Core class oriented to manage the comunication and execution messages pass for all components on system orchestrating the execution of actions activities (train and prediction) on specific frameworks. More...

Inheritance diagram for gdayf.core.controller.Controller:

[legend]

Collaboration diagram for gdayf.core.controller.Controller:

[legend]

Public Member Functions
def	__init__ (self, e_c=None, user_id='PoC_gDayF', workflow_id='default')
	Constructor.

def	config_checks (self)
	Method leading configurations coherence checks. More...

def	exec_prediction (self, datapath, armetadata=None, model_file=None)
	Method leading and controlling prediction's executions on all frameworks. More...

def	clean_handler (self, fw)
	Method focus on cleaning handler objects. More...

def	init_handler (self, fw)
	Method oriented to init handler objects. More...

def	clean_handlers (self)
	Method oriented to shutdown localClusters.

def	exec_analysis (self, datapath, objective_column, amode=POC, metric='test_accuracy', deep_impact=3, kwargs)
	Method leading and controlling analysis's executions on all frameworks. More...

def	log_model_list (self, ar_list, metric)
	Method oriented to log leaderboard against selected metrics. More...

def	table_model_list (self, ar_list, metric)
	Method oriented to log leaderboard against selected metrics on dataframe. More...

def	exec_sanalysis (self, datapath, list_ar_metadata, metric='combined_accuracy', deep_impact=1, kwargs)
	Method leading and controlling analysis's executions on specific analysis. More...

def	get_external_model (self, armetadata, type='pojo')
	Method leading and controlling coversion to java model. More...

def	save_models (self, arlist, mode=BEST, metric='accuracy')
	Method leading and controlling model savings. More...

def	load_models (self, arlist)
	Method leading and controlling model loads. More...

def	remove_models (self, arlist, mode=ALL)
	Method leading and controlling model removing from server. More...

def	reconstruct_execution_tree (self, arlist=None, metric='combined', store=True)
	Method oriented to generate execution tree for visualizations and analysis issues. More...

def	priorize_list (self, arlist, metric)
	Method oriented to priorize ARlist. More...

def	get_ar_from_engine (self, path)
	Method base to get an ArMetadata Structure from file. More...

Public Attributes
	timestamp

	analysis_list

	model_handler

	adviser

Detailed Description

Core class oriented to manage the comunication and execution messages pass for all components on system orchestrating the execution of actions activities (train and prediction) on specific frameworks.

Definition at line 41 of file controller.py.

Member Function Documentation

◆ clean_handler()

def gdayf.core.controller.Controller.clean_handler	(	self,
		fw
	)

Method focus on cleaning handler objects.

Parameters

fw framework

Definition at line 276 of file controller.py.

Here is the caller graph for this function:

◆ config_checks()

def gdayf.core.controller.Controller.config_checks ( self )

Method leading configurations coherence checks.

Parameters

self	object pointer

Returns: True if OK / False if wrong

Definition at line 66 of file controller.py.

Here is the call graph for this function:

◆ exec_analysis()

def gdayf.core.controller.Controller.exec_analysis	(	self,
		datapath,
		objective_column,
		amode = `POC`,
		metric = `'test_accuracy'`,
		deep_impact = `3`,
		kwargs
	)

Method leading and controlling analysis's executions on all frameworks.

Parameters

self	object pointer
datapath	String Path indicating file to be analyzed or DataFrame
objective_column	string indicating objective column
amode	Analysis mode of execution [0,1,2,3,4,5,6]
metric	to evalute models ['train_accuracy', 'train_rmse', 'test_accuracy', 'combined_accuracy', 'test_rmse', 'cdistance']
deep_impact	deep analysis

Returns: status, adviser.analysis_recommendation_order

Definition at line 321 of file controller.py.

Here is the call graph for this function:

◆ exec_prediction()

def gdayf.core.controller.Controller.exec_prediction	(	self,
		datapath,
		armetadata = `None`,
		model_file = `None`
	)

Method leading and controlling prediction's executions on all frameworks.

Parameters

self	object pointer
datapath	String Path indicating file to be analyzed or Dataframe
armetadata
model_file	String Path indicating model_file ArMetadata.json structure

Definition at line 211 of file controller.py.

Here is the call graph for this function:

◆ exec_sanalysis()

def gdayf.core.controller.Controller.exec_sanalysis	(	self,
		datapath,
		list_ar_metadata,
		metric = `'combined_accuracy'`,
		deep_impact = `1`,
		kwargs
	)

Method leading and controlling analysis's executions on specific analysis.

Parameters

self	object pointer
datapath	String Path indicating file to be analyzed or DataFrame
list_ar_metadata	list of models to execute
metric	to evalute models
deep_impact	deep analysis

Returns: status, adviser.analysis_recommendation_order

Definition at line 554 of file controller.py.

