{"id":13444557,"url":"https://github.com/tleyden/neurgo","last_synced_at":"2025-05-10T09:30:36.858Z","repository":{"id":8802562,"uuid":"10497244","full_name":"tleyden/neurgo","owner":"tleyden","description":"Neural Network toolkit in Go","archived":false,"fork":false,"pushed_at":"2016-03-09T18:24:39.000Z","size":365,"stargazers_count":36,"open_issues_count":2,"forks_count":3,"subscribers_count":4,"default_branch":"master","last_synced_at":"2024-10-28T08:41:41.518Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tleyden.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2013-06-05T07:24:52.000Z","updated_at":"2024-09-20T02:26:31.000Z","dependencies_parsed_at":"2022-09-02T11:51:10.128Z","dependency_job_id":null,"html_url":"https://github.com/tleyden/neurgo","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tleyden%2Fneurgo","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tleyden%2Fneurgo/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tleyden%2Fneurgo/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tleyden%2Fneurgo/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tleyden","download_url":"https://codeload.github.com/tleyden/neurgo/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224937283,"owners_count":17395121,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-07-31T04:00:30.109Z","updated_at":"2024-11-16T16:15:41.648Z","avatar_url":"https://github.com/tleyden.png","language":"Go","funding_links":[],"categories":["Neural Networks"],"sub_categories":["Vector Database"],"readme":"# Neurgo\n\n[![Build Status](https://drone.io/github.com/tleyden/neurgo/status.png)](https://drone.io/github.com/tleyden/neurgo/latest) [![GoDoc](https://godoc.org/github.com/tleyden/neurgo?status.png)](https://godoc.org/github.com/tleyden/neurgo)\n\nA library for constructing Neural Networks in [Go](http://golang.org/), where Neurons are goroutines that communicate with each other via channels.\n\n\n![architecture_diagram.png](http://cl.ly/image/0a1Y0e0B0P1m/Screen%20Shot%202013-10-09%20at%209.22.11%20PM.png)\n\n## What it can do\n\n* Feedforward networks\n* Recurrent networks\n* JSON Marshal/Unmarshal ([example json](https://drone.io/github.com/tleyden/neurgo/files/xnor.json))\n* Visualization network topology in SVG ([example svg](https://drone.io/github.com/tleyden/neurgo/files/xnor.svg))\n\n## Learning mechanism\n\nNeurgo does _not_ contain any code for learning/training.  \n\nThe idea is to have a separation of concerns such that the code that does the training will live in it's own repo.  Currently, there is only one training module:\n\n* [neurvolve](https://github.com/tleyden/neurvolve) - An evolution based trainer that is essentially a port of [DXNN2](https://github.com/CorticalComputer/DXNN2) (a Topology \u0026 Parameter Evolving Universal Learning Network in Erlang).\n\n## Roadmap\n\n* Training module for Backpropagation based learning (contributions welcome!)\n* Stress testing / benchmarks\n\n## Example applications\n\n* [Checkerlution - A Checkers Bot](https://github.com/tleyden/checkerlution)\n\n## Example code\n\nThe following code creates a neural net with [this topology](https://drone.io/github.com/tleyden/neurgo/files/xnor.svg).  It does not actually run the network (eg, feed inputs), so for a more complete example see `cortex_test.go`.\n\n```go\nsensor := \u0026Sensor{\n\tNodeId:       NewSensorId(\"sensor\", 0.0),\n\tVectorLength: 2,\n}\nsensor.Init()\nhiddenNeuron1 := \u0026Neuron{\n\tActivationFunction: EncodableSigmoid(),\n\tNodeId:             NewNeuronId(\"hidden-neuron1\", 0.25),\n\tBias:               -30,\n}\nhiddenNeuron1.Init()\nhiddenNeuron2 := \u0026Neuron{\n\tActivationFunction: EncodableSigmoid(),\n\tNodeId:             NewNeuronId(\"hidden-neuron2\", 0.25),\n\tBias:               10,\n}\nhiddenNeuron2.Init()\noutputNeuron := \u0026Neuron{\n\tActivationFunction: EncodableSigmoid(),\n\tNodeId:             NewNeuronId(\"output-neuron\", 0.35),\n\tBias:               -10,\n}\noutputNeuron.Init()\nactuator := \u0026Actuator{\n\tNodeId:       NewActuatorId(\"actuator\", 0.5),\n\tVectorLength: 1,\n}\nactuator.Init()\n\n// wire up connections\nsensor.ConnectOutbound(hiddenNeuron1)\nhiddenNeuron1.ConnectInboundWeighted(sensor, []float64{20, 20})\nsensor.ConnectOutbound(hiddenNeuron2)\nhiddenNeuron2.ConnectInboundWeighted(sensor, []float64{-20, -20})\nhiddenNeuron1.ConnectOutbound(outputNeuron)\noutputNeuron.ConnectInboundWeighted(hiddenNeuron1, []float64{20})\nhiddenNeuron2.ConnectOutbound(outputNeuron)\noutputNeuron.ConnectInboundWeighted(hiddenNeuron2, []float64{20})\noutputNeuron.ConnectOutbound(actuator)\nactuator.ConnectInbound(outputNeuron)\n\n// create cortex\nnodeId := NewCortexId(\"cortex\")\ncortex := \u0026Cortex{\n\tNodeId: nodeId,\n}\ncortex.SetSensors([]*Sensor{sensor})\ncortex.SetNeurons([]*Neuron{hiddenNeuron1, hiddenNeuron2, outputNeuron})\ncortex.SetActuators([]*Actuator{actuator})\n```\n\n## Getting Started\n\n* [Install Go](http://golang.org/doc/install)\n\n* Clone repository with `$ git clone git://github.com/tleyden/neurgo.git`\n\n* Run tests with `$ go test`\n\n* To write code that uses neurgo, your code will need `import \"github.com/tleyden/neurgo\"` as described in the [API documentation](http://godoc.org/github.com/tleyden/neurgo)\n\n## Documentation\n\n* This README file\n\n* [API documentation](http://godoc.org/github.com/tleyden/neurgo)\n\n\n## Libraries that build on Neurgo\n\n* [neurvolve](https://github.com/tleyden/neurvolve) builds on this library to support evolution-based learning.\n\n## Related Work\n\n[DXNN2](https://github.com/CorticalComputer/DXNN2) - Pure Erlang TPEULN (Topology \u0026 Parameter Evolving Universal Learning Network).  \n\n\n## Related Publications\n\n[Handbook of Neuroevolution Through Erlang](http://www.amazon.com/Handbook-Neuroevolution-Through-Erlang-Gene/dp/1461444624) _by Gene Sher_.\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftleyden%2Fneurgo","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftleyden%2Fneurgo","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftleyden%2Fneurgo/lists"}