 // batch_size
boundaries = [epocs * 12, epocs * 19] boundaries = [epocs * 12, epocs * 19]
values = [learning_rate, learning_rate * 0.5, values = [
learning_rate * 0.25] learning_rate, learning_rate * 0.5, learning_rate * 0.25
]
elif 'pascalvoc' in data_args.dataset: elif 'pascalvoc' in data_args.dataset:
epocs = 19200 // batch_size // devices_num epocs = 19200 // batch_size
test_epocs = 4952 // batch_size
boundaries = [epocs * 40, epocs * 60, epocs * 80, epocs * 100] boundaries = [epocs * 40, epocs * 60, epocs * 80, epocs * 100]
values = [ values = [
learning_rate, learning_rate * 0.5, learning_rate * 0.25, learning_rate, learning_rate * 0.5, learning_rate * 0.25,
learning_rate * 0.1, learning_rate * 0.01] learning_rate * 0.1, learning_rate * 0.01
]
optimizer = fluid.optimizer.RMSProp(
learning_rate=fluid.layers.piecewise_decay(boundaries, values),
regularization=fluid.regularizer.L2Decay(0.00005), )
if args.enable_ce: optimizer.minimize(loss)
startup_prog.random_seed = 111
train_prog.random_seed = 111
test_prog.random_seed = 111
train_py_reader, loss = build_program(
is_train=True,
main_prog=train_prog,
startup_prog=startup_prog,
args=args,
data_args=data_args,
values = values,
boundaries = boundaries,
train_file_list=train_file_list)
test_py_reader, map_eval = build_program(
is_train=False,
main_prog=test_prog,
startup_prog=startup_prog,
args=args,
data_args=data_args)
place = fluid.CUDAPlace(0) if args.use_gpu else fluid.CPUPlace() place = fluid.CUDAPlace(0) if args.use_gpu else fluid.CPUPlace()
exe = fluid.Executor(place) exe = fluid.Executor(place)
exe.run(startup_prog) exe.run(fluid.default_startup_program())
if pretrained_model: if pretrained_model:
def if_exist(var): def if_exist(var):
return os.path.exists(os.path.join(pretrained_model, var.name)) return os.path.exists(os.path.join(pretrained_model, var.name))
fluid.io.load_vars(exe, pretrained_model, main_program=train_prog, predicate=if_exist) fluid.io.load_vars(exe, pretrained_model, predicate=if_exist)
if args.parallel: if args.parallel:
train_exe = fluid.ParallelExecutor(main_program=train_prog, train_exe = fluid.ParallelExecutor(
use_cuda=args.use_gpu, loss_name=loss.name) use_cuda=args.use_gpu, loss_name=loss.name)
test_exe = fluid.ParallelExecutor(main_program=test_prog,
use_cuda=args.use_gpu, share_vars_from=train_exe)
if not args.enable_ce: if not args.enable_ce:
train_reader = reader.batch_reader(data_args, train_file_list, batch_size, "train") train_reader = paddle.batch(
reader.train(data_args, train_file_list), batch_size=batch_size)
else: else:
import random import random
random.seed(0) random.seed(0)
np.random.seed(0) np.random.seed(0)
train_reader = reader.batch_reader(data_args, train_file_list, batch_size, "train", shuffle=False) train_reader = paddle.batch(
test_reader = reader.batch_reader(data_args, val_file_list, batch_size, "test") reader.train(data_args, train_file_list, False), batch_size=batch_size)
train_py_reader.decorate_paddle_reader(train_reader) test_reader = paddle.batch(
test_py_reader.decorate_paddle_reader(test_reader) reader.test(data_args, val_file_list), batch_size=batch_size)
feeder = fluid.DataFeeder(
def save_model(postfix, main_prog): place=place, feed_list=[image, gt_box, gt_label, difficult])
def save_model(postfix):
model_path = os.path.join(model_save_dir, postfix) model_path = os.path.join(model_save_dir, postfix)
if os.path.isdir(model_path): if os.path.isdir(model_path):
shutil.rmtree(model_path) shutil.rmtree(model_path)
print('save models to %s' % (model_path)) print('save models to %s' % (model_path))
fluid.io.save_persistables(exe, model_path, main_program=main_prog) fluid.io.save_persistables(exe, model_path)
best_map = 0. best_map = 0.
