Caffe2 - Python API
A deep learning, cross platform ML framework
Classes | Functions
net_printer Namespace Reference

Module caffe2.python.net_printer. More...

Classes

class  Analyzer
 
class  Printer
 
class  Text
 
class  Visitor
 

Functions

def analyze_op (analyzer, op)
 
def analyze_net (analyzer, net)
 
def analyze_step (analyzer, step)
 
def analyze_task (analyzer, task)
 
def analyze_task_group (analyzer, tg)
 
def analyze_job (analyzer, job)
 
def analyze (obj)
 
def commonprefix (m)
 
def factor_prefix (vals, do_it)
 
def call (op, inputs=None, outputs=None, factor_prefixes=False)
 
def print_op (text, op)
 
def print_net (text, net)
 
def print_step (text, step)
 
def print_task (text, task)
 
def print_task_group (text, tg, header=None)
 
def print_job (text, job)
 
def to_string (obj)
 
def debug_net (net)
 

Detailed Description

Module caffe2.python.net_printer.

Function Documentation

◆ analyze()

def net_printer.analyze (   obj)
Given a Job, visits all the execution steps making sure that:
  - no undefined blobs will be found during excution
  - no blob with same name is defined in concurrent steps

Definition at line 135 of file net_printer.py.

◆ debug_net()

def net_printer.debug_net (   net)
Given a Net, produce another net that logs info about the operator call
before each operator execution. Use for debugging purposes.

Definition at line 328 of file net_printer.py.

◆ to_string()

def net_printer.to_string (   obj)
Given a Net, ExecutionStep, Task, TaskGroup or Job, produces a string
with detailed description of the execution steps.

Definition at line 318 of file net_printer.py.