Detecting Objects on Tello Drone

Let’s Get Started

In this tutorial, we will be using Node-RED. Designed and built by IBM, Node-RED is a free open source logic engine that allows programmers of any level to interconnect IoT, cloud-based systems, web services, databases, API’s and more!

Step 1: Get started by installing Node-RED locally

You can check out this tutorial on how to install Node-RED on your local computer , or follow the commands below :

sudo npm install -g --unsafe-perm node-red

This should install Node-RED, once installed run this command :

node-red

This should point you to a Node-RED URL http://127.0.0.1:1880/ that you can put in your browser to see the Node-RED editor

Note: In later steps, we will be adjusting the settings.js file that comes when you install Node-RED

Node-RED editor in your browser

Step 2: Install FFMpeg on your local computer

FFMpeg is a very powerful command-line tool that is used for performing various conversion operations on audio and video files. This tool will be the bread and butter for video streaming! It is free to use and is available for Windows, Linux and Mac operating systems

If you have a Mac, the best way to install it is by using homebrew.

Run: brew install ffmpeg to install FFmpeg locally to your computer.

This will be critical to see streams that are flowing in from your local computer!

Step 3: Install the FFMpeg Node in Node-RED

There are two ways you can install this node

Do npm install node-red-contrib-ffmpeg in the same directory that node-RED (from step 1) was installed
Open your node-RED browser and find node-red-contrib-ffmpeg in manage pallet

Open hamburger menu on right hand side > manage palette

Click install and you should see the ffmpeg node in your palette

Step 4: Train a Model

Follow this tutorial to train an object detection model .

Once you finish this tutorial you should end up with a model_web folder.

Step 5: Copy `model_web` into Node-RED and adjust settings

Once you have a model_web directory with your trained object model you will need to copy that into the directory that you installed Node-RED

We will then adjust the settings.js file in your node-RED director

Open settings.js and search for and uncomment out the following

httpAdminRoot: ‘/editor’,
httpStatic: ‘

I set the endpoint of httpAdminRoot to editor, so now when you run node-red command to see your node-RED editor you will now go to this url http://127.0.0.1:1880/editor

httpStatic will include the location of your model_web folder

Step 6: Configure your Tello Drone

Before you can use your Tello Drone you MUST activate it in the official Tello Drone app. Once your drone is activated, you can connect to it’s WiFi Network TELLO-XXXXXX and send it commands via UDP

Step 7: Import Flow into Node-RED

Start Node-RED locally and Import the following flow :

