add evaluation dashboard via github.io page publication

[keighrim]

Epic issue to track progress and issues related to web presence of evaluation results, and possibly evaluation invokers.

[MrSqually]

Considering the use cases of an evaluation dashboard, the following features seem particularly salient:

Evaluation Dashboard Functions

• quick lookup and enumeration of evaluated / "production ready" apps within a given problem space
• comparison of the various CLAMS-ready tools for a given task, across all available metrics
• comparison of prior evaluations to more recent app updates.

As such, I've come up with the following general structure for the evaluation dashboard:

Proposed Evaluation Dashboard Features

• "task oriented" grouping of evaluations, similar to the directory structure
• comparison table between similar app evaluations within a given task + eval dataset
• link or reference to the annotations used by eval (for cross-reference, validation, etc.)

Example Layout

Optical Character Recognition [task heading]

Dataset: aapb-collaboration-batch-xx (symlink to annotations?)

Metric	Tesseract	Parseq	docTR
CER	75	60	30
WER	60	60	20

etc.

Let me know if there's anything missing here that would be important for this page. I'm also playing with the idea of possibly adding external links to documentation for the metrics themselves (e.g., link to a wikipedia explanation of what CER actually is), but I'm not sure there's a programmatic way to generate that kind of documentation, and doing it manually seems like it defeats the purpose.