EDIT2:下面的Github链接包含从进程调用TF模型的问题的可能解决方案。它们包括渴望执行和专用服务器进程,通过http请求提供TF模型预测。与每次初始化全局变量并调用tf.train.Server
相比,我想知道是否使用自定义服务器和请求都能赢得胜利,但这似乎是一种更优雅的方式。
我将调查内存泄漏,如果不存在,请关闭此问题。
编辑:添加了该问题的简单可复制示例:
https://github.com/hcl14/Tensorflow-server-launched-from-child-process
背景:我正在运行Tensorflow服务器,并通过“分支”进程连接到它。动态创建(销毁)进程对我来说至关重要-由于weird memory leak,Python探查器不可见(线程无法解决问题),因此将代码的高负载部分移到了那里。因此,我希望快速初始化进程并立即开始工作。仅当进程被破坏时才释放内存。
做实验时,我找到了一个解决方案,将加载的模型和图形保存到全局变量中,然后由子进程(默认情况下使用“ fork”模式)采用,然后调用服务器。
问题:对我来说,奇怪的是,在加载keras模型后,我无法锁定不希望修改的图,并且每次我都需要运行tf.global_variables_initializer()
在子进程中打开新会话。但是,在没有任何会话创建的情况下在主流程中运行虚拟运行就可以了。我知道在这种情况下,tensorflow使用默认会话,但是图上的所有变量都应在模型运行后初始化,因此我希望新的会话可以与先前定义的图一起使用。
因此,我认为修改模型使Python会对子进程(“ fork”模式)产生很多不满,这会增加计算和内存开销。
请原谅我很多代码。我使用的模型对我来说是旧的黑匣子,所以我的问题可能与此有关。 Tensorflow版本为1.2 (我无法升级,模型不兼容), Python 3.6.5 。
此外,也许我的解决方案效率低下,并且有更好的解决方案,对于您的建议我将不胜感激。
我的设置如下:
1.Tensorflow服务器在主进程中启动:
初始化服务器:
def start_tf_server():
import tensorflow as tf
cluster = tf.train.ClusterSpec({"local": [tf_server_address]})
server = tf.train.Server(cluster, job_name="local", task_index=0)
server.join() # block process from exiting
在主要过程中:
p = multiprocessing.Process(target=start_tf_server)
p.daemon=True
p.start() # this process never ends, unless tf server crashes
# WARNING! Graph initialization must be made only after Tf server start!
# Otherwise everything will hang
# I suppose this is because of another session will be
# created before the server one
# init model graph before branching processes
# share graph in the current process scope
interests = init_interests_for_process()
global_vars.multiprocess_globals["interests"] = interests
2。init_interests_for_process()
是模型初始化程序,它加载我的旧模型并在全局变量中共享它。我进行了一次虚拟模型传递,以在图形上初始化所有内容,然后想要锁定图形。但这不起作用:
def init_interests_for_process():
# Prevent errors on my GPU and disable tensorflow
# complaining about CPU instructions
import os
os.environ["CUDA_VISIBLE_DEVICES"]= ""
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
import tensorflow as tf
from tensorflow.contrib.keras import models
# create tensorflow graph
graph = tf.get_default_graph()
with graph.as_default():
TOKENIZER = joblib.load(TOKENIZER_FILE)
NN1_MODEL = models.load_model(NN1_MODEL_FILE)
with open(NN1_CATEGORY_NAMES_FILE, 'r') as f:
NN1_CATEGORY_NAMES = f.read().splitlines()
NN2_MODEL = models.load_model(NN2_MODEL_FILE)
with open(NN2_CATEGORY_NAMES_FILE, 'r') as f:
NN2_CATEGORY_NAMES = f.read().splitlines()
