Custom Op - Channels Last

Channels Last Custom Ops

class qonnx.custom_op.channels_last.base_wrapped_op.ChannelsLastWrappedOp(onnx_node, onnx_opset_version=11)

execute_node(context, graph): Execute this CustomOp instance, given the execution context and ONNX graph.

infer_node_datatype(model): Set the DataType annotations corresponding to the outputs of this node.

verify_node(): Verifies that all attributes the node needs are there and that particular attributes are set correctly. Also checks if the number of inputs is equal to the expected number.

qonnx.custom_op.channels_last.base_wrapped_op.to_channels_first_args(ndim): Returns the tuple of parameters to transpose a channels last tensor to a channels first one. :param ndim: Number of dimensions of the tensor to be transposed.

qonnx.custom_op.channels_last.base_wrapped_op.to_channels_last_args(ndim): Returns the tuple of parameters to transpose a channels first tensor to a channels last one. :param ndim: Number of dimensions of the tensor to be transposed.

class qonnx.custom_op.channels_last.batch_normalization.BatchNormalization(onnx_node, onnx_opset_version=11)

get_nodeattr_types(): Returns a dict of permitted attributes for node, where: ret_dict[attribute_name] = (dtype, require, default_value, <allowed_values>) - dtype indicates which member of the ONNX AttributeProto will be utilized - require indicates whether this attribute is required - default_val indicates the default value that will be used if the attribute is not set - <allowed_values> (if specified) indicates that this attribute can only be set to one of the values in the set <allowed_values>. If not specified, all values permitted by dtype are allowed.

make_shape_compatible_op(model): Returns a standard ONNX op which is compatible with this CustomOp for performing shape inference.

verify_node(): Verifies that all attributes the node needs are there and that particular attributes are set correctly. Also checks if the number of inputs is equal to the expected number.

class qonnx.custom_op.channels_last.conv.Conv(onnx_node, onnx_opset_version=11)

get_nodeattr_types(): Returns a dict of permitted attributes for node, where: ret_dict[attribute_name] = (dtype, require, default_value, <allowed_values>) - dtype indicates which member of the ONNX AttributeProto will be utilized - require indicates whether this attribute is required - default_val indicates the default value that will be used if the attribute is not set - <allowed_values> (if specified) indicates that this attribute can only be set to one of the values in the set <allowed_values>. If not specified, all values permitted by dtype are allowed.

make_shape_compatible_op(model): Returns a standard ONNX op which is compatible with this CustomOp for performing shape inference.

verify_node(): Verifies that all attributes the node needs are there and that particular attributes are set correctly. Also checks if the number of inputs is equal to the expected number.

class qonnx.custom_op.channels_last.max_pool.MaxPool(onnx_node, onnx_opset_version=11)

get_nodeattr_types(): Returns a dict of permitted attributes for node, where: ret_dict[attribute_name] = (dtype, require, default_value, <allowed_values>) - dtype indicates which member of the ONNX AttributeProto will be utilized - require indicates whether this attribute is required - default_val indicates the default value that will be used if the attribute is not set - <allowed_values> (if specified) indicates that this attribute can only be set to one of the values in the set <allowed_values>. If not specified, all values permitted by dtype are allowed.

make_shape_compatible_op(model): Returns a standard ONNX op which is compatible with this CustomOp for performing shape inference.

verify_node(): Verifies that all attributes the node needs are there and that particular attributes are set correctly. Also checks if the number of inputs is equal to the expected number.