我能得到的最接近的例子是在这个问题上:https://github.com/tensorflow/tensorflow/issues/899
有了这个最低限度的可复制代码:
import tensorflow as tf
import tensorflow.python.framework.ops as ops
g = tf.Graph()
with g.as_default():
A = tf.Variable(tf.random_normal( [25,16] ))
B = tf.Variable(tf.random_normal( [16,9] ))
C = tf.matmul(A,B) # shape=[25,9]
for op in g.get_operations():
flops = ops.get_stats_for_node_def(g, op.node_def, 'flops').value
if flops is not None:
print 'Flops should be ~',2*25*16*9
print '25 x 25 x 9 would be',2*25*25*9 # ignores internal dim, repeats first
print 'TF stats gives',flops但是,返回的失败始终为零。是否有一种方法可以具体地测量触发器,特别是使用PB文件?
发布于 2017-11-29 20:23:19
有点晚了,但也许这对将来的一些游客有帮助。对于您的示例,我成功地测试了以下代码段:
g = tf.Graph()
run_meta = tf.RunMetadata()
with g.as_default():
A = tf.Variable(tf.random_normal( [25,16] ))
B = tf.Variable(tf.random_normal( [16,9] ))
C = tf.matmul(A,B) # shape=[25,9]
opts = tf.profiler.ProfileOptionBuilder.float_operation()
flops = tf.profiler.profile(g, run_meta=run_meta, cmd='op', options=opts)
if flops is not None:
print('Flops should be ~',2*25*16*9)
print('25 x 25 x 9 would be',2*25*25*9) # ignores internal dim, repeats first
print('TF stats gives',flops.total_float_ops)还可以结合使用分析器和Keras,如下所示:
import tensorflow as tf
import keras.backend as K
from keras.applications.mobilenet import MobileNet
run_meta = tf.RunMetadata()
with tf.Session(graph=tf.Graph()) as sess:
K.set_session(sess)
net = MobileNet(alpha=.75, input_tensor=tf.placeholder('float32', shape=(1,32,32,3)))
opts = tf.profiler.ProfileOptionBuilder.float_operation()
flops = tf.profiler.profile(sess.graph, run_meta=run_meta, cmd='op', options=opts)
opts = tf.profiler.ProfileOptionBuilder.trainable_variables_parameter()
params = tf.profiler.profile(sess.graph, run_meta=run_meta, cmd='op', options=opts)
print("{:,} --- {:,}".format(flops.total_float_ops, params.total_parameters))希望我能帮上忙!
发布于 2019-11-30 07:25:38
上述方法不再适用于TF2.0,因为分析器方法已被废弃,并在compat.v1下移动。看来这个特性还需要实现。
下面是关于Github的一个问题:https://github.com/tensorflow/tensorflow/issues/32809
https://stackoverflow.com/questions/45085938
复制相似问题