models._graph_visualization

Functions related to graph visualization of mlmodels

Module Contents

Functions

_calculate_edges(cy_nodes,cy_edges,shape_dict=None) Parameters
_layer_specific_info(layer) Parameters
_get_custom_layer_value(parameter)
_pipeline_component_info(model,info) Parameters
_neural_network_node_info(nn_spec,cy_nodes,child=False,parent=None) Parameters
_neural_network_nodes_and_edges(nn_spec,cy_nodes,cy_edges,spec_outputs,input_spec,input_shape_dict=None) Parameters
_pipeline_nodes_and_edges(cy_nodes,cy_edges,pipeline_spec,spec_outputs) Parameters
_start_server(port,web_dir) Parameters
_calculate_edges(cy_nodes, cy_edges, shape_dict=None)

cy_nodes: list of nodes for graph cy_edges: list of edges to be updated for graph shape_dict: shape_dict required for inferring shape information

cy_nodes: list of nodes for graph cy_edges: list of edges to be updated for graph

_layer_specific_info(layer)

layer : Can be one of : ‘activation’, ‘add’, ‘average’, ‘batchnorm’, ‘biDirectionalLSTM’, ‘bias’, ‘concat’, ‘convolution’, ‘crop’, ‘dot’, ‘embedding’, ‘flatten’, ‘gru’, ‘innerProduct’, ‘input’, ‘l2normalize’, ‘loadConstant’, ‘lrn’, ‘max’, ‘min’, ‘multiply’, ‘mvn’, ‘name’, ‘output’, ‘padding’, permute’, ‘pooling’, ‘reduce’, ‘reorganizeData’, ‘reshape’, ‘scale’, ‘sequenceRepeat’, ‘simpleRecurrent’, ‘slice’, ‘softmax’, ‘split’,

‘unary’, ‘uniDirectionalLSTM’, ‘upsample’

info : info specific to layer type

_get_custom_layer_value(parameter)
_pipeline_component_info(model, info)

model : pipeline model info : info dict to dump model related info into

model can be one of ‘arrayFeatureExtractor’, ‘categoricalMapping’, ‘dictVectorizer’, ‘featureVectorizer’, ‘glmClassifier’, ‘glmRegressor’, ‘identity’, ‘imputer’, ‘neuralNetwork’, ‘neuralNetworkClassifier’, ‘neuralNetworkRegressor’, ‘normalizer’, ‘oneHotEncoder’, ‘scaler’, ‘supportVectorClassifier’, ‘supportVectorRegressor’, ‘treeEnsembleClassifier’, ‘treeEnsembleRegressor’

info : info dict with required info for model

_neural_network_node_info(nn_spec, cy_nodes, child=False, parent=None)

nn_spec : Neural Network spec of mlmodel cy_nodes: list of nodes to update with nn layers child: If child of a parent pipeline component parent : Parent node of the Neural Network spec

cy_nodes: Updated with layer specific information

_neural_network_nodes_and_edges(nn_spec, cy_nodes, cy_edges, spec_outputs, input_spec, input_shape_dict=None)

nn_spec : Neural Network Spec cy_nodes : list to add nn nodes to cy_edges : list to add edges for nn nodes to spec_outputs : outputs of nn spec input_spec : input spec of Neural Network

cy_data : concatenated list of updated cy_nodes and cy_edges

_pipeline_nodes_and_edges(cy_nodes, cy_edges, pipeline_spec, spec_outputs)

cy_nodes : list to add nn nodes to cy_edges : list to add edges for nn nodes to pipeline_spec: Spec of pipeline mlmodel spec_outputs: spec outputs of pipeline mlmodel

cy_data : concatenated list of updated cy_nodes and cy_edges

_start_server(port, web_dir)

port : localhost port to start server on web_dir: directory containing server files

None