[{"id":"d7ec3041.ef1a7","type":"tab","label":"Flow 1","disabled":false,"info":""},{"id":"1e2f8497.0f0bb3","type":"http in","z":"d7ec3041.ef1a7","name":"","url":"/dashboard","method":"get","upload":false,"swaggerDoc":"","x":160,"y":260,"wires":[["e63ccee5.1b84a8"]]},{"id":"e63ccee5.1b84a8","type":"template","z":"d7ec3041.ef1a7","name":"","field":"payload","fieldType":"msg","format":"handlebars","syntax":"mustache","template":"<!DOCTYPE html>\n<html>\n<meta charset=\"utf-8\" />\n<body>\n<canvas id=\"video-canvas\" style=\"position: fixed; top: 0; left: 0;\"></canvas>\n<canvas id=\"draw-canvas\" style=\"position: fixed; top: 0; left: 0;\"></canvas>\n<script src=\"https://cdn.jsdelivr.net/npm/@cloud-annotations/object-detection@0.0.8\"></script>\n<script>\n\nconst renderPredictions = predictions => {\n    const canvas = document.getElementById('draw-canvas')\n    const videoCanvas = document.getElementById('video-canvas')\n    canvas.width = videoCanvas.width\n    canvas.height = videoCanvas.height\n    const ctx = canvas.getContext('2d')\n    ctx.clearRect(0, 0, ctx.canvas.width, ctx.canvas.height)\n    // Font options.\n    const font = '16px sans-serif'\n    ctx.font = font\n    ctx.textBaseline = 'top'\n    predictions.forEach(prediction => {\n        const x = prediction.bbox[0]\n        const y = prediction.bbox[1]\n        const width = prediction.bbox[2]\n        const height = prediction.bbox[3]\n        // Draw the bounding box.\n        ctx.strokeStyle = '#00FFFF'\n        ctx.lineWidth = 4\n        ctx.strokeRect(x, y, width, height)\n        // Draw the label background.\n        ctx.fillStyle = '#00FFFF'\n        const textWidth = ctx.measureText(prediction.class).width\n        const textHeight = parseInt(font, 10) // base 10\n        ctx.fillRect(x, y, textWidth + 4, textHeight + 4)\n    })\n\n    predictions.forEach(prediction => {\n        const x = prediction.bbox[0]\n        const y = prediction.bbox[1]\n        // Draw the text last to ensure it's on top.\n        ctx.fillStyle = '#000000'\n        ctx.fillText(prediction.class, x, y)\n    })\n}\n\nconst detect = async model => {\n    const videoCanvas = document.getElementById('video-canvas')\n    const predictions = await model.detect(videoCanvas)\n    renderPredictions(predictions)\n    requestAnimationFrame(() => {\n        detect(model)\n    })\n}\n\nconst init = async () => {\n    const canvas = document.getElementById('draw-canvas')\n    const videoCanvas = document.getElementById('video-canvas')\n    const url = `ws://${window.location.hostname}:${window.location.port}/stream`\n    new JSMpeg.Player(url, { canvas: videoCanvas })\n    \n    const model = await objectDetector.load('/model_web')\n    detect(model)\n}\n\nif (document.readyState === 'loading') {\n    document.addEventListener('DOMContentLoaded', init)\n} else {\n    setTimeout(init, 500)\n}\n</script>\n<script>\nvar JSMpeg={Player:null,VideoElement:null,BitBuffer:null,Source:{},Demuxer:{},Decoder:{},Renderer:{},AudioOutput:{},Now:function(){return window.performance?window.performance.now()/1e3:Date.now()/1e3},CreateVideoElements:function(){var elements=document.querySelectorAll(\".jsmpeg\");for(var i=0;i<elements.length;i++){new JSMpeg.VideoElement(elements[i])}},Fill:function(array,value){if(array.fill){array.fill(value)}else{for(var i=0;i<array.length;i++){array[i]=value}}},Base64ToArrayBuffer:function(base64){var binary=window.atob(base64);var length=binary.length;var bytes=new Uint8Array(length);for(var i=0;i<length;i++){bytes[i]=binary.charCodeAt(i)}return bytes.buffer},WASM_BINARY_INLINED:null};if(document.readyState===\"complete\"){JSMpeg.CreateVideoElements()}else{document.addEventListener(\"DOMContentLoaded\",JSMpeg.CreateVideoElements)}JSMpeg.VideoElement=function(){\"use strict\";var VideoElement=function(element){var url=element.dataset.url;if(!url){throw\"VideoElement has no `data-url` attribute\"}var addStyles=function(element,styles){for(var name in styles){element.style[name]=styles[name]}};this.container=element;addStyles(this.container,{display:\"inline-block\",position:\"relative\",minWidth:\"80px\",minHeight:\"80px\"});this.canvas=document.createElement(\"canvas\");this.canvas.width=960;this.canvas.height=540;addStyles(this.canvas,{display:\"block\",width:\"100%\"});this.container.appendChild(this.canvas);this.playButton=document.createElement(\"div\");this.playButton.innerHTML=VideoElement.PLAY_BUTTON;addStyles(this.playButton,{zIndex:2,position:\"absolute\",top:\"0\",bottom:\"0\",left:\"0\",right:\"0\",maxWidth:\"75px\",maxHeight:\"75px\",margin:\"auto\",opacity:\"0.7\",cursor:\"pointer\"});this.container.appendChild(this.playButton);var options={canvas:this.canvas};for(var option in element.dataset){try{options[option]=JSON.parse(element.dataset[option])}catch(err){options[option]=element.dataset[option]}}this.player=new JSMpeg.Player(url,options);element.playerInstance=this.player;if(options.poster&&!options.autoplay&&!this.player.options.streaming){options.decodeFirstFrame=false;this.poster=new Image;this.poster.src=options.poster;this.poster.addEventListener(\"load\",this.posterLoaded);addStyles(this.poster,{display:\"block\",zIndex:1,position:\"absolute\",top:0,left:0,bottom:0,right:0});this.container.appendChild(this.poster)}if(!this.player.options.streaming){this.container.addEventListener(\"click\",this.onClick.bind(this))}if(options.autoplay||this.player.options.streaming){this.playButton.style.display=\"none\"}if(this.player.audioOut&&!this.player.audioOut.unlocked){var unlockAudioElement=this.container;if(options.autoplay||this.player.options.streaming){this.unmuteButton=document.createElement(\"div\");this.unmuteButton.innerHTML=VideoElement.UNMUTE_BUTTON;addStyles(this.unmuteButton,{zIndex:2,position:\"absolute\",bottom:\"10px\",right:\"20px\",width:\"75px\",height:\"75px\",margin:\"auto\",opacity:\"0.7\",cursor:\"pointer\"});this.container.appendChild(this.unmuteButton);unlockAudioElement=this.unmuteButton}this.unlockAudioBound=this.onUnlockAudio.bind(this,unlockAudioElement);unlockAudioElement.addEventListener(\"touchstart\",this.unlockAudioBound,false);unlockAudioElement.addEventListener(\"click\",this.unlockAudioBound,true)}};VideoElement.prototype.onUnlockAudio=function(element,ev){if(this.unmuteButton){ev.preventDefault();ev.stopPropagation()}this.player.audioOut.unlock(function(){if(this.unmuteButton){this.unmuteButton.style.display=\"none\"}element.removeEventListener(\"touchstart\",this.unlockAudioBound);element.removeEventListener(\"click\",this.unlockAudioBound)}.bind(this))};VideoElement.prototype.onClick=function(ev){if(this.player.isPlaying){this.player.pause();this.playButton.style.display=\"block\"}else{this.player.play();this.playButton.style.display=\"none\";if(this.poster){this.poster.style.display=\"none\"}}};VideoElement.PLAY_BUTTON='<svg style=\"max-width: 75px; max-height: 75px;\" '+'viewBox=\"0 0 200 200\" alt=\"Play video\">'+'<circle cx=\"100\" cy=\"100\" r=\"90\" fill=\"none\" '+'stroke-width=\"15\" stroke=\"#fff\"/>'+'<polygon points=\"70, 55 70, 145 145, 100\" fill=\"#fff\"/>'+\"</svg>\";VideoElement.UNMUTE_BUTTON='<svg style=\"max-width: 75px; max-height: 75px;\" viewBox=\"0 0 75 75\">'+'<polygon class=\"audio-speaker\" stroke=\"none\" fill=\"#fff\" '+'points=\"39,13 22,28 6,28 6,47 21,47 39,62 39,13\"/>'+'<g stroke=\"#fff\" stroke-width=\"5\">'+'<path d=\"M 49,50 69,26\"/>'+'<path d=\"M 69,50 49,26\"/>'+\"</g>\"+\"</svg>\";return VideoElement}();JSMpeg.Player=function(){\"use strict\";var Player=function(url,options){this.options=options||{};if(options.source){this.source=new options.source(url,options);options.streaming=!!this.source.streaming}else if(url.match(/^wss?:\\/\\//)){this.source=new JSMpeg.Source.WebSocket(url,options);options.streaming=true}else if(options.progressive!==false){this.source=new JSMpeg.Source.AjaxProgressive(url,options);options.streaming=false}else{this.source=new JSMpeg.Source.Ajax(url,options);options.streaming=false}this.maxAudioLag=options.maxAudioLag||.25;this.loop=options.loop!==false;this.autoplay=!!options.autoplay||options.streaming;this.demuxer=new JSMpeg.Demuxer.TS(options);this.source.connect(this.demuxer);if(!options.disableWebAssembly&&JSMpeg.WASMModule.IsSupported()){this.wasmModule=new JSMpeg.WASMModule;options.wasmModule=this.wasmModule}if(options.video!==false){this.video=options.wasmModule?new JSMpeg.Decoder.MPEG1VideoWASM(options):new JSMpeg.Decoder.MPEG1Video(options);this.renderer=!options.disableGl&&JSMpeg.Renderer.WebGL.IsSupported()?new JSMpeg.Renderer.WebGL(options):new JSMpeg.Renderer.Canvas2D(options);this.demuxer.connect(JSMpeg.Demuxer.TS.STREAM.VIDEO_1,this.video);this.video.connect(this.renderer)}if(options.audio!