Get the raw data size of the underlying buffer.

Drops common suffixes like _test.cc or -inl.h from filename. For example: >>> _DropCommonSuffixes('foo/foo-inl.h') 'foo/foo' >>> _DropC

File extension (excluding dot) matches a source file extension.

Sets the module's verbosity, and returns the previous setting.

Returns the module's verbosity setting.

Construct pywrap.Source

Make rpc request.

Append a `Layer` to layers

append description of multivalent feature.

Build the dict of Data from flatten value lists. Returns: dict: key is alias, value is `Data`.

Check barrier set by DataLoader. @param name: barrier name @return ErrorCode.OK if barrier reaches READY state, else return ErrorCode.NOT_READY_ERROR.

evaluate accuracy

Reshape and format attributes with int_attr_num, float_attr_num and string_attr_num calculated by decoder.attr_types. Return: Reshaped

Builds `BatchGraph` object from flatten tensors. The order of the build and the order of the flattening needs to be strictly consistent A

edge data edge count: 300 src: range(0, 100) each src has 3 out edges, whose dst are random int in range(0, 100)

induce edge_index.

induce edge_index and implement connected path node labeling. here we only use common neighbor.

Get dict of `Data` in numpy format from flatten values.

get a dict of tensor format `Data` corresponding the given query. Keys of the dict is the aliaes in query.

Get and reformat the raw flatten numpy values list from query.