Profiler Constructor

Applicability

Description

Constructs an object of the Profiler class, which is used to enable the profiling function locally. For example, you can collect the profile data of a local subgraph on the TensorFlow network or a specified step.

Prototype

1
2
3
4
5
6
7
8

class Profiler(object):
    def __init__(
        self,
        *,
        level: str = "L0",
        aic_metrics: str = "",
        output_path: str = ""
    )

Parameters

Returns

None

Restrictions

• The Profiler class needs to be called using the with statement, and the profile data collection function takes effect in the corresponding scope.
• The Profiler class can be called only in session mode.
• The Profiler class cannot be nested.

  The following is an incorrect calling example:

  1 2 3

  with profiler.Profiler(level="L1", aic_metrics="ArithmeticUtilization", output_path = "./"): with profiler.Profiler(level="L1", aic_metrics="ArithmeticUtilization", output_path = "./"): sess.run(add)

• The Profiler class cannot be used together with parameters profiling_mode and profiling_options in Session Configuration, parameters enable_profiling and profiling_options in NPURunConfig Configuration, and environment variables PROFILING_MODE and PROFILING_OPTIONS. For details about the environment variables, see Environment Variables.
• The Profiler class does not support multi-thread calling.

Example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

import tensorflow as tf
from npu_bridge.npu_init import *

......
a = tf.placeholder(tf.int32, (None,None))
b = tf.constant([[1,2],[2,3]], dtype=tf.int32, shape=(2,2))
add = tf.add(a, b)

with tf.Session(config=session_config, graph=g) as sess:
  with profiler.Profiler(level="L1", aic_metrics=str("ArithmeticUtilization"), output_path = "./"):
    result=sess.run(add, feed_dict={a: [[-20, 2],[1,3]],c: [[1],[-21]]})