==false&&JSMpeg.AudioOutput.WebAudio.IsSupported()){this.audio=options.wasmModule?new JSMpeg.Decoder.MP2AudioWASM(options):new JSMpeg.Decoder.MP2Audio(options);this.audioOut=new JSMpeg.AudioOutput.WebAudio(options);this.demuxer.connect(JSMpeg.Demuxer.TS.STREAM.AUDIO_1,this.audio);this.audio.connect(this.audioOut)}Object.defineProperty(this,\"currentTime\",{get:this.getCurrentTime,set:this.setCurrentTime});Object.defineProperty(this,\"volume\",{get:this.getVolume,set:this.setVolume});this.paused=true;this.unpauseOnShow=false;if(options.pauseWhenHidden!==false){document.addEventListener(\"visibilitychange\",this.showHide.bind(this))}if(this.wasmModule){if(JSMpeg.WASM_BINARY_INLINED){var wasm=JSMpeg.Base64ToArrayBuffer(JSMpeg.WASM_BINARY_INLINED);this.wasmModule.loadFromBuffer(wasm,this.startLoading.bind(this))}else{this.wasmModule.loadFromFile(\"jsmpeg.wasm\",this.startLoading.bind(this))}}else{this.startLoading()}};Player.prototype.startLoading=function(){this.source.start();if(this.autoplay){this.play()}};Player.prototype.showHide=function(ev){if(document.visibilityState===\"hidden\"){this.unpauseOnShow=this.wantsToPlay;this.pause()}else if(this.unpauseOnShow){this.play()}};Player.prototype.play=function(ev){if(this.animationId){return}this.animationId=requestAnimationFrame(this.update.bind(this));this.wantsToPlay=true;this.paused=false};Player.prototype.pause=function(ev){if(this.paused){return}cancelAnimationFrame(this.animationId);this.animationId=null;this.wantsToPlay=false;this.isPlaying=false;this.paused=true;if(this.audio&&this.audio.canPlay){this.audioOut.stop();this.seek(this.currentTime)}if(this.options.onPause){this.options.onPause(this)}};Player.prototype.getVolume=function(){return this.audioOut?this.audioOut.volume:0};Player.prototype.setVolume=function(volume){if(this.audioOut){this.audioOut.volume=volume}};Player.prototype.stop=function(ev){this.pause();this.seek(0);if(this.video&&this.options.decodeFirstFrame!==false){this.video.decode()}};Player.prototype.destroy=function(){this.pause();this.source.destroy();this.video&&this.video.destroy();this.renderer&&this.renderer.destroy();this.audio&&this.audio.destroy();this.audioOut&&this.audioOut.destroy()};Player.prototype.seek=function(time){var startOffset=this.audio&&this.audio.canPlay?this.audio.startTime:this.video.startTime;if(this.video){this.video.seek(time+startOffset)}if(this.audio){this.audio.seek(time+startOffset)}this.startTime=JSMpeg.Now()-time};Player.prototype.getCurrentTime=function(){return this.audio&&this.audio.canPlay?this.audio.currentTime-this.audio.startTime:this.video.currentTime-this.video.startTime};Player.prototype.setCurrentTime=function(time){this.seek(time)};Player.prototype.update=function(){this.animationId=requestAnimationFrame(this.update.bind(this));if(!this.source.established){if(this.renderer){this.renderer.renderProgress(this.source.progress)}return}if(!this.isPlaying){this.isPlaying=true;this.startTime=JSMpeg.Now()-this.currentTime;if(this.options.onPlay){this.options.onPlay(this)}}if(this.options.streaming){this.updateForStreaming()}else{this.updateForStaticFile()}};Player.prototype.updateForStreaming=function(){if(this.video){this.video.decode()}if(this.audio){var decoded=false;do{if(this.audioOut.enqueuedTime>this.maxAudioLag){this.audioOut.resetEnqueuedTime();this.audioOut.enabled=false}decoded=this.audio.decode()}while(decoded);this.audioOut.enabled=true}};Player.prototype.nextFrame=function(){if(this.source.established&&this.video){return this.video.decode()}return false};Player.prototype.updateForStaticFile=function(){var notEnoughData=false,headroom=0;if(this.audio&&this.audio.canPlay){while(!notEnoughData&&this.audio.decodedTime-this.audio.currentTime<.25){notEnoughData=!this.audio.decode()}if(this.video&&this.video.currentTime<this.audio.currentTime){notEnoughData=!this.video.decode()}headroom=this.demuxer.currentTime-this.audio.currentTime}else if(this.video){var targetTime=JSMpeg.Now()-this.startTime+this.video.startTime,lateTime=targetTime-this.video.currentTime,frameTime=1/this.video.frameRate;if(this.video&&lateTime>0){if(lateTime>frameTime*2){this.startTime+=lateTime}notEnoughData=!this.video.decode()}headroom=this.demuxer.currentTime-targetTime}this.source.resume(headroom);if(notEnoughData&&this.source.completed){if(this.loop){this.seek(0)}else{this.pause();if(this.options.onEnded){this.options.onEnded(this)}}}else if(notEnoughData&&this.options.onStalled){this.options.onStalled(this)}};return Player}();JSMpeg.BitBuffer=function(){\"use strict\";var BitBuffer=function(bufferOrLength,mode){if(typeof bufferOrLength===\"object\"){this.bytes=bufferOrLength instanceof Uint8Array?bufferOrLength:new Uint8Array(bufferOrLength);this.byteLength=this.bytes.length}else{this.bytes=new Uint8Array(bufferOrLength||1024*1024);this.byteLength=0}this.mode=mode||BitBuffer.MODE.EXPAND;this.index=0};BitBuffer.prototype.resize=function(size){var newBytes=new Uint8Array(size);if(this.byteLength!==0){this.byteLength=Math.min(this.byteLength,size);newBytes.set(this.bytes,0,this.byteLength)}this.bytes=newBytes;this.index=Math.min(this.index,this.byteLength<<3)};BitBuffer.prototype.evict=function(sizeNeeded){var bytePos=this.index>>3,available=this.bytes.length-this.byteLength;if(this.index===this.byteLength<<3||sizeNeeded>available+bytePos){this.byteLength=0;this.index=0;return}else if(bytePos===0){return}if(this.bytes.copyWithin){this.bytes.copyWithin(0,bytePos,this.byteLength)}else{this.bytes.set(this.bytes.subarray(bytePos,this.byteLength))}this.byteLength=this.byteLength-bytePos;this.index-=bytePos<<3;return};BitBuffer.prototype.write=function(buffers){var isArrayOfBuffers=typeof buffers[0]===\"object\",totalLength=0,available=this.bytes.length-this.byteLength;if(isArrayOfBuffers){var totalLength=0;for(var i=0;i<buffers.length;i++){totalLength+=buffers[i].byteLength}}else{totalLength=buffers.byteLength}if(totalLength>available){if(this.mode===BitBuffer.MODE.EXPAND){var newSize=Math.max(this.bytes.length*2,totalLength-available);this.resize(newSize)}else{this.evict(totalLength)}}if(isArrayOfBuffers){for(var i=0;i<buffers.length;i++){this.appendSingleBuffer(buffers[i])}}else{this.appendSingleBuffer(buffers)}return totalLength};BitBuffer.prototype.appendSingleBuffer=function(buffer){buffer=buffer instanceof Uint8Array?buffer:new Uint8Array(buffer);this.bytes.set(buffer,this.byteLength);this.byteLength+=buffer.length};BitBuffer.prototype.findNextStartCode=function(){for(var i=this.index+7>>3;i<this.byteLength;i++){if(this.bytes[i]==0&&this.bytes[i+1]==0&&this.bytes[i+2]==1){this.index=i+4<<3;return this.bytes[i+3]}}this.index=this.byteLength<<3;return-1};BitBuffer.prototype.findStartCode=function(code){var current=0;while(true){current=this.findNextStartCode();if(current===code||current===-1){return current}}return-1};BitBuffer.prototype.nextBytesAreStartCode=function(){var i=this.index+7>>3;return i>=this.byteLength||this.bytes[i]==0&&this.bytes[i+1]==0&&this.bytes[i+2]==1};BitBuffer.prototype.peek=function(count){var offset=this.index;var value=0;while(count){var currentByte=this.bytes[offset>>3],remaining=8-(offset&7),read=remaining<count?remaining:count,shift=remaining-read,mask=255>>8-read;value=value<<read|(currentByte&mask<<shift)>>shift;offset+=read;count-=read}return value};BitBuffer.prototype.read=function(count){var value=this.peek(count);this.index+=count;return value};BitBuffer.prototype.skip=function(count){return this.index+=count};BitBuffer.prototype.rewind=function(count){this.index=Math.max(this.index-count,0)};BitBuffer.prototype.has=function(count){return(this.byteLength<<3)-this.index>=count};BitBuffer.MODE={EVICT:1,EXPAND:2};return BitBuffer}();JSMpeg.Source.Ajax=function(){\"use strict\";var AjaxSource=function(url,options){this.url=url;this.destination=null;this.request=null;this.streaming=false;this.completed=false;this.established=false;this.progress=0;this.onEstablishedCallback=options.onSourceEstablished;this.onCompletedCallback=options.onSourceCompleted};AjaxSource.prototype.connect=function(destination){this.destination=destination};AjaxSource.prototype.start=function(){this.request=new XMLHttpRequest;this.request.onreadystatechange=function(){if(this.request.readyState===this.request.DONE&&this.request.status===200){this.onLoad(this.request.response)}}.bind(this);this.request.onprogress=this.onProgress.bind(this);this.request.open(\"GET\",this.url);this.request.responseType=\"arraybuffer\";this.request.send()};AjaxSource.prototype.resume=function(secondsHeadroom){};AjaxSource.prototype.destroy=function(){this.request.abort()};AjaxSource.prototype.onProgress=function(ev){this.progress=ev.loaded/ev.total};AjaxSource.prototype.onLoad=function(data){this.established=true;this.completed=true;this.progress=1;if(this.onEstablishedCallback){this.onEstablishedCallback(this)}if(this.onCompletedCallback){this.onCompletedCallback(this)}if(this.destination){this.destination.write(data)}};return