# global variable with all the data to be shared
interests = {}
interests["TOKENIZER"] = TOKENIZER
interests["NN1_MODEL"] = NN1_MODEL
interests["NN1_CATEGORY_NAMES"] = NN1_CATEGORY_NAMES
interests["NN2_MODEL"] = NN2_MODEL
interests["NN2_CATEGORY_NAMES"] = NN2_CATEGORY_NAMES
interests['all_category_names'] = NN1_CATEGORY_NAMES + \
NN2_CATEGORY_NAMES
# Reconstruct a Python object from a file persisted with joblib.dump.
interests["INTEREST_SETTINGS"] = joblib.load(INTEREST_SETTINGS_FILE)
# dummy run to create graph
x = tf.contrib.keras.preprocessing.sequence.pad_sequences(
TOKENIZER.texts_to_sequences("Dummy srting"),
maxlen=interests["INTEREST_SETTINGS"]["INPUT_LENGTH"]
)
y1 = NN1_MODEL.predict(x)
y2 = NN2_MODEL.predict(x)
# PROBLEM: I want, but cannot lock graph, as child process
# wants to run its own tf.global_variables_initializer()
# graph.finalize()
interests["GRAPH"] = graph
return interests
3。现在我生成该进程(实际上,该进程是从另一个进程生成的-层次结构很复杂):
def foo(q):
result = call_function_which_uses_interests_model(some_data)
q.put(result)
return # I've read it is essential for destroying local variables
q = Queue()
p = Process(target=foo,args=(q,))
p.start()
p.join()
result = q.get() # retrieve data
4。在此过程中,我将模型称为:
# retrieve model from global variable
interests = global_vars.multiprocess_globals["interests"]
tokenizer = interests["TOKENIZER"]
nn1_model = interests["NN1_MODEL"]
nn1_category_names = interests["NN1_CATEGORY_NAMES"]
nn2_model = interests["NN2_MODEL"]
nn2_category_names = interests["NN2_CATEGORY_NAMES"]
input_length = interests["INTEREST_SETTINGS"]["INPUT_LENGTH"]
# retrieve graph
graph = interests["GRAPH"]
# open session for server
logger.debug('Trying tf server at ' + 'grpc://'+tf_server_address)
sess = tf.Session('grpc://'+tf_server_address, graph=graph)
# PROBLEM: and I need to run variables initializer:
sess.run(tf.global_variables_initializer())
tf.contrib.keras.backend.set_session(sess)
# finally, make a call to server:
with sess.as_default():
x = tf.contrib.keras.preprocessing.sequence.pad_sequences(
tokenizer.texts_to_sequences(input_str),
maxlen=input_length)
y1 = nn1_model.predict(x)
y2 = nn2_model.predict(x)
一切正常,如果每次生成新进程时我都不锁定图形并运行变量初始化器,那么一切正常。 (除了每个调用大约30-90 MB的内存泄漏,对于python内存分析器不可见)。当我想锁定图形时,会出现有关未初始化变量的错误:
FailedPreconditionError (see above for traceback):
Attempting to use uninitialized value gru_1/bias
[[Node: gru_1/bias/read = Identity[T=DT_FLOAT, _class=["loc:@gru_1/bias"],
_device="/job:local/replica:0/task:0/cpu:0"](gru_1/bias)]]
谢谢!
答案 0 :(得分:1)
您是否考虑过TensorFlow服务? https://www.tensorflow.org/serving/
通常,您需要缓存会话,我认为这是TF Serving使用的策略。到目前为止,这将是将TF模型部署到数据中心的最佳体验。
您也可以选择另一个方向,例如tf.enable_eager_execution()
,这消除了会话的需要。变量仍然被初始化,尽管它是在创建Python变量对象后立即发生的。
但是,如果您确实要创建和销毁Session,则可以用常量("freeze" it)替换图形中的变量。在这种情况下,我还会考虑禁用图形优化,因为默认情况下,使用新的提要和获取集进行的第一个session.run
调用会默认花费一些时间来优化图形(通过{内的RewriterConfig
配置{1}}原型。
(从对问题的评论中展开)
答案 1 :(得分:0)
我不确定这是否能帮到您,但您需要知道在tensorflow中,变量仅针对给定的Session
进行初始化。需要在每个使用的Session
中初始化一个变量-即使在最简单的情况下也是如此:
import tensorflow as tf
x = tf.Variable(0.)
with tf.Session() as sess:
tf.global_variables_initializer().run()
# x is initialized -- no issue here
x.eval()
with tf.Session() as sess:
x.eval()
# Error -- x was never initialized in this session, even though
# it has been initialized before in another session