Caffe2 - Python API
A deep learning, cross platform ML framework
Public Member Functions | List of all members
checkpoint.MultiNodeCheckpointManager Class Reference
Inheritance diagram for checkpoint.MultiNodeCheckpointManager:

Public Member Functions

def __init__ (self, db_prefix, db_type, node_manager_class=CheckpointManager)
 
def init (self, nodes, retrieve_from_epoch=None)
 
def load (self, epoch)
 
def load_blobs_locally (self, nodes, blob_names, epoch, session)
 
def save (self, epoch)
 

Detailed Description

Coordinates checkpointing and checkpointing across multiple nodes.
Each of `init`, `load` and `save` will build TaskGroups which will
trigger checkpointing on each of the nodes involved in a distributed job.

Definition at line 221 of file checkpoint.py.

Member Function Documentation

◆ load_blobs_locally()

def checkpoint.MultiNodeCheckpointManager.load_blobs_locally (   self,
  nodes,
  blob_names,
  epoch,
  session 
)
Loads the necessary blobs from the checkpoints to the current node.

Args:
    blob_names: A list of strings. Each string is the name of a
blob.
    epoch: An integer. The checkpoint epoch to load from.
    session: A Session object to execute the Load ops.

Definition at line 261 of file checkpoint.py.


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