AjaxSource}();JSMpeg.Source.Fetch=function(){\"use strict\";var FetchSource=function(url,options){this.url=url;this.destination=null;this.request=null;this.streaming=true;this.completed=false;this.established=false;this.progress=0;this.aborted=false;this.onEstablishedCallback=options.onSourceEstablished;this.onCompletedCallback=options.onSourceCompleted};FetchSource.prototype.connect=function(destination){this.destination=destination};FetchSource.prototype.start=function(){var params={method:\"GET\",headers:new Headers,cache:\"default\"};self.fetch(this.url,params).then(function(res){if(res.ok&&(res.status>=200&&res.status<=299)){this.progress=1;this.established=true;return this.pump(res.body.getReader())}else{}}.bind(this)).catch(function(err){throw err})};FetchSource.prototype.pump=function(reader){return reader.read().then(function(result){if(result.done){this.completed=true}else{if(this.aborted){return reader.cancel()}if(this.destination){this.destination.write(result.value.buffer)}return this.pump(reader)}}.bind(this)).catch(function(err){throw err})};FetchSource.prototype.resume=function(secondsHeadroom){};FetchSource.prototype.abort=function(){this.aborted=true};return FetchSource}();JSMpeg.Source.AjaxProgressive=function(){\"use strict\";var AjaxProgressiveSource=function(url,options){this.url=url;this.destination=null;this.request=null;this.streaming=false;this.completed=false;this.established=false;this.progress=0;this.fileSize=0;this.loadedSize=0;this.chunkSize=options.chunkSize||1024*1024;this.isLoading=false;this.loadStartTime=0;this.throttled=options.throttled!==false;this.aborted=false;this.onEstablishedCallback=options.onSourceEstablished;this.onCompletedCallback=options.onSourceCompleted};AjaxProgressiveSource.prototype.connect=function(destination){this.destination=destination};AjaxProgressiveSource.prototype.start=function(){this.request=new XMLHttpRequest;this.request.onreadystatechange=function(){if(this.request.readyState===this.request.DONE){this.fileSize=parseInt(this.request.getResponseHeader(\"Content-Length\"));this.loadNextChunk()}}.bind(this);this.request.onprogress=this.onProgress.bind(this);this.request.open(\"HEAD\",this.url);this.request.send()};AjaxProgressiveSource.prototype.resume=function(secondsHeadroom){if(this.isLoading||!this.throttled){return}var worstCaseLoadingTime=this.loadTime*8+2;if(worstCaseLoadingTime>secondsHeadroom){this.loadNextChunk()}};AjaxProgressiveSource.prototype.destroy=function(){this.request.abort();this.aborted=true};AjaxProgressiveSource.prototype.loadNextChunk=function(){var start=this.loadedSize,end=Math.min(this.loadedSize+this.chunkSize-1,this.fileSize-1);if(start>=this.fileSize||this.aborted){this.completed=true;if(this.onCompletedCallback){this.onCompletedCallback(this)}return}this.isLoading=true;this.loadStartTime=JSMpeg.Now();this.request=new XMLHttpRequest;this.request.onreadystatechange=function(){if(this.request.readyState===this.request.DONE&&this.request.status>=200&&this.request.status<300){this.onChunkLoad(this.request.response)}else if(this.request.readyState===this.request.DONE){if(this.loadFails++<3){this.loadNextChunk()}}}.bind(this);if(start===0){this.request.onprogress=this.onProgress.bind(this)}this.request.open(\"GET\",this.url+\"?\"+start+\"-\"+end);this.request.setRequestHeader(\"Range\",\"bytes=\"+start+\"-\"+end);this.request.responseType=\"arraybuffer\";this.request.send()};AjaxProgressiveSource.prototype.onProgress=function(ev){this.progress=ev.loaded/ev.total};AjaxProgressiveSource.prototype.onChunkLoad=function(data){var isFirstChunk=!this.established;this.established=true;this.progress=1;this.loadedSize+=data.byteLength;this.loadFails=0;this.isLoading=false;if(isFirstChunk&&this.onEstablishedCallback){this.onEstablishedCallback(this)}if(this.destination){this.destination.write(data)}this.loadTime=JSMpeg.Now()-this.loadStartTime;if(!this.throttled){this.loadNextChunk()}};return AjaxProgressiveSource}();JSMpeg.Source.WebSocket=function(){\"use strict\";var WSSource=function(url,options){this.url=url;this.options=options;this.socket=null;this.streaming=true;this.callbacks={connect:[],data:[]};this.destination=null;this.reconnectInterval=options.reconnectInterval!==undefined?options.reconnectInterval:5;this.shouldAttemptReconnect=!!this.reconnectInterval;this.completed=false;this.established=false;this.progress=0;this.reconnectTimeoutId=0;this.onEstablishedCallback=options.onSourceEstablished;this.onCompletedCallback=options.onSourceCompleted};WSSource.prototype.connect=function(destination){this.destination=destination};WSSource.prototype.destroy=function(){clearTimeout(this.reconnectTimeoutId);this.shouldAttemptReconnect=false;this.socket.close()};WSSource.prototype.start=function(){this.shouldAttemptReconnect=!!this.reconnectInterval;this.progress=0;this.established=false;this.socket=new WebSocket(this.url,this.options.protocols||null);this.socket.binaryType=\"arraybuffer\";this.socket.onmessage=this.onMessage.bind(this);this.socket.onopen=this.onOpen.bind(this);this.socket.onerror=this.onClose.bind(this);this.socket.onclose=this.onClose.bind(this)};WSSource.prototype.resume=function(secondsHeadroom){};WSSource.prototype.onOpen=function(){this.progress=1};WSSource.prototype.onClose=function(){if(this.shouldAttemptReconnect){clearTimeout(this.reconnectTimeoutId);this.reconnectTimeoutId=setTimeout(function(){this.start()}.bind(this),this.reconnectInterval*1e3)}};WSSource.prototype.onMessage=function(ev){var isFirstChunk=!this.established;this.established=true;if(isFirstChunk&&this.onEstablishedCallback){this.onEstablishedCallback(this)}if(this.destination){this.destination.write(ev.data)}};return WSSource}();JSMpeg.Demuxer.TS=function(){\"use strict\";var TS=function(options){this.bits=null;this.leftoverBytes=null;this.guessVideoFrameEnd=true;this.pidsToStreamIds={};this.pesPacketInfo={};this.startTime=0;this.currentTime=0};TS.prototype.connect=function(streamId,destination){this.pesPacketInfo[streamId]={destination:destination,currentLength:0,totalLength:0,pts:0,buffers:[]}};TS.prototype.write=function(buffer){if(this.leftoverBytes){var totalLength=buffer.byteLength+this.leftoverBytes.byteLength;this.bits=new JSMpeg.BitBuffer(totalLength);this.bits.write([this.leftoverBytes,buffer])}else{this.bits=new JSMpeg.BitBuffer(buffer)}while(this.bits.has(188<<3)&&this.parsePacket()){}var leftoverCount=this.bits.byteLength-(this.bits.index>>3);this.leftoverBytes=leftoverCount>0?this.bits.bytes.subarray(this.bits.index>>3):null};TS.prototype.parsePacket=function(){if(this.bits.read(8)!==71){if(!this.resync()){return false}}var end=(this.bits.index>>3)+187;var transportError=this.bits.read(1),payloadStart=this.bits.read(1),transportPriority=this.bits.read(1),pid=this.bits.read(13),transportScrambling=this.bits.read(2),adaptationField=this.bits.read(2),continuityCounter=this.bits.read(4);var streamId=this.pidsToStreamIds[pid];if(payloadStart&&streamId){var pi=this.pesPacketInfo[streamId];if(pi&&pi.currentLength){this.packetComplete(pi)}}if(adaptationField&1){if(adaptationField&2){var adaptationFieldLength=this.bits.read(8);this.bits.skip(adaptationFieldLength<<3)}if(payloadStart&&this.bits.nextBytesAreStartCode()){this.bits.skip(24);streamId=this.bits.read(8);this.pidsToStreamIds[pid]=streamId;var packetLength=this.bits.read(16);this.bits.skip(8);var ptsDtsFlag=this.bits.read(2);this.bits.skip(6);var headerLength=this.bits.read(8);var payloadBeginIndex=this.bits.index+(headerLength<<3);var pi=this.pesPacketInfo[streamId];if(pi){var pts=0;if(ptsDtsFlag&2){this.bits.skip(4);var p32_30=this.bits.read(3);this.bits.skip(1);var p29_15=this.bits.read(15);this.bits.skip(1);var p14_0=this.bits.read(15);this.bits.skip(1);pts=(p32_30*1073741824+p29_15*32768+p14_0)/9e4;this.currentTime=pts;if(this.startTime===-1){this.startTime=pts}}var payloadLength=packetLength?packetLength-headerLength-3:0;this.packetStart(pi,pts,payloadLength)}this.bits.index=payloadBeginIndex}if(streamId){var pi=this.pesPacketInfo[streamId];if(pi){var start=this.bits.index>>3;var complete=this.packetAddData(pi,start,end);var hasPadding=!payloadStart&&adaptationField&2;if(complete||this.guessVideoFrameEnd&&hasPadding){this.packetComplete(pi)}}}}this.bits.index=end<<3;return true};TS.prototype.resync=function(){if(!this.bits.has(188*6<<3)){return false}var byteIndex=this.bits.index>>3;for(var i=0;i<187;i++){if(this.bits.bytes[byteIndex+i]===71){var foundSync=true;for(var j=1;j<5;j++){if(this.bits.bytes[byteIndex+i+188*j]!