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

Module caffe2.experiments.python.convnet_benchmarks. More...

Functions

def MLP (order)
 
def AlexNet (order)
 
def OverFeat (order)
 
def VGGA (order)
 
def net_DAG_Builder (model)
 
def Inception (order)
 
def AddInput (model, batch_size, db, db_type)
 
def AddParameterUpdate (model)
 
def Benchmark (model_gen, arg)
 
def GetArgumentParser ()
 
def MLP (order, cudnn_ws, mkl)
 
def ResNet50 (order, cudnn_ws, mkl)
 
def AlexNet (order, cudnn_ws, mkl)
 
def OverFeat (order, cudnn_ws, mkl)
 
def VGGA (order, cudnn_ws, mkl)
 
def Inception (order, cudnn_ws, mkl)
 

Variables

 args = GetArgumentParser().parse_args()
 
dictionary model_map
 
 extra_args
 

Detailed Description

Module caffe2.experiments.python.convnet_benchmarks.

Module caffe2.python.convnet_benchmarks.

Function Documentation

def convnet_benchmarks.AddInput (   model,
  batch_size,
  db,
  db_type 
)
Adds the data input part.

Definition at line 529 of file convnet_benchmarks.py.

def convnet_benchmarks.AddParameterUpdate (   model)
Simple plain SGD update -- not tuned to actually train the models 

Definition at line 542 of file convnet_benchmarks.py.

Variable Documentation

dictionary convnet_benchmarks.model_map
Initial value:
1 = {
2  'AlexNet': AlexNet,
3  'OverFeat': OverFeat,
4  'VGGA': VGGA,
5  'Inception': Inception,
6  'MLP': MLP,
7  }

Definition at line 693 of file convnet_benchmarks.py.