Merge pull request #47 from JuliaRobotics/feature/cloudgraphdfg

Updating README

Author: GearsAD

README.md

@@ -15,9 +15,58 @@ DistributedFactorGraphs.jl provides a flexible factor graph API for use in the [
 Please see the [documentation](http://juliarobotics.github.io/DistributedFactorGraphs.jl/latest/) and the [unit tests](https://github.com/JuliaRobotics/DistributedFactorGraphs.jl/tree/master/test) for examples on using DistributedFactorGraphs.jl.
 
 # Installation
-DistributedFactorGraphs.jl has not been registered yet, and should be installed by cloning this repository:
+DistributedFactorGraphs can be installed from Julia packages using:
+```julia
+add DistributedFactorGraphs
+```
+
+# Usage
+DistributedFactorGraphs (DFG) currently supports two implementations:
+* An in-memory factor graph based on Graphs.jl
+* A Neo4j-based factor graph based on Neo4j.jl
+
+The in-memory implementation is the default. The Neo4j driver can be enabled by importing Neo4j before DFG:
 
 ```julia
-Pkg.clone("https://github.com/JuliaRobotics/DistributedFactorGraphs.jl.git")
+# To enable the Neo4j driver, import Neo4j.jl first
+using Neo4j
+using DistributedFactorGraphs
 ```
 
+Both drivers support the same functions, so choose which you want to use when creating your initial DFG. For example:
+
+```julia
+# In-memory DFG
+dfg = GraphsDFG{NoSolverParams}()
+addVariable!(dfg, DFGVariable(:a))
+addVariable!(dfg, DFGVariable(:b))
+addFactor!(dfg, [v1, v2], DFGFactor{Int, :Symbol}(:f1)) # Rather use a RoME-type factor here (e.g. Pose2Pose2) rather than an Int, this is just for demonstrative purposes.
+```
+
+```julia
+# Neo4j-based DFG
+dfg = CloudGraphsDFG{NoSolverParams}("localhost", 7474, "neo4j", "test",
+      "testUser", "testRobot", "testSession",
+      nothing,
+      nothing,
+      IncrementalInference.decodePackedType)
+addVariable!(dfg, DFGVariable(:a))
+addVariable!(dfg, DFGVariable(:b))
+addFactor!(dfg, [v1, v2], DFGFactor{Int, :Symbol}(:f1)) # Rather use a RoME-type factor here (e.g. Pose2Pose2) rather than an Int, this is just for demonstrative purposes.
+```
+
+Please see the documentation for more information on interacting with the factor graph.
+
+## Setting up a Quick Neo4j Database
+The simplest way to set up a test database is with Docker.
+
+To pull the Neo4j image:
+```bash
+docker pull neo4j
+```
+
+To run the image with user `neo4j` and password `test`:
+
+```bash
+docker run --publish=7474:7474 --publish=7687:7687 --env NEO4J_AUTH=neo4j/test neo4j
+```