==71){foundSync=false;break}}if(foundSync){this.bits.index=byteIndex+i+1<<3;return true}}}console.warn(\"JSMpeg: Possible garbage data. Skipping.\");this.bits.skip(187<<3);return false};TS.prototype.packetStart=function(pi,pts,payloadLength){pi.totalLength=payloadLength;pi.currentLength=0;pi.pts=pts};TS.prototype.packetAddData=function(pi,start,end){pi.buffers.push(this.bits.bytes.subarray(start,end));pi.currentLength+=end-start;var complete=pi.totalLength!==0&&pi.currentLength>=pi.totalLength;return complete};TS.prototype.packetComplete=function(pi){pi.destination.write(pi.pts,pi.buffers);pi.totalLength=0;pi.currentLength=0;pi.buffers=[]};TS.STREAM={PACK_HEADER:186,SYSTEM_HEADER:187,PROGRAM_MAP:188,PRIVATE_1:189,PADDING:190,PRIVATE_2:191,AUDIO_1:192,VIDEO_1:224,DIRECTORY:255};return TS}();JSMpeg.Decoder.Base=function(){\"use strict\";var BaseDecoder=function(options){this.destination=null;this.canPlay=false;this.collectTimestamps=!options.streaming;this.bytesWritten=0;this.timestamps=[];this.timestampIndex=0;this.startTime=0;this.decodedTime=0;Object.defineProperty(this,\"currentTime\",{get:this.getCurrentTime})};BaseDecoder.prototype.destroy=function(){};BaseDecoder.prototype.connect=function(destination){this.destination=destination};BaseDecoder.prototype.bufferGetIndex=function(){return this.bits.index};BaseDecoder.prototype.bufferSetIndex=function(index){this.bits.index=index};BaseDecoder.prototype.bufferWrite=function(buffers){return this.bits.write(buffers)};BaseDecoder.prototype.write=function(pts,buffers){if(this.collectTimestamps){if(this.timestamps.length===0){this.startTime=pts;this.decodedTime=pts}this.timestamps.push({index:this.bytesWritten<<3,time:pts})}this.bytesWritten+=this.bufferWrite(buffers);this.canPlay=true};BaseDecoder.prototype.seek=function(time){if(!this.collectTimestamps){return}this.timestampIndex=0;for(var i=0;i<this.timestamps.length;i++){if(this.timestamps[i].time>time){break}this.timestampIndex=i}var ts=this.timestamps[this.timestampIndex];if(ts){this.bufferSetIndex(ts.index);this.decodedTime=ts.time}else{this.bufferSetIndex(0);this.decodedTime=this.startTime}};BaseDecoder.prototype.decode=function(){this.advanceDecodedTime(0)};BaseDecoder.prototype.advanceDecodedTime=function(seconds){if(this.collectTimestamps){var newTimestampIndex=-1;var currentIndex=this.bufferGetIndex();for(var i=this.timestampIndex;i<this.timestamps.length;i++){if(this.timestamps[i].index>currentIndex){break}newTimestampIndex=i}if(newTimestampIndex!==-1&&newTimestampIndex!==this.timestampIndex){this.timestampIndex=newTimestampIndex;this.decodedTime=this.timestamps[this.timestampIndex].time;return}}this.decodedTime+=seconds};BaseDecoder.prototype.getCurrentTime=function(){return this.decodedTime};return BaseDecoder}();JSMpeg.Decoder.MPEG1Video=function(){\"use strict\";var MPEG1=function(options){JSMpeg.Decoder.Base.call(this,options);this.onDecodeCallback=options.onVideoDecode;var bufferSize=options.videoBufferSize||512*1024;var bufferMode=options.streaming?JSMpeg.BitBuffer.MODE.EVICT:JSMpeg.BitBuffer.MODE.EXPAND;this.bits=new JSMpeg.BitBuffer(bufferSize,bufferMode);this.customIntraQuantMatrix=new Uint8Array(64);this.customNonIntraQuantMatrix=new Uint8Array(64);this.blockData=new Int32Array(64);this.currentFrame=0;this.decodeFirstFrame=options.decodeFirstFrame!==false};MPEG1.prototype=Object.create(JSMpeg.Decoder.Base.prototype);MPEG1.prototype.constructor=MPEG1;MPEG1.prototype.write=function(pts,buffers){JSMpeg.Decoder.Base.prototype.write.call(this,pts,buffers);if(!this.hasSequenceHeader){if(this.bits.findStartCode(MPEG1.START.SEQUENCE)===-1){return false}this.decodeSequenceHeader();if(this.decodeFirstFrame){this.decode()}}};MPEG1.prototype.decode=function(){var startTime=JSMpeg.Now();if(!this.hasSequenceHeader){return false}if(this.bits.findStartCode(MPEG1.START.PICTURE)===-1){var bufferedBytes=this.bits.byteLength-(this.bits.index>>3);return false}this.decodePicture();this.advanceDecodedTime(1/this.frameRate);var elapsedTime=JSMpeg.Now()-startTime;if(this.onDecodeCallback){this.onDecodeCallback(this,elapsedTime)}return true};MPEG1.prototype.readHuffman=function(codeTable){var state=0;do{state=codeTable[state+this.bits.read(1)]}while(state>=0&&codeTable[state]!==0);return codeTable[state+2]};MPEG1.prototype.frameRate=30;MPEG1.prototype.decodeSequenceHeader=function(){var newWidth=this.bits.read(12),newHeight=this.bits.read(12);this.bits.skip(4);this.frameRate=MPEG1.PICTURE_RATE[this.bits.read(4)];this.bits.skip(18+1+10+1);if(newWidth!==this.width||newHeight!==this.height){this.width=newWidth;this.height=newHeight;this.initBuffers();if(this.destination){this.destination.resize(newWidth,newHeight)}}if(this.bits.read(1)){for(var i=0;i<64;i++){this.customIntraQuantMatrix[MPEG1.ZIG_ZAG[i]]=this.bits.read(8)}this.intraQuantMatrix=this.customIntraQuantMatrix}if(this.bits.read(1)){for(var i=0;i<64;i++){var idx=MPEG1.ZIG_ZAG[i];this.customNonIntraQuantMatrix[idx]=this.bits.read(8)}this.nonIntraQuantMatrix=this.customNonIntraQuantMatrix}this.hasSequenceHeader=true};MPEG1.prototype.initBuffers=function(){this.intraQuantMatrix=MPEG1.DEFAULT_INTRA_QUANT_MATRIX;this.nonIntraQuantMatrix=MPEG1.DEFAULT_NON_INTRA_QUANT_MATRIX;this.mbWidth=this.width+15>>4;this.mbHeight=this.height+15>>4;this.mbSize=this.mbWidth*this.mbHeight;this.codedWidth=this.mbWidth<<4;this.codedHeight=this.mbHeight<<4;this.codedSize=this.codedWidth*this.codedHeight;this.halfWidth=this.mbWidth<<3;this.halfHeight=this.mbHeight<<3;this.currentY=new Uint8ClampedArray(this.codedSize);this.currentY32=new Uint32Array(this.currentY.buffer);this.currentCr=new Uint8ClampedArray(this.codedSize>>2);this.currentCr32=new Uint32Array(this.currentCr.buffer);this.currentCb=new Uint8ClampedArray(this.codedSize>>2);this.currentCb32=new Uint32Array(this.currentCb.buffer);this.forwardY=new Uint8ClampedArray(this.codedSize);this.forwardY32=new Uint32Array(this.forwardY.buffer);this.forwardCr=new Uint8ClampedArray(this.codedSize>>2);this.forwardCr32=new Uint32Array(this.forwardCr.buffer);this.forwardCb=new Uint8ClampedArray(this.codedSize>>2);this.forwardCb32=new Uint32Array(this.forwardCb.buffer)};MPEG1.prototype.currentY=null;MPEG1.prototype.currentCr=null;MPEG1.prototype.currentCb=null;MPEG1.prototype.pictureType=0;MPEG1.prototype.forwardY=null;MPEG1.prototype.forwardCr=null;MPEG1.prototype.forwardCb=null;MPEG1.prototype.fullPelForward=false;MPEG1.prototype.forwardFCode=0;MPEG1.prototype.forwardRSize=0;MPEG1.prototype.forwardF=0;MPEG1.prototype.decodePicture=function(skipOutput){this.currentFrame++;this.bits.skip(10);this.pictureType=this.bits.read(3);this.bits.skip(16);if(this.pictureType<=0||this.pictureType>=MPEG1.PICTURE_TYPE.B){return}if(this.pictureType===MPEG1.PICTURE_TYPE.PREDICTIVE){this.fullPelForward=this.bits.read(1);this.forwardFCode=this.bits.read(3);if(this.forwardFCode===0){return}this.forwardRSize=this.forwardFCode-1;this.forwardF=1<<this.forwardRSize}var code=0;do{code=this.bits.findNextStartCode()}while(code===MPEG1.START.EXTENSION||code===MPEG1.START.USER_DATA);while(code>=MPEG1.START.SLICE_FIRST&&code<=MPEG1.START.SLICE_LAST){this.decodeSlice(code&255);code=this.bits.findNextStartCode()}if(code!==-1){this.bits.rewind(32)}if(this.destination){this.destination.render(this.currentY,this.currentCr,this.currentCb,true)}if(this.pictureType===MPEG1.PICTURE_TYPE.INTRA||this.pictureType===MPEG1.PICTURE_TYPE.PREDICTIVE){var tmpY=this.forwardY,tmpY32=this.forwardY32,tmpCr=this.forwardCr,tmpCr32=this.forwardCr32,tmpCb=this.forwardCb,tmpCb32=this.forwardCb32;this.forwardY=this.currentY;this.forwardY32=this.currentY32;this.forwardCr=this.currentCr;this.forwardCr32=this.currentCr32;this.forwardCb=this.currentCb;this.forwardCb32=this.currentCb32;this.currentY=tmpY;this.currentY32=tmpY32;this.currentCr=tmpCr;this.currentCr32=tmpCr32;this.currentCb=tmpCb;this.currentCb32=tmpCb32}};MPEG1.prototype.quantizerScale=0;MPEG1.prototype.sliceBegin=false;MPEG1.prototype.decodeSlice=function(slice){this.sliceBegin=true;this.macroblockAddress=(slice-1)*this.mbWidth-1;this.motionFwH=this.motionFwHPrev=0;this.motionFwV=this.motionFwVPrev=0;this.dcPredictorY=128;this.dcPredictorCr=128;this.dcPredictorCb=128;this.quantizerScale=this.bits.read(5);while(this.bits.read(1)){this.bits.skip(8)}do{this.decodeMacroblock()}while(!this.bits.nextBytesAreStartCode())};MPEG1.prototype.macroblockAddress=0;MPEG1.prototype.mbRow=0;MPEG1.prototype.mbCol=0;MPEG1.prototype.macroblockType=0;MPEG1.prototype.macroblockIntra=false;MPEG1.prototype.macroblockMotFw=false;MPEG1.prototype.motionFwH=0;MPEG1.prototype.motionFwV=0;MPEG1.prototype.motionFwHPrev=0;MPEG1.prototype.motionFwVPrev=0;MPEG1.prototype.decodeMacroblock=function(){var increment=0,t=this.readHuffman(MPEG1.MACROBLOCK_ADDRESS_INCREMENT);while(t===34){t=this.readHuffman(MPEG1.MACROBLOCK_ADDRESS_INCREMENT)}while(t===35){increment+=33;\nt=this.readHuffman(MPEG1.MACROBLOCK_ADDRESS_INCREMENT)}increment+=t;if(this.sliceBegin){this.sliceBegin=false;this.macroblockAddress+=increment}else{if(this.macroblockAddress+increment>=this.mbSize){return}if(increment>1){this.dcPredictorY=128;this.dcPredictorCr=128;this.dcPredictorCb=128;if(this.pictureType===MPEG1.PICTURE_TYPE.PREDICTIVE){this.motionFwH=this.motionFwHPrev=0;this.motionFwV=this.motionFwVPrev=0}}while(increment>1){this.macroblockAddress++;this.mbRow=this.macroblockAddress/this.mbWidth|0;this.mbCol=this.macroblockAddress%this.mbWidth;this.copyMacroblock(this.motionFwH,this.motionFwV,this.forwardY,this.forwardCr,this.forwardCb);increment--}this.macroblockAddress++}this.mbRow=this.macroblockAddress/this.mbWidth|0;this.mbCol=this.macroblockAddress%this.mbWidth;var mbTable=MPEG1.MACROBLOCK_TYPE[this.pictureType];this.macroblockType=this.readHuffman(mbTable);this.macroblockIntra=this.macroblockType&1;this.macroblockMotFw=this.macroblockType&8;if((this.macroblockType&16)!