caffe2.python.modeling.compute_norm_for_blobs.ComputeNormForBlobs Class Reference

Inheritance diagram for caffe2.python.modeling.compute_norm_for_blobs.ComputeNormForBlobs:

Public Member Functions
def	__init__ (self, blobs, logging_frequency, p=2, compute_averaged_norm=False)
def	modify_net (self, net, init_net=None, grad_map=None, blob_to_device=None, modify_output_record=False)
def	field_name_suffix (self)
Public Member Functions inherited from caffe2.python.modeling.net_modifier.NetModifier
def	__init__ (self)
def	modify_net (self, net, init_net=None, grad_map=None, blob_to_device=None)
def	__call__ (self, net, init_net=None, grad_map=None, blob_to_device=None, modify_output_record=False)

Detailed Description

This class modifies the net passed in by adding ops to compute norms for
certain blobs.

Args:
    blobs: list of blobs to compute norm for
    logging_frequency: frequency for printing norms to logs
    p: type of norm. Currently it supports p=1 or p=2
    compute_averaged_norm: norm or averaged_norm (averaged_norm = norm/size)

Definition at line 13 of file compute_norm_for_blobs.py.

---

The documentation for this class was generated from the following file:

• caffe2/python/modeling/compute_norm_for_blobs.py