jsmorph

RemPattern for Quamina

Basic problem and proposed solution

Quamina doesn't currently support RemPattern. Some of the contemplated implementations of RemPattern are difficult. At least one approach is pretty easy: Wrap the current matcher to filter removed patterns from match results and periodically rebuild the matcher from scrach with the live patterns. More specifically:

jsmorph

CSS operations

CSS = Conjunction Streaming Service, which consumes orbital data (currently TLEs) and emits conjunction reports for objects on orbit (or crossing orbits).

Contact

Email contact: info@morphism.com.

jsmorph

To run a model, you need a configuration. The structure of a configuration consists of various Go structures, maps, and other types. We use a fork (with only minor changes) of github.com/alecthomas/jsonschema to generate a JSON schema for model configurations. For one GUI, we then use github.com/json-editor/json-editor to generate a complex HTML/Javascript configuration form dynamically from that JSON schema. A Go program serves the HTML and Javascript, and that program also runs then model. The model output is rendered using d3js.org.

jsmorph

// Package main demos TAOCP, Vol. 2, p. 232 algorithm for incremental variance.
package main

import (
	"fmt"
	"math/rand"
)

type Var struct {
	N int

jsmorph

---- MODULE Kyle1 ----
(* Kyle has a rule that turns on his garage light for 10 minutes every time
   he opens his garage door.  Unfortunately, Kyle has to go into his garage
   frequently (in order to get the next exotic tool he urgently needs
   for a project).  Does his rule really do what he wants? *)
   
EXTENDS Naturals, Sequences

VARIABLES lts,     \* Light timers
          o,       \* Interal state for deciding whether to open the door

jsmorph

jsmorph

Notes on representing and computing uncertainty

Approach

A simple approach that uses sampling from either probability distributions or approximations of distributions as input and output from deterministic functions.

Terminology

As examples, we'll consider observations and predictions. A prediction is computed from a set of observations. We'll work with observations and predictions of probability distributions of tuples .

jsmorph

A Python script for generating KML with timestamps from Pixhawk logs.

See the script for a few comments.

jsmorph

;; Ugly mitigation of http://dev.clojure.org/jira/browse/CLJ-1152 in
;; the protocols case.  Motivation: Need to eval lots domain-specific
;; language code rendered in core.logic.  Approach: Start a thread to
;; purge protocol caches every second.  You should probable write a
;; fancier version.

(defn protocol? [x]
  (and (instance? clojure.lang.PersistentArrayMap x)
       (boolean (:on-interface x))))

jsmorph

java -jar kg.jar --threads 2 --partitioned-grandchildren-test
:kc-open 000-foo.kct
:kc-open 001-foo.kct
:load-synthetic-data #<DB 000-foo.kct: 0: 400128> :start
:load-synthetic-data #<DB 000-foo.kct: 0: 400128> :nodes 0
:load-synthetic-data #<DB 001-foo.kct: 0: 400128> :start
:load-synthetic-data #<DB 001-foo.kct: 0: 400128> :nodes 0
:load-synthetic-data #<DB 001-foo.kct: 44505: 400128> :done
{:triples-added 44505, :nodes 10000, :triples-total 44505, :triples-per-second 35718.29855537721}
:load-synthetic-data #<DB 000-foo.kct: 45235: 400128> :done