==0){this.quantizerScale=this.bits.read(5)}if(this.macroblockIntra){this.motionFwH=this.motionFwHPrev=0;this.motionFwV=this.motionFwVPrev=0}else{this.dcPredictorY=128;this.dcPredictorCr=128;this.dcPredictorCb=128;this.decodeMotionVectors();this.copyMacroblock(this.motionFwH,this.motionFwV,this.forwardY,this.forwardCr,this.forwardCb)}var cbp=(this.macroblockType&2)!==0?this.readHuffman(MPEG1.CODE_BLOCK_PATTERN):this.macroblockIntra?63:0;for(var block=0,mask=32;block<6;block++){if((cbp&mask)!==0){this.decodeBlock(block)}mask>>=1}};MPEG1.prototype.decodeMotionVectors=function(){var code,d,r=0;if(this.macroblockMotFw){code=this.readHuffman(MPEG1.MOTION);if(code!==0&&this.forwardF!==1){r=this.bits.read(this.forwardRSize);d=(Math.abs(code)-1<<this.forwardRSize)+r+1;if(code<0){d=-d}}else{d=code}this.motionFwHPrev+=d;if(this.motionFwHPrev>(this.forwardF<<4)-1){this.motionFwHPrev-=this.forwardF<<5}else if(this.motionFwHPrev<-this.forwardF<<4){this.motionFwHPrev+=this.forwardF<<5}this.motionFwH=this.motionFwHPrev;if(this.fullPelForward){this.motionFwH<<=1}code=this.readHuffman(MPEG1.MOTION);if(code!==0&&this.forwardF!==1){r=this.bits.read(this.forwardRSize);d=(Math.abs(code)-1<<this.forwardRSize)+r+1;if(code<0){d=-d}}else{d=code}this.motionFwVPrev+=d;if(this.motionFwVPrev>(this.forwardF<<4)-1){this.motionFwVPrev-=this.forwardF<<5}else if(this.motionFwVPrev<-this.forwardF<<4){this.motionFwVPrev+=this.forwardF<<5}this.motionFwV=this.motionFwVPrev;if(this.fullPelForward){this.motionFwV<<=1}}else if(this.pictureType===MPEG1.PICTURE_TYPE.PREDICTIVE){this.motionFwH=this.motionFwHPrev=0;this.motionFwV=this.motionFwVPrev=0}};MPEG1.prototype.copyMacroblock=function(motionH,motionV,sY,sCr,sCb){var width,scan,H,V,oddH,oddV,src,dest,last;var dY=this.currentY32,dCb=this.currentCb32,dCr=this.currentCr32;width=this.codedWidth;scan=width-16;H=motionH>>1;V=motionV>>1;oddH=(motionH&1)===1;oddV=(motionV&1)===1;src=((this.mbRow<<4)+V)*width+(this.mbCol<<4)+H;dest=this.mbRow*width+this.mbCol<<2;last=dest+(width<<2);var x,y1,y2,y;if(oddH){if(oddV){while(dest<last){y1=sY[src]+sY[src+width];src++;for(x=0;x<4;x++){y2=sY[src]+sY[src+width];src++;y=y1+y2+2>>2&255;y1=sY[src]+sY[src+width];src++;y|=y1+y2+2<<6&65280;y2=sY[src]+sY[src+width];src++;y|=y1+y2+2<<14&16711680;y1=sY[src]+sY[src+width];src++;y|=y1+y2+2<<22&4278190080;dY[dest++]=y}dest+=scan>>2;src+=scan-1}}else{while(dest<last){y1=sY[src++];for(x=0;x<4;x++){y2=sY[src++];y=y1+y2+1>>1&255;y1=sY[src++];y|=y1+y2+1<<7&65280;y2=sY[src++];y|=y1+y2+1<<15&16711680;y1=sY[src++];y|=y1+y2+1<<23&4278190080;dY[dest++]=y}dest+=scan>>2;src+=scan-1}}}else{if(oddV){while(dest<last){for(x=0;x<4;x++){y=sY[src]+sY[src+width]+1>>1&255;src++;y|=sY[src]+sY[src+width]+1<<7&65280;src++;y|=sY[src]+sY[src+width]+1<<15&16711680;src++;y|=sY[src]+sY[src+width]+1<<23&4278190080;src++;dY[dest++]=y}dest+=scan>>2;src+=scan}}else{while(dest<last){for(x=0;x<4;x++){y=sY[src];src++;y|=sY[src]<<8;src++;y|=sY[src]<<16;src++;y|=sY[src]<<24;src++;dY[dest++]=y}dest+=scan>>2;src+=scan}}}width=this.halfWidth;scan=width-8;H=motionH/2>>1;V=motionV/2>>1;oddH=(motionH/2&1)===1;oddV=(motionV/2&1)===1;src=((this.mbRow<<3)+V)*width+(this.mbCol<<3)+H;dest=this.mbRow*width+this.mbCol<<1;last=dest+(width<<1);var cr1,cr2,cr,cb1,cb2,cb;if(oddH){if(oddV){while(dest<last){cr1=sCr[src]+sCr[src+width];cb1=sCb[src]+sCb[src+width];src++;for(x=0;x<2;x++){cr2=sCr[src]+sCr[src+width];cb2=sCb[src]+sCb[src+width];src++;cr=cr1+cr2+2>>2&255;cb=cb1+cb2+2>>2&255;cr1=sCr[src]+sCr[src+width];cb1=sCb[src]+sCb[src+width];src++;cr|=cr1+cr2+2<<6&65280;cb|=cb1+cb2+2<<6&65280;cr2=sCr[src]+sCr[src+width];cb2=sCb[src]+sCb[src+width];src++;cr|=cr1+cr2+2<<14&16711680;cb|=cb1+cb2+2<<14&16711680;cr1=sCr[src]+sCr[src+width];cb1=sCb[src]+sCb[src+width];src++;cr|=cr1+cr2+2<<22&4278190080;cb|=cb1+cb2+2<<22&4278190080;dCr[dest]=cr;dCb[dest]=cb;dest++}dest+=scan>>2;src+=scan-1}}else{while(dest<last){cr1=sCr[src];cb1=sCb[src];src++;for(x=0;x<2;x++){cr2=sCr[src];cb2=sCb[src++];cr=cr1+cr2+1>>1&255;cb=cb1+cb2+1>>1&255;cr1=sCr[src];cb1=sCb[src++];cr|=cr1+cr2+1<<7&65280;cb|=cb1+cb2+1<<7&65280;cr2=sCr[src];cb2=sCb[src++];cr|=cr1+cr2+1<<15&16711680;cb|=cb1+cb2+1<<15&16711680;cr1=sCr[src];cb1=sCb[src++];cr|=cr1+cr2+1<<23&4278190080;cb|=cb1+cb2+1<<23&4278190080;dCr[dest]=cr;dCb[dest]=cb;dest++}dest+=scan>>2;src+=scan-1}}}else{if(oddV){while(dest<last){for(x=0;x<2;x++){cr=sCr[src]+sCr[src+width]+1>>1&255;cb=sCb[src]+sCb[src+width]+1>>1&255;src++;cr|=sCr[src]+sCr[src+width]+1<<7&65280;cb|=sCb[src]+sCb[src+width]+1<<7&65280;src++;cr|=sCr[src]+sCr[src+width]+1<<15&16711680;cb|=sCb[src]+sCb[src+width]+1<<15&16711680;src++;cr|=sCr[src]+sCr[src+width]+1<<23&4278190080;cb|=sCb[src]+sCb[src+width]+1<<23&4278190080;src++;dCr[dest]=cr;dCb[dest]=cb;dest++}dest+=scan>>2;src+=scan}}else{while(dest<last){for(x=0;x<2;x++){cr=sCr[src];cb=sCb[src];src++;cr|=sCr[src]<<8;cb|=sCb[src]<<8;src++;cr|=sCr[src]<<16;cb|=sCb[src]<<16;src++;cr|=sCr[src]<<24;cb|=sCb[src]<<24;src++;dCr[dest]=cr;dCb[dest]=cb;dest++}dest+=scan>>2;src+=scan}}}};MPEG1.prototype.dcPredictorY=0;MPEG1.prototype.dcPredictorCr=0;MPEG1.prototype.dcPredictorCb=0;MPEG1.prototype.blockData=null;MPEG1.prototype.decodeBlock=function(block){var n=0,quantMatrix;if(this.macroblockIntra){var predictor,dctSize;if(block<4){predictor=this.dcPredictorY;dctSize=this.readHuffman(MPEG1.DCT_DC_SIZE_LUMINANCE)}else{predictor=block===4?this.dcPredictorCr:this.dcPredictorCb;dctSize=this.readHuffman(MPEG1.DCT_DC_SIZE_CHROMINANCE)}if(dctSize>0){var differential=this.bits.read(dctSize);if((differential&1<<dctSize-1)!==0){this.blockData[0]=predictor+differential}else{this.blockData[0]=predictor+(-1<<dctSize|differential+1)}}else{this.blockData[0]=predictor}if(block<4){this.dcPredictorY=this.blockData[0]}else if(block===4){this.dcPredictorCr=this.blockData[0]}else{this.dcPredictorCb=this.blockData[0]}this.blockData[0]<<=3+5;quantMatrix=this.intraQuantMatrix;n=1}else{quantMatrix=this.nonIntraQuantMatrix}var level=0;while(true){var run=0,coeff=this.readHuffman(MPEG1.DCT_COEFF);if(coeff===1&&n>0&&this.bits.read(1)===0){break}if(coeff===65535){run=this.bits.read(6);level=this.bits.read(8);if(level===0){level=this.bits.read(8)}else if(level===128){level=this.bits.read(8)-256}else if(level>128){level=level-256}}else{run=coeff>>8;level=coeff&255;if(this.bits.read(1)){level=-level}}n+=run;var dezigZagged=MPEG1.ZIG_ZAG[n];n++;level<<=1;if(!this.macroblockIntra){level+=level<0?-1:1}level=level*this.quantizerScale*quantMatrix[dezigZagged]>>4;if((level&1)===0){level-=level>0?1:-1}if(level>2047){level=2047}else if(level<-2048){level=-2048}this.blockData[dezigZagged]=level*MPEG1.PREMULTIPLIER_MATRIX[dezigZagged]}var destArray,destIndex,scan;if(block<4){destArray=this.currentY;scan=this.codedWidth-8;destIndex=this.mbRow*this.codedWidth+this.mbCol<<4;if((block&1)!==0){destIndex+=8}if((block&2)!