def test(pass_id, best_map): def test(pass_id, best_map):
_, accum_map = map_eval.get_map_var() _, accum_map = map_eval.get_map_var()
map_eval.reset(exe) map_eval.reset(exe)
every_pass_map=[] every_pass_map=[]
test_py_reader.start() for batch_id, data in enumerate(test_reader()):
batch_id = 0 test_map, = exe.run(test_program,
try: feed=feeder.feed(data),
while True: fetch_list=[accum_map])
test_map, = exe.run(test_prog, if batch_id % 20 == 0:
fetch_list=[accum_map]) every_pass_map.append(test_map)
if batch_id % 20 == 0: print("Batch {0}, map {1}".format(batch_id, test_map))
every_pass_map.append(test_map)
print("Batch {0}, map {1}".format(batch_id, test_map))
batch_id += 1
if batch_id > test_epocs:
break
except fluid.core.EOFException:
test_py_reader.reset()
mean_map = np.mean(every_pass_map) mean_map = np.mean(every_pass_map)
if test_map[0] > best_map: if test_map[0] > best_map:
best_map = test_map[0] best_map = test_map[0]
save_model('best_model', test_prog) save_model('best_model')
print("Pass {0}, test map {1}".format(pass_id, test_map)) print("Pass {0}, test map {1}".format(pass_id, test_map))
return best_map, mean_map return best_map, mean_map
for pass_id in range(num_passes): for pass_id in range(num_passes):
batch_begin = time.time() batch_begin = time.time()
start_time = time.time() start_time = time.time()
train_py_reader.start() prev_start_time = start_time
every_pass_loss = [] every_pass_loss = []
batch_id = 0 for batch_id, data in enumerate(train_reader()):
try: prev_start_time = start_time
while True: start_time = time.time()
prev_start_time = start_time if len(data) < (devices_num * 2):
start_time = time.time() print("There are too few data to train on all devices.")
continue
if args.parallel: if args.parallel:
loss_v, = train_exe.run(fetch_list=[loss.name]) loss_v, = train_exe.run(fetch_list=[loss.name],
else: feed=feeder.feed(data))
loss_v, = exe.run(train_prog, fetch_list=[loss]) else:
loss_v = np.mean(np.array(loss_v)) loss_v, = exe.run(fluid.default_main_program(),
every_pass_loss.append(loss_v) feed=feeder.feed(data),
if batch_id % 20 == 0: fetch_list=[loss])
print("Pass {0}, batch {1}, loss {2}, time {3}".format( loss_v = np.mean(np.array(loss_v))
pass_id, batch_id, loss_v, start_time - prev_start_time)) every_pass_loss.append(loss_v)
batch_id += 1 if batch_id % 20 == 0:
if batch_id > epocs: print("Pass {0}, batch {1}, loss {2}, time {3}".format(
break pass_id, batch_id, loss_v, start_time - prev_start_time))
except fluid.core.EOFException:
train_py_reader.reset() end_time = time.time()
batch_end = time.time()
best_map, mean_map = test(pass_id, best_map) best_map, mean_map = test(pass_id, best_map)
if args.enable_ce and pass_id == num_passes - 1: if args.enable_ce and pass_id == args.num_passes - 1:
total_time = batch_end - batch_begin total_time = end_time - start_time
train_avg_loss = np.mean(every_pass_loss) train_avg_loss = np.mean(every_pass_loss)
if devices_num == 1: if devices_num == 1:
print("kpis train_cost %s" % train_avg_loss) print("kpis train_cost %s" % train_avg_loss)
...@@ -240,10 +200,11 @@ def train(args, ...@@ -240,10 +200,11 @@ def train(args,
print("kpis test_acc_card%s %s" % print("kpis test_acc_card%s %s" %
(devices_num, mean_map)) (devices_num, mean_map))
print("kpis train_speed_card%s %f" % print("kpis train_speed_card%s %f" %
(devices_num, test_epocs / total_time)) (devices_num, epocs / total_time))
if pass_id % 10 == 0 or pass_id == num_passes - 1: if pass_id % 10 == 0 or pass_id == num_passes - 1:
save_model(str(pass_id), train_prog) save_model(str(pass_id))
print("Best test map {0}".format(best_map)) print("Best test map {0}".format(best_map))
if __name__ == '__main__': if __name__ == '__main__':
......
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