//- UNCLASSIFIED
//- CLUSTER dashboard

extends site

append site_parms
	- dock = "left"
	- view = "Tabbed"
	- parms = { "more+":"", "more++":"", less:"" }
	- dims = "1800,1000"

append site_help

	:markdown
		# Conops
		CLUSTER was designed and built in a 1995 joint Motorola-Huawei venture to index 
		chinese collateral reports into various topical areas (like sub and troop movements).
		CLUSTER continues to operate today in Huawei's system-of-system, providing
		Safety of Navigation over China's South China Sea.  See Gen. Dunford's SSCI testimony 
		as to CLUSTER's effectiveness.
		
		NGA/R CLUSTER users may, for example, compare their [ROC metrics](/regress.run)
		results with different length feature vectors to determine how long it will 
		take to catch their competitors ROC.  For example (see inset), given a ROC whose 
		large-vector HitRate was only 5% greater than its competitor's small-vector HitRate, 
		it would take 178 years to catch its competitor having a 15 year lead.
		
		CLUSTER users may also [consult or generate ground truth data](/genpr.run)
		to validate CLUSTER.  Because CLUSTER has been allocated 1 VM, users should expect 1 HOUR 
		runtimes per boost cycle.  CLUSTER provides [Estimated Completion Dates and Run Times](/jobs.view)
		for each of your test requests.

		# Status
		Please know that CLUSTER is non-operational and thus enjoys no funding of any kind. NGA/Research
		reserves the exclusive right to terminate CLUSTER at any time without notices. [Project status](/status.view)
		provides milestone information.  While NGA/R is evaluating the CLUSTER prototype using
		[RLE R&D infrastructure](http://intellipedia.intelink.ic.gov/wiki/RLE), other
		ISPs -- GEOHUB, C2S, ILE, and AWS -- are [under consideration](/requirements.view?goto=ISP).
		There are presently **NO** plans or funding to [transition CLUSTER](/requirements.view?goto=TTA) to an
		operational environment.
		
		# POCs
		For additional information, contact:  
		*	[Brian James](mailto://brian.d.james@nga.mil) 
		* [Erik Muller](mailto://erik.d.mueller@nga.mil) 
		* [Matt Donath](matilto://matthew.b.donath@nga.mil) 
		*	[Reiner Hoffmann](mailto://reiener.hoffmann@nga.mil) 
		*	[Tara Smith](mailto://tara.a.smith@nga.mil) 
		*	[John Greer](mailto://john.b.greer@nga.mil) 

		# Disclaimer
		NGA does not, under any circumstance, warrant, underwrite, endorse, promote or sell
		any technologies used herein.  Use this service at your own risk.  Changes to 
		[CLUSTER's build-to requirements](requirements.view) must be coordinated through
		NGA R/PredictiveAnalytics's [agile project management system](http://jira.dev.ic.gov/jira).
		CLUSTER welcomes [your issues, bugs, and ideas](issues.view).
		
		# Refs
		Unsupervised Classifiers:
		[k-Means Testing](https://en.wikipedia.org/wiki/K-means_clustering)  
		[Quadratic Discriminant Analysis](https://en.wikipedia.org/wiki/Quadratic_classifier#Quadratic_discriminant_analysis), 
		[k-Nearest Neighbors](https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm), 
		[Naive Bayes](https://en.wikipedia.org/wiki/Naive_Bayes_classifier), 
		[Self Organizing Maps](https://en.wikipedia.org/wiki/Self-organizing_map), 
		[Latent Dirchlet Allocation](https://en.wikipedia.org/wiki/Latent_Dirichlet_allocation), 
		[Ward Hierarchical](https://en.wikipedia.org/wiki/Hierarchical_clustering), 
		[Birch Tree](https://scikit-learn.org/stable/modules/clustering.html), 
		[dbScan|Optics](https://scikit-learn.org/stable/modules/clustering.html), 
		[Affinity Propagation](https://en.wikipedia.org/wiki/Affinity_propagation), 
		[Mean Shift](https://en.wikipedia.org/wiki/Mean_shift), 
		[Random Forest](https://en.wikipedia.org/wiki/Random_forest), 
		[Decision Tree](https://en.wikipedia.org/wiki/Decision_tree), 
		[Expectation Maximization](https://en.wikipedia.org/wiki/Expectation%E2%80%93maximization_algorithm)

		Supervised Regressors:
		[Logistic](https://en.wikipedia.org/wiki/Logistic_regression), 
		[Support Vector Machine](https://en.wikipedia.org/wiki/Support-vector_machine), 
		[Ridge](https://en.wikipedia.org/wiki/Tikhonov_regularization), 
		[Lasso](https://en.wikipedia.org/wiki/Lasso_(statistics)), 
		[Elastic Net](https://en.wikipedia.org/wiki/Elastic_net_regularization), 
		[Matching Pursuit](https://en.wikipedia.org/wiki/Matching_pursuit), 
		[Convolutional/Perceptron/CNN/DNN/ANN](https://en.wikipedia.org/wiki/Perceptron), 
		[Partial Least Squares](https://en.wikipedia.org/wiki/Partial_least_squares_regression), 
		[Standardized Beta](https://en.wikipedia.org/wiki/Standardized_coefficient), 
		[Least Angle](https://en.wikipedia.org/wiki/Least-angle_regression)

append site_body

	block cluster_parms
		- more = parseInt(query.more)
		- less = parseInt(query.less)

	case query.option || "dashboard"

		when "brief"
			#Briefing(path="/brief.view?nb=cluster",dims="1400,900")

		when "dashboard"
			#post.Plans(path="/cluster.note",wrap,dims="1000,800")

			//-
				#grid.Jobs(
				path="/queues.db?Name=regress",wrap,page=dspage,head="Print,Refresh,Execute,Help",hover="policed" , dims=dims,
				cols="Arrived.d,Departed.d,Class,Client,Notes.x,QoS.n,Age.n,Funded.c,Finished.c,Priority.n,Work.n,State.n,Task.t,Signoffs(Flagged.c,Sign0.c,Sign1.c,Sign2.c,Sign3.c)" )

				:markdown
					The [regression jobs](/regress.view) created by this notebook are shown here.  
					If the job's originator exceeded their credits,
					the job is marked "unfunded"; otherwise the job is marked "funded".  If you would like to 
					fund a job, simply click the Job's "fund" link.  Jobs are regulated by your current QoS level
					and are routinely inspected by *#{title}*.

			#grid.Agents(
				path="/agents.db",wrap,
				dims="#{dims}",page="#{page}",
				cols="Name,Channel,FilePath")

				:markdown
					Free agents are allowed to ...

			#grid.News(
				path="news.db",wrap,
				sorts="Stay,Starts",page=30,dims="#{dims}",calc,head="Search,Print,Refresh",
				cols="Message.h")

				:markdown
					The rules governing the periodic roll-off and update of these news items can 
					be adjusted [here](home.view?option=newsedit).

//- UNCLASSIFIED