==0){destIndex+=this.codedWidth<<3}}else{destArray=block===4?this.currentCb:this.currentCr;scan=(this.codedWidth>>1)-8;destIndex=(this.mbRow*this.codedWidth<<2)+(this.mbCol<<3)}if(this.macroblockIntra){if(n===1){MPEG1.CopyValueToDestination(this.blockData[0]+128>>8,destArray,destIndex,scan);this.blockData[0]=0}else{MPEG1.IDCT(this.blockData);MPEG1.CopyBlockToDestination(this.blockData,destArray,destIndex,scan);JSMpeg.Fill(this.blockData,0)}}else{if(n===1){MPEG1.AddValueToDestination(this.blockData[0]+128>>8,destArray,destIndex,scan);this.blockData[0]=0}else{MPEG1.IDCT(this.blockData);MPEG1.AddBlockToDestination(this.blockData,destArray,destIndex,scan);JSMpeg.Fill(this.blockData,0)}}n=0};MPEG1.CopyBlockToDestination=function(block,dest,index,scan){for(var n=0;n<64;n+=8,index+=scan+8){dest[index+0]=block[n+0];dest[index+1]=block[n+1];dest[index+2]=block[n+2];dest[index+3]=block[n+3];dest[index+4]=block[n+4];dest[index+5]=block[n+5];dest[index+6]=block[n+6];dest[index+7]=block[n+7]}};MPEG1.AddBlockToDestination=function(block,dest,index,scan){for(var n=0;n<64;n+=8,index+=scan+8){dest[index+0]+=block[n+0];dest[index+1]+=block[n+1];dest[index+2]+=block[n+2];dest[index+3]+=block[n+3];dest[index+4]+=block[n+4];dest[index+5]+=block[n+5];dest[index+6]+=block[n+6];dest[index+7]+=block[n+7]}};MPEG1.CopyValueToDestination=function(value,dest,index,scan){for(var n=0;n<64;n+=8,index+=scan+8){dest[index+0]=value;dest[index+1]=value;dest[index+2]=value;dest[index+3]=value;dest[index+4]=value;dest[index+5]=value;dest[index+6]=value;dest[index+7]=value}};MPEG1.AddValueToDestination=function(value,dest,index,scan){for(var n=0;n<64;n+=8,index+=scan+8){dest[index+0]+=value;dest[index+1]+=value;dest[index+2]+=value;dest[index+3]+=value;dest[index+4]+=value;dest[index+5]+=value;dest[index+6]+=value;dest[index+7]+=value}};MPEG1.IDCT=function(block){var b1,b3,b4,b6,b7,tmp1,tmp2,m0,x0,x1,x2,x3,x4,y3,y4,y5,y6,y7;for(var i=0;i<8;++i){b1=block[4*8+i];b3=block[2*8+i]+block[6*8+i];b4=block[5*8+i]-block[3*8+i];tmp1=block[1*8+i]+block[7*8+i];tmp2=block[3*8+i]+block[5*8+i];b6=block[1*8+i]-block[7*8+i];b7=tmp1+tmp2;m0=block[0*8+i];x4=(b6*473-b4*196+128>>8)-b7;x0=x4-((tmp1-tmp2)*362+128>>8);x1=m0-b1;x2=((block[2*8+i]-block[6*8+i])*362+128>>8)-b3;x3=m0+b1;y3=x1+x2;y4=x3+b3;y5=x1-x2;y6=x3-b3;y7=-x0-(b4*473+b6*196+128>>8);block[0*8+i]=b7+y4;block[1*8+i]=x4+y3;block[2*8+i]=y5-x0;block[3*8+i]=y6-y7;block[4*8+i]=y6+y7;block[5*8+i]=x0+y5;block[6*8+i]=y3-x4;block[7*8+i]=y4-b7}for(var i=0;i<64;i+=8){b1=block[4+i];b3=block[2+i]+block[6+i];b4=block[5+i]-block[3+i];tmp1=block[1+i]+block[7+i];tmp2=block[3+i]+block[5+i];b6=block[1+i]-block[7+i];b7=tmp1+tmp2;m0=block[0+i];x4=(b6*473-b4*196+128>>8)-b7;x0=x4-((tmp1-tmp2)*362+128>>8);x1=m0-b1;x2=((block[2+i]-block[6+i])*362+128>>8)-b3;x3=m0+b1;y3=x1+x2;y4=x3+b3;y5=x1-x2;y6=x3-b3;y7=-x0-(b4*473+b6*196+128>>8);block[0+i]=b7+y4+128>>8;block[1+i]=x4+y3+128>>8;block[2+i]=y5-x0+128>>8;block[3+i]=y6-y7+128>>8;block[4+i]=y6+y7+128>>8;block[5+i]=x0+y5+128>>8;block[6+i]=y3-x4+128>>8;block[7+i]=y4-b7+128>>8}};MPEG1.PICTURE_RATE=[0,23.976,24,25,29.97,30,50,59.94,60,0,0,0,0,0,0,0];MPEG1.ZIG_ZAG=new Uint8Array([0,1,8,16,9,2,3,10,17,24,32,25,18,11,4,5,12,19,26,33,40,48,41,34,27,20,13,6,7,14,21,28,35,42,49,56,57,50,43,36,29,22,15,23,30,37,44,51,58,59,52,45,38,31,39,46,53,60,61,54,47,55,62,63]);MPEG1.DEFAULT_INTRA_QUANT_MATRIX=new Uint8Array([8,16,19,22,26,27,29,34,16,16,22,24,27,29,34,37,19,22,26,27,29,34,34,38,22,22,26,27,29,34,37,40,22,26,27,29,32,35,40,48,26,27,29,32,35,40,48,58,26,27,29,34,38,46,56,69,27,29,35,38,46,56,69,83]);MPEG1.DEFAULT_NON_INTRA_QUANT_MATRIX=new Uint8Array([16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]);MPEG1.PREMULTIPLIER_MATRIX=new Uint8Array([32,44,42,38,32,25,17,9,44,62,58,52,44,35,24,12,42,58,55,49,42,33,23,12,38,52,49,44,38,30,20,10,32,44,42,38,32,25,17,9,25,35,33,30,25,20,14,7,17,24,23,20,17,14,9,5,9,12,12,10,9,7,5,2]);MPEG1.MACROBLOCK_ADDRESS_INCREMENT=new Int16Array([1*3,2*3,0,3*3,4*3,0,0,0,1,5*3,6*3,0,7*3,8*3,0,9*3,10*3,0,11*3,12*3,0,0,0,3,0,0,2,13*3,14*3,0,15*3,16*3,0,0,0,5,0,0,4,17*3,18*3,0,19*3,20*3,0,0,0,7,0,0,6,21*3,22*3,0,23*3,24*3,0,25*3,26*3,0,27*3,28*3,0,-1,29*3,0,-1,30*3,0,31*3,32*3,0,33*3,34*3,0,35*3,36*3,0,37*3,38*3,0,0,0,9,0,0,8,39*3,40*3,0,41*3,42*3,0,43*3,44*3,0,45*3,46*3,0,0,0,15,0,0,14,0,0,13,0,0,12,0,0,11,0,0,10,47*3,-1,0,-1,48*3,0,49*3,50*3,0,51*3,52*3,0,53*3,54*3,0,55*3,56*3,0,57*3,58*3,0,59*3,60*3,0,61*3,-1,0,-1,62*3,0,63*3,64*3,0,65*3,66*3,0,67*3,68*3,0,69*3,70*3,0,71*3,72*3,0,73*3,74*3,0,0,0,21,0,0,20,0,0,19,0,0,18,0,0,17,0,0,16,0,0,35,0,0,34,0,0,33,0,0,32,0,0,31,0,0,30,0,0,29,0,0,28,0,0,27,0,0,26,0,0,25,0,0,24,0,0,23,0,0,22]);MPEG1.MACROBLOCK_TYPE_INTRA=new Int8Array([1*3,2*3,0,-1,3*3,0,0,0,1,0,0,17]);MPEG1.MACROBLOCK_TYPE_PREDICTIVE=new Int8Array([1*3,2*3,0,3*3,4*3,0,0,0,10,5*3,6*3,0,0,0,2,7*3,8*3,0,0,0,8,9*3,10*3,0,11*3,12*3,0,-1,13*3,0,0,0,18,0,0,26,0,0,1,0,0,17]);MPEG1.MACROBLOCK_TYPE_B=new Int8Array([1*3,2*3,0,3*3,5*3,0,4*3,6*3,0,8*3,7*3,0,0,0,12,9*3,10*3,0,0,0,14,13*3,14*3,0,12*3,11*3,0,0,0,4,0,0,6,18*3,16*3,0,15*3,17*3,0,0,0,8,0,0,10,-1,19*3,0,0,0,1,20*3,21*3,0,0,0,30,0,0,17,0,0,22,0,0,26]);MPEG1.MACROBLOCK_TYPE=[null,MPEG1.MACROBLOCK_TYPE_INTRA,MPEG1.MACROBLOCK_TYPE_PREDICTIVE,MPEG1.MACROBLOCK_TYPE_B];MPEG1.CODE_BLOCK_PATTERN=new Int16Array([2*3,1*3,0,3*3,6*3,0,4*3,5*3,0,8*3,11*3,0,12*3,13*3,0,9*3,7*3,0,10*3,14*3,0,20*3,19*3,0,18*3,16*3,0,23*3,17*3,0,27*3,25*3,0,21*3,28*3,0,15*3,22*3,0,24*3,26*3,0,0,0,60,35*3,40*3,0,44*3,48*3,0,38*3,36*3,0,42*3,47*3,0,29*3,31*3,0,39*3,32*3,0,0,0,32,45*3,46*3,0,33*3,41*3,0,43*3,34*3,0,0,0,4,30*3,37*3,0,0,0,8,0,0,16,0,0,44,50*3,56*3,0,0,0,28,0,0,52,0,0,62,61*3,59*3,0,52*3,60*3,0,0,0,1,55*3,54*3,0,0,0,61,0,0,56,57*3,58*3,0,0,0,2,0,0,40,51*3,62*3,0,0,0,48,64*3,63*3,0,49*3,53*3,0,0,0,20,0,0,12,80*3,83*3,0,0,0,63,77*3,75*3,0,65*3,73*3,0,84*3,66*3,0,0,0,24,0,0,36,0,0,3,69*3,87*3,0,81*3,79*3,0,68*3,71*3,0,70*3,78*3,0,67*3,76*3,0,72*3,74*3,0,86*3,85*3,0,88*3,82*3,0,-1,94*3,0,95*3,97*3,0,0,0,33,0,0,9,106*3,110*3,0,102*3,116*3,0,0,0,5,0,0,10,93*3,89*3,0,0,0,6,0,0,18,0,0,17,0,0,34,113*3,119*3,0,103*3,104*3,0,90*3,92*3,0,109*3,107*3,0,117*3,118*3,0,101*3,99*3,0,98*3,96*3,0,100*3,91*3,0,114*3,115*3,0,105*3,108*3,0,112*3,111*3,0,121*3,125*3,0,0,0,41,0,0,14,0,0,21,124*3,122*3,0,120*3,123*3,0,0,0,11,0,0,19,0,0,7,0,0,35,0,0,13,0,0,50,0,0,49,0,0,58,0,0,37,0,0,25,0,0,45,0,0,57,0,0,26,0,0,29,0,0,38,0,0,53,0,0,23,0,0,43,0,0,46,0,0,42,0,0,22,0,0,54,0,0,51,0,0,15,0,0,30,0,0,39,0,0,47,0,0,55,0,0,27,0,0,59,0,0,31]);MPEG1.MOTION=new Int16Array([1*3,2*3,0,4*3,3*3,0,0,0,0,6*3,5*3,0,8*3,7*3,0,0,0,-1,0,0,1,9*3,10*3,0,12*3,11*3,0,0,0,2,0,0,-2,14*3,15*3,0,16*3,13*3,0,20*3,18*3,0,0,0,3,0,0,-3,17*3,19*3,0,-1,23*3,0,27*3,25*3,0,26*3,21*3,0,24*3,22*3,0,32*3,28*3,0,29*3,31*3,0,-1,33*3,0,36*3,35*3,0,0,0,-4,30*3,34*3,0,0,0,4,0,0,-7,0,0,5,37*3,41*3,0,0,0,-5,0,0,7,38*3,40*3,0,42*3,39*3,0,0,0,-6,0,0,6,51*3,54*3,0,50*3,49*3,0,45*3,46*3,0,52*3,47*3,0,43*3,53*3,0,44*3,48*3,0,0,0,10,0,0,9,0,0,8,0,0,-8,57*3,66*3,0,0,0,-9,60*3,64*3,0,56*3,61*3,0,55*3,62*3,0,58*3,63*3,0,0,0,-10,59*3,65*3,0,0,0,12,0,0,16,0,0,13,0,0,14,0,0,11,0,0,15,0,0,-16,0,0,-12,0,0,-14,0,0,-15,0,0,-11,0,0,-13]);MPEG1.DCT_DC_SIZE_LUMINANCE=new Int8Array([2*3,1*3,0,6*3,5*3,0,3*3,4*3,0,0,0,1,0,0,2,9*3,8*3,0,7*3,10*3,0,0,0,0,12*3,11*3,0,0,0,4,0,0,3,13*3,14*3,0,0,0,5,0,0,6,16*3,15*3,0,17*3,-1,0,0,0,7,0,0,8]);MPEG1.DCT_DC_SIZE_CHROMINANCE=new Int8Array([2*3,1*3,0,4*3,3*3,0,6*3,5*3,0,8*3,7*3,0,0,0,2,0,0,1,0,0,0,10*3,9*3,0,0,0,3,12*3,11*3,0,0,0,4,14*3,13*3,0,0,0,5,16*3,15*3,0,0,0,6,17*3,-1,0,0,0,7,0,0,8]);MPEG1.DCT_COEFF=new Int32Array([1*3,2*3,0,4*3,3*3,0,0,0,1,7*3,8*3,0,6*3,5*3,0,13*3,9*3,0,11*3,10*3,0,14*3,12*3,0,0,0,257,20*3,22*3,0,18*3,21*3,0,16*3,19*3,0,0,0,513,17*3,15*3,0,0,0,2,0,0,3,27*3,25*3,0,29*3,31*3,0,24*3,26*3,0,32*3,30*3,0,0,0,1025,23*3,28*3,0,0,0,769,0,0,258,0,0,1793,0,0,65535,0,0,1537,37*3,36*3,0,0,0,1281,35*3,34*3,0,39*3,38*3,0,33*3,42*3,0,40*3,41*3,0,52*3,50*3,0,54*3,53*3,0,48*3,49*3,0,43*3,45*3,0,46*3,44*3,0,0,0,2049,0,0,4,0,0,514,0,0,2305,51*3,47*3,0,55*3,57*3,0,60*3,56*3,0,59*3,58*3,0,61*3,62*3,0,0,0,2561,0,0,3329,0,0,6,0,0,259,0,0,5,0,0,770,0,0,2817,0,0,3073,76*3,75*3,0,67*3,70*3,0,73*3,71*3,0,78*3,74*3,0,72*3,77*3,0,69*3,64*3,0,68*3,63*3,0,66*3,65*3,0,81*3,87*3,0,91*3,80*3,0,82*3,79*3,0,83*3,86*3,0,93*3,92*3,0,84*3,85*3,0,90*3,94*3,0,88*3,89*3,0,0,0,515,0,0,260,0,0,7,0,0,1026,0,0,1282,0,0,4097,0,0,3841,0,0,3585,105*3,107*3,0,111*3,114*3,0,104*3,97*3,0,125*3,119*3,0,96*3,98*3,0,-1,123*3,0,95*3,101*3,0,106*3,121*3,0,99*3,102*3,0,113*3,103*3,0,112*3,116*3,0,110*3,100*3,0,124*3,115*3,0,117*3,122*3,0,109*3,118*3,0,120*3,108*3,0,127*3,136*3,0,139*3,140*3,0,130*3,126*3,0,145*3,146*3,0,128*3,129*3,0,0,0,2050,132*3,134*3,0,155*3,154*3,0,0,0,8,137*3,133*3,0,143*3,144*3,0,151*3,138*3,0,142*3,141*3,0,0,0,10,0,0,9,0,0,11,0,0,5377,0,0,1538,0,0,771,0,0,5121,0,0,1794,0,0,4353,0,0,4609,0,0,4865,148*3,152*3,0,0,0,1027,153*3,150*3,0,0,0,261,131*3,135*3,0,0,0,516,149*3,147*3,0,172*3,173*3,0,162*3,158*3,0,170*3,161*3,0,168*3,166*3,0,157*3,179*3,0,169*3,167*3,0,174*3,171*3,0,178*3,177*3,0,156*3,159*3,0,164*3,165*3,0,183*3,182*3,0,175*3,176*3,0,0,0,263,0,0,2562,0,0,2306,0,0,5633,0,0,5889,0,0,6401,0,0,6145,0,0,1283,0,0,772,0,0,13,0,0,12,0,0,14,0,0,15,0,0,517,0,0,6657,0,0,262,180*3,181*3,0,160*3,163*3,0,196*3,199*3,0,0,0,27,203*3,185*3,0,202*3,201*3,0,0,0,19,0,0,22,197*3,207*3,0,0,0,18,191*3,192*3,0,188*3,190*3,0,0,0,20,184*3,194*3,0,0,0,21,186*3,193*3,0,0,0,23,204*3,198*3,0,0,0,25,0,0,24,200*3,205*3,0,0,0,31,0,0,30,0,0,28,0,0,29,0,0,26,0,0,17,0,0,16,189*3,206*3,0,187*3,195*3,0,218*3,211*3,0,0,0,37,215*3,216*3,0,0,0,36,210*3,212*3,0,0,0,34,213*3,209*3,0,221*3,222*3,0,219*3,208*3,0,217*3,214*3,0,223*3,220*3,0,0,0,35,0,0,267,0,0,40,0,0,268,0,0,266,0,0,32,0,0,264,0,0,265,0,0,38,0,0,269,0,0,270,0,0,33,0,0,39,0,0,7937,0,0,6913,0,0,7681,0,0,4098,0,0,7425,0,0,7169,0,0,271,0,0,274,0,0,273,0,0,272,0,0,1539,0,0,2818,0,0,3586,0,0,3330,0,0,3074,0,0,3842]);MPEG1.PICTURE_TYPE={INTRA:1,PREDICTIVE:2,B:3};MPEG1.START={SEQUENCE:179,SLICE_FIRST:1,SLICE_LAST:175,PICTURE:0,EXTENSION:181,USER_DATA:178};return MPEG1}();JSMpeg.Decoder.MPEG1VideoWASM=function(){\"use strict\";var MPEG1WASM=function(options){JSMpeg.Decoder.Base.call(this,options);this.onDecodeCallback=options.onVideoDecode;this.module=options.wasmModule;this.bufferSize=options.videoBufferSize||512*1024;this.bufferMode=options.streaming?JSMpeg.BitBuffer.MODE.EVICT:JSMpeg.BitBuffer.MODE.EXPAND;this.decodeFirstFrame=options.decodeFirstFrame!