Here is the call graph for this function:

◆ get_ar_from_engine()

def gdayf.core.controller.Controller.get_ar_from_engine	(	self,
		path
	)

Method base to get an ArMetadata Structure from file.

Parameters

self	object pointer
path	FilePath

Returns: operation status (0 success /1 error, ArMetadata/None)

Definition at line 834 of file controller.py.

◆ get_external_model()

def gdayf.core.controller.Controller.get_external_model	(	self,
		armetadata,
		type = `'pojo'`
	)

Method leading and controlling coversion to java model.

Parameters

self	object pointer
armetadata	Armetada object
type	base type if is possible

Returns: download_path, hash MD5 key

Definition at line 647 of file controller.py.

Here is the call graph for this function:

◆ init_handler()

def gdayf.core.controller.Controller.init_handler	(	self,
		fw
	)

Method oriented to init handler objects.

Parameters

fw framework

Definition at line 283 of file controller.py.

Here is the caller graph for this function:

◆ load_models()

def gdayf.core.controller.Controller.load_models	(	self,
		arlist
	)

Method leading and controlling model loads.

Parameters

self	object pointer
arlist	List of armetadata

Returns: list of ar_descriptors of models correctly loaded

Definition at line 688 of file controller.py.

Here is the call graph for this function:

◆ log_model_list()

def gdayf.core.controller.Controller.log_model_list	(	self,
		ar_list,
		metric
	)

Method oriented to log leaderboard against selected metrics.

Parameters

ar_list	List of AR models Execution Data
metric	to execute order ['train_accuracy', 'train_rmse', 'test_accuracy', 'combined_accuracy', 'test_rmse', 'cdistance']

Definition at line 447 of file controller.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ priorize_list()

def gdayf.core.controller.Controller.priorize_list	(	self,
		arlist,
		metric
	)

Method oriented to priorize ARlist.

Parameters

self	object pointer
analysis_id
arlist	Priorized ArMetadata list
metric	['accuracy', 'combined', 'test_accuracy', 'rmse']

Returns: OrderedDict() with execution tree data Analysis

Definition at line 824 of file controller.py.

Here is the caller graph for this function:

◆ reconstruct_execution_tree()

def gdayf.core.controller.Controller.reconstruct_execution_tree	(	self,
		arlist = `None`,
		metric = `'combined'`,
		store = `True`
	)

Method oriented to generate execution tree for visualizations and analysis issues.

Parameters

arlist	Priorized ArMetadata list
metric	['accuracy', 'combined', 'test_accuracy', 'rmse']
store	True/False
experiment	analysys_id for mongoDB recovery
user	user_id for mongoDB recovery

Returns: OrderedDict() with execution tree data Analysis

Definition at line 740 of file controller.py.

Here is the call graph for this function:

◆ remove_models()

def gdayf.core.controller.Controller.remove_models	(	self,
		arlist,
		mode = `ALL`
	)

Method leading and controlling model removing from server.

Parameters

self	object pointer
mode	to be keeped in memory [BEST, BEST_3, EACH_BEST, ALL,NONE]
arlist	List of armetadata

Definition at line 704 of file controller.py.

Here is the call graph for this function:

◆ save_models()

def gdayf.core.controller.Controller.save_models	(	self,
		arlist,
		mode = `BEST`,
		metric = `'accuracy'`
	)

Method leading and controlling model savings.

Parameters

self	object pointer
mode	[BEST, BEST_3, EACH_BEST, ALL]
arlist	List of armetadata
metric	['accuracy', 'combined', 'test_accuracy', 'rmse']

Definition at line 659 of file controller.py.

Here is the call graph for this function:

◆ table_model_list()

def gdayf.core.controller.Controller.table_model_list	(	self,
		ar_list,
		metric
	)

Method oriented to log leaderboard against selected metrics on dataframe.

Parameters

analysis_id
ar_list	List of AR models Execution Data
metric	to execute order ['train_accuracy', 'train_rmse', 'test_accuracy', 'combined_accuracy', 'test_rmse', 'cdistance']

Returns: Dataframe performance model list

Definition at line 495 of file controller.py.

Here is the call graph for this function:

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

D:/OneDrive/DayF/git/gdayf-core/src/gdayf/core/controller.py

Public Member Functions

Public Attributes

Detailed Description

Member Function Documentation

◆ clean_handler()

◆ config_checks()

◆ exec_analysis()

◆ exec_prediction()

◆ exec_sanalysis()

◆ get_ar_from_engine()

◆ get_external_model()

◆ init_handler()

◆ load_models()

◆ log_model_list()

◆ priorize_list()

◆ reconstruct_execution_tree()

◆ remove_models()

◆ save_models()

◆ table_model_list()