PyTimeloop
|
Classes | |
class | Architecture |
An architecture. More... | |
class | ArchNode |
A node in the architecture hierarchy. More... | |
class | ArchNodes |
A collection of architectural nodes. More... | |
class | Attributes |
A class representing attributes for a node in the architecture. More... | |
class | Branch |
A branch in the architecture. More... | |
class | Component |
A component in the architecture. More... | |
class | Compute |
A compute component. More... | |
class | Container |
A container in the architecture. More... | |
class | Hierarchical |
A hierarchical branch in the architecture. More... | |
class | Leaf |
A leaf node in the architecture hierarchy. More... | |
class | Network |
A network component. More... | |
class | Networks |
A list of networks in the architecture. More... | |
class | Nothing |
A class representing a 'nothing' component. More... | |
class | Parallel |
A parallel branch in the architecture. More... | |
class | Pipelined |
" A pipelined branch in the architecture. More... | |
class | Spatial |
A spatial configuration in a system architecture. More... | |
class | Storage |
A storage component. More... | |
class | StorageAttributes |
Represents the attributes of a storage element. More... | |
Functions | |
"Component" | component_factory (*args, **kwargs) |
Factory function for creating components based on the provided arguments. | |
"Storage" | dummy_storage (str name) |
Create a dummy storage component. | |
Variables | |
tuple | BUFFER_CLASSES = ("DRAM", "SRAM", "regfile", "smartbuffer", "storage") |
tuple | COMPUTE_CLASSES = ("mac", "intmac", "fpmac", "compute") |
tuple | NETWORK_CLASSES = ("XY_NoC", "Legacy", "ReductionTree", "SimpleMulticast") |
tuple | NOTHING_CLASSES = ("nothing",) |
"Component" pytimeloop.timeloopfe.v4.arch.component_factory | ( | * | args, |
** | kwargs ) |
Factory function for creating components based on the provided arguments.
Args: *args: Variable length arguments. Either a single dictionary or keyword arguments. **kwargs: Keyword arguments. Either a single dictionary or keyword arguments.
Returns: Component: The created component.
Raises: TypeError: If both a dictionary and keyword arguments are provided, or if no dictionary is provided. TypeError: If the provided argument is not a dictionary. AssertionError: If the 'class' attribute is missing in the provided dictionary. AssertionError: If the 'class' attribute is not a string. ValueError: If the element class is unknown.
"Storage" pytimeloop.timeloopfe.v4.arch.dummy_storage | ( | str | name | ) |
Create a dummy storage component.
Args: name (str): The name of the storage component.
Returns: Storage: The created dummy storage component.
tuple pytimeloop.timeloopfe.v4.arch.BUFFER_CLASSES = ("DRAM", "SRAM", "regfile", "smartbuffer", "storage") |
tuple pytimeloop.timeloopfe.v4.arch.COMPUTE_CLASSES = ("mac", "intmac", "fpmac", "compute") |
tuple pytimeloop.timeloopfe.v4.arch.NETWORK_CLASSES = ("XY_NoC", "Legacy", "ReductionTree", "SimpleMulticast") |
tuple pytimeloop.timeloopfe.v4.arch.NOTHING_CLASSES = ("nothing",) |