==false;this.hasSequenceHeader=false};MPEG1WASM.prototype=Object.create(JSMpeg.Decoder.Base.prototype);MPEG1WASM.prototype.constructor=MPEG1WASM;MPEG1WASM.prototype.initializeWasmDecoder=function(){if(!this.module.instance){console.warn(\"JSMpeg: WASM module not compiled yet\");return}this.instance=this.module.instance;this.functions=this.module.instance.exports;this.decoder=this.functions._mpeg1_decoder_create(this.bufferSize,this.bufferMode)};MPEG1WASM.prototype.destroy=function(){if(!this.decoder){return}this.functions._mpeg1_decoder_destroy(this.decoder)};MPEG1WASM.prototype.bufferGetIndex=function(){if(!this.decoder){return}return this.functions._mpeg1_decoder_get_index(this.decoder)};MPEG1WASM.prototype.bufferSetIndex=function(index){if(!this.decoder){return}this.functions._mpeg1_decoder_set_index(this.decoder,index)};MPEG1WASM.prototype.bufferWrite=function(buffers){if(!this.decoder){this.initializeWasmDecoder()}var totalLength=0;for(var i=0;i<buffers.length;i++){totalLength+=buffers[i].length}var ptr=this.functions._mpeg1_decoder_get_write_ptr(this.decoder,totalLength);for(var i=0;i<buffers.length;i++){this.instance.heapU8.set(buffers[i],ptr);ptr+=buffers[i].length}this.functions._mpeg1_decoder_did_write(this.decoder,totalLength);return totalLength};MPEG1WASM.prototype.write=function(pts,buffers){JSMpeg.Decoder.Base.prototype.write.call(this,pts,buffers);if(!this.hasSequenceHeader&&this.functions._mpeg1_decoder_has_sequence_header(this.decoder)){this.loadSequnceHeader()}};MPEG1WASM.prototype.loadSequnceHeader=function(){this.hasSequenceHeader=true;this.frameRate=this.functions._mpeg1_decoder_get_frame_rate(this.decoder);this.codedSize=this.functions._mpeg1_decoder_get_coded_size(this.decoder);if(this.destination){var w=this.functions._mpeg1_decoder_get_width(this.decoder);var h=this.functions._mpeg1_decoder_get_height(this.decoder);this.destination.resize(w,h)}if(this.decodeFirstFrame){this.decode()}};MPEG1WASM.prototype.decode=function(){var startTime=JSMpeg.Now();if(!this.decoder){return false}var didDecode=this.functions._mpeg1_decoder_decode(this.decoder);if(!didDecode){return false}if(this.destination){var ptrY=this.functions._mpeg1_decoder_get_y_ptr(this.decoder),ptrCr=this.functions._mpeg1_decoder_get_cr_ptr(this.decoder),ptrCb=this.functions._mpeg1_decoder_get_cb_ptr(this.decoder);var dy=this.instance.heapU8.subarray(ptrY,ptrY+this.codedSize);var dcr=this.instance.heapU8.subarray(ptrCr,ptrCr+(this.codedSize>>2));var dcb=this.instance.heapU8.subarray(ptrCb,ptrCb+(this.codedSize>>2));this.destination.render(dy,dcr,dcb,false)}this.advanceDecodedTime(1/this.frameRate);var elapsedTime=JSMpeg.Now()-startTime;if(this.onDecodeCallback){this.onDecodeCallback(this,elapsedTime)}return true};return MPEG1WASM}();JSMpeg.Decoder.MP2Audio=function(){\"use strict\";var MP2=function(options){JSMpeg.Decoder.Base.call(this,options);this.onDecodeCallback=options.onAudioDecode;var bufferSize=options.audioBufferSize||128*1024;var bufferMode=options.streaming?JSMpeg.BitBuffer.MODE.EVICT:JSMpeg.BitBuffer.MODE.EXPAND;this.bits=new JSMpeg.BitBuffer(bufferSize,bufferMode);this.left=new Float32Array(1152);this.right=new Float32Array(1152);this.sampleRate=44100;this.D=new Float32Array(1024);this.D.set(MP2.SYNTHESIS_WINDOW,0);this.D.set(MP2.SYNTHESIS_WINDOW,512);this.V=new Float32Array(1024);this.U=new Int32Array(32);this.VPos=0;this.allocation=[new Array(32),new Array(32)];this.scaleFactorInfo=[new Uint8Array(32),new Uint8Array(32)];this.scaleFactor=[new Array(32),new Array(32)];this.sample=[new Array(32),new Array(32)];for(var j=0;j<2;j++){for(var i=0;i<32;i++){this.scaleFactor[j][i]=[0,0,0];this.sample[j][i]=[0,0,0]}}};MP2.prototype=Object.create(JSMpeg.Decoder.Base.prototype);MP2.prototype.constructor=MP2;MP2.prototype.decode=function(){var startTime=JSMpeg.Now();var pos=this.bits.index>>3;if(pos>=this.bits.byteLength){return false}var decoded=this.decodeFrame(this.left,this.right);this.bits.index=pos+decoded<<3;if(!decoded){return false}if(this.destination){this.destination.play(this.sampleRate,this.left,this.right)}this.advanceDecodedTime(this.left.length/this.sampleRate);var elapsedTime=JSMpeg.Now()-startTime;if(this.onDecodeCallback){this.onDecodeCallback(this,elapsedTime)}return true};MP2.prototype.getCurrentTime=function(){var enqueuedTime=this.destination?this.destination.enqueuedTime:0;return this.decodedTime-enqueuedTime};MP2.prototype.decodeFrame=function(left,right){var sync=this.bits.read(11),version=this.bits.read(2),layer=this.bits.read(2),hasCRC=!this.bits.read(1);if(sync!==MP2.FRAME_SYNC||version!==MP2.VERSION.MPEG_1||layer!==MP2.LAYER.II){return 0}var bitrateIndex=this.bits.read(4)-1;if(bitrateIndex>13){return 0}var sampleRateIndex=this.bits.read(2);var sampleRate=MP2.SAMPLE_RATE[sampleRateIndex];if(sampleRateIndex===3){return 0}if(version===MP2.VERSION.MPEG_2){sampleRateIndex+=4;bitrateIndex+=14}var padding=this.bits.read(1),privat=this.bits.read(1),mode=this.bits.read(2);var bound=0;if(mode===MP2.MODE.JOINT_STEREO){bound=this.bits.read(2)+1<<2}else{this.bits.skip(2);bound=mode===MP2.MODE.MONO?0:32}this.bits.skip(4);if(hasCRC){this.bits.skip(16)}var bitrate=MP2.BIT_RATE[bitrateIndex],sampleRate=MP2.SAMPLE_RATE[sampleRateIndex],frameSize=144e3*bitrate/sampleRate+padding|0;var tab3=0;var sblimit=0;if(version===MP2.VERSION.MPEG_2){tab3=2;sblimit=30}else{var tab1=mode===MP2.MODE.MONO?0:1;var tab2=MP2.QUANT_LUT_STEP_1[tab1][bitrateIndex];tab3=MP2.QUANT_LUT_STEP_2[tab2][sampleRateIndex];sblimit=tab3&63;tab3>>=6}if(bound>sblimit){bound=sblimit}for(var sb=0;sb<bound;sb++){this.allocation[0][sb]=this.readAllocation(sb,tab3);this.allocation[1][sb]=this.readAllocation(sb,tab3)}for(var sb=bound;sb<sblimit;sb++){this.allocation[0][sb]=this.allocation[1][sb]=this.readAllocation(sb,tab3)}var channels=mode===MP2.MODE.MONO?1:2;for(var sb=0;sb<sblimit;sb++){for(ch=0;ch<channels;ch++){if(this.allocation[ch][sb]){this.scaleFactorInfo[ch][sb]=this.bits.read(2)}}if(mode===MP2.MODE.MONO){this.scaleFactorInfo[1][sb]=this.scaleFactorInfo[0][sb]}}for(var sb=0;sb<sblimit;sb++){for(var ch=0;ch<channels;ch++){if(this.allocation[ch][sb]){var sf=this.scaleFactor[ch][sb];switch(this.scaleFactorInfo[ch][sb]){case 0:sf[0]=this.bits.read(6);sf[1]=this.bits.read(6);sf[2]=this.bits.read(6);break;case 1:sf[0]=sf[1]=this.bits.read(6);sf[2]=this.bits.read(6);break;case 2:sf[0]=sf[1]=sf[2]=this.bits.read(6);break;case 3:sf[0]=this.bits.read(6);sf[1]=sf[2]=this.bits.read(6);break}}}if(mode===MP2.MODE.MONO){this.scaleFactor[1][sb][0]=this.scaleFactor[0][sb][0];this.scaleFactor[1][sb][1]=this.scaleFactor[0][sb][1];this.scaleFactor[1][sb][2]=this.scaleFactor[0][sb][2]}}var outPos=0;for(var part=0;part<3;part++){for(var granule=0;granule<4;granule++){for(var sb=0;sb<bound;sb++){this.readSamples(0,sb,part);this.readSamples(1,sb,part)}for(var sb=bound;sb<sblimit;sb++){this.readSamples(0,sb,part);this.sample[1][sb][0]=this.sample[0][sb][0];this.sample[1][sb][1]=this.sample[0][sb][1];this.sample[1][sb][2]=this.sample[0][sb][2]}for(var sb=sblimit;sb<32;sb++){this.sample[0][sb][0]=0;this.sample[0][sb][1]=0;this.sample[0][sb][2]=0;this.sample[1][sb][0]=0;this.sample[1][sb][1]=0;this.sample[1][sb][2]=0}for(var p=0;p<3;p++){this.VPos=this.VPos-64&1023;for(var ch=0;ch<2;ch++){MP2.MatrixTransform(this.sample[ch],p,this.V,this.VPos);JSMpeg.Fill(this.U,0);var dIndex=512-(this.VPos>>1);var vIndex=this.VPos%128>>1;while(vIndex<1024){for(var i=0;i<32;++i){this.U[i]+=this.D[dIndex++]*this.V[vIndex++]}vIndex+=128-32;dIndex+=64-32}vIndex=128-32+1024-vIndex;dIndex-=512-32;while(vIndex<1024){for(var i=0;i<32;++i){this.U[i]+=this.D[dIndex++]*this.V[vIndex++]}vIndex+=128-32;dIndex+=64-32}var outChannel=ch===0?left:right;for(var j=0;j<32;j++){outChannel[outPos+j]=this.U[j]/2147418112}}outPos+=32}}}this.sampleRate=sampleRate;return frameSize};MP2.prototype.readAllocation=function(sb,