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

Module caffe2.python.workspace. More...

Classes

class  _BlobDict
 

Functions

def GetCudaPeerAccessPattern ()
 
def StartMint (root_folder=None, port=None)
 
def StringifyProto (obj)
 
def ResetWorkspace (root_folder=None)
 
def CreateNet (net, overwrite=False, input_blobs=None)
 
def RunOperatorOnce (operator)
 
def RunOperatorsOnce (operators)
 
def RunNetOnce (net)
 
def RunNet (name, num_iter=1)
 
def RunPlan (plan_or_step)
 
def InferShapesAndTypes (nets, blob_dimensions=None)
 
def StringifyBlobName (name)
 
def StringifyNetName (name)
 
def FeedBlob (name, arr, device_option=None)
 
def FetchBlobs (names)
 
def FetchBlob (name)
 
def GetNameScope ()
 
def IsImmediate ()
 
def WorkspaceGuard (workspace_name)
 
def StartImmediate (i_know=False)
 
def StopImmediate ()
 
def ImmediateBlobs ()
 
def RunOperatorImmediate (op)
 
def FetchImmediate (args, kwargs)
 
def FeedImmediate (args, kwargs)
 

Variables

 logger = logging.getLogger(__name__)
 
 Blobs = C.blobs
 
 CreateBlob = C.create_blob
 
 CurrentWorkspace = C.current_workspace
 
 DeserializeBlob = C.deserialize_blob
 
 GlobalInit = C.global_init
 
 HasBlob = C.has_blob
 
 RegisteredOperators = C.registered_operators
 
 SerializeBlob = C.serialize_blob
 
 SwitchWorkspace = C.switch_workspace
 
 RootFolder = C.root_folder
 
 Workspaces = C.workspaces
 
 BenchmarkNet = C.benchmark_net
 
 Predictor = C.Predictor
 
 is_asan = C.is_asan
 
 has_gpu_support = C.has_gpu_support
 
 NumCudaDevices = C.num_cuda_devices
 
 SetDefaultGPUID = C.set_default_gpu_id
 
 GetDefaultGPUID = C.get_default_gpu_id
 
 GetCuDNNVersion = C.get_cudnn_version
 
 GetCudaPeerAccessPattern = lambda: np.array([])
 
 basestring = str
 
 blobs = _BlobDict()
 

Detailed Description

Module caffe2.python.workspace.

Function Documentation

◆ FeedBlob()

def workspace.FeedBlob (   name,
  arr,
  device_option = None 
)
Feeds a blob into the workspace.

Inputs:
  name: the name of the blob.
  arr: either a TensorProto object or a numpy array object to be fed into
      the workspace.
  device_option (optional): the device option to feed the data with.
Returns:
  True or False, stating whether the feed is successful.

Definition at line 229 of file workspace.py.

◆ FetchBlob()

def workspace.FetchBlob (   name)
Fetches a blob from the workspace.

Inputs:
  name: the name of the blob - a string or a BlobReference
Returns:
  Fetched blob (numpy array or string) if successful

Definition at line 276 of file workspace.py.

◆ FetchBlobs()

def workspace.FetchBlobs (   names)
Fetches a list of blobs from the workspace.

Inputs:
    names: list of names of blobs - strings or BlobReferences
Returns:
    list of fetched blobs

Definition at line 265 of file workspace.py.

◆ GetNameScope()

def workspace.GetNameScope ( )
Return the current namescope string. To be used to fetch blobs

Definition at line 287 of file workspace.py.

◆ InferShapesAndTypes()

def workspace.InferShapesAndTypes (   nets,
  blob_dimensions = None 
)
Infers the shapes and types for the specified nets.

Inputs:
  nets: the list of nets
  blob_dimensions (optional): a dictionary of blobs and their dimensions.
      If not specified, the workspace blobs are used.
Returns:
  A tuple of (shapes, types) dictionaries keyed by blob name.

Definition at line 184 of file workspace.py.

◆ RunNet()

def workspace.RunNet (   name,
  num_iter = 1 
)
Runs a given net.

Inputs:
  name: the name of the net, or a reference to the net.
  num_iter: number of iterations to run
Returns:
  True or an exception.

Definition at line 164 of file workspace.py.

◆ StartMint()

def workspace.StartMint (   root_folder = None,
  port = None 
)
Start a mint instance.

TODO(Yangqing): this does not work well under ipython yet. According to
    https://github.com/ipython/ipython/issues/5862
writing up some fix is a todo item.

Definition at line 81 of file workspace.py.

◆ StopImmediate()

def workspace.StopImmediate ( )
Stops an immediate mode run.

Definition at line 401 of file workspace.py.

◆ StringifyProto()

def workspace.StringifyProto (   obj)
Stringify a protocol buffer object.

  Inputs:
obj: a protocol buffer object, or a Pycaffe2 object that has a Proto()
    function.
  Outputs:
string: the output protobuf string.
  Raises:
AttributeError: if the passed in object does not have the right attribute.

Definition at line 105 of file workspace.py.