ZigZag - Deep Learning Hardware Design Space Exploration
This repository presents the novel version of our tried-and-tested hardware Architecture-Mapping Design Space Exploration (DSE) Framework for Deep Learning (DL) accelerators. ZigZag bridges the gap between algorithmic DL decisions and their acceleration cost on specialized accelerators through a fast and accurate hardware cost estimation.
DefaultNodeParser Class Reference

This class parses an ONNX node into a DummyNode. More...

Inheritance diagram for DefaultNodeParser:
Collaboration diagram for DefaultNodeParser:

Public Member Functions

DummyNode run (self)
 Run the parser. More...
 
DummyNode generate_dummy_node (self)
 
- Public Member Functions inherited from ONNXOperatorParser
None __init__ (self, int node_id, NodeProto node, dict[int, Any] nodes_outputs, ModelProto onnx_model, *list[dict[str, Any]]|None mapping_data=None, Accelerator|None accelerator=None)
 
def get_input_output_weight_data_type (self)
 Return the data type of the input, output and weight tensors of this node. More...
 
def get_weight_name (self, NodeProto node)
 Return the name of the weight input of this node depending on its operator type. More...
 
list[int] get_node_predecessors (self)
 Compute node input sources. More...
 
def get_operand_source_user_format (self, list[int] predecessors)
 Set input source and indicate constant operands. More...
 
def get_weight_precision (self)
 Return the weight precision for this node. More...
 
def get_activation_precision (self)
 Return the activation precision for this node. More...
 
def get_intermediate_output_precision (self)
 Return the intermediate output precision for this node. More...
 

Additional Inherited Members

- Public Attributes inherited from ONNXOperatorParser
 node_id
 
 node
 
 nodes_outputs
 
 onnx_model
 
 mapping_data
 
 accelerator
 
- Static Public Attributes inherited from ONNXOperatorParser
string CUSTOM_WEIGHT_SIZE_ATTR = "weight_size"
 
string CUSTOM_ACT_SIZE_ATTR = "act_size"
 
string CUSTOM_OUTPUT_SIZE_ATTR = "output_size"
 

Detailed Description

This class parses an ONNX node into a DummyNode.

Member Function Documentation

◆ generate_dummy_node()

DummyNode generate_dummy_node (   self)
Here is the call graph for this function:
Here is the caller graph for this function:

◆ run()

DummyNode run (   self)

Run the parser.

Reimplemented from ONNXOperatorParser.

Here is the call graph for this function:
Here is the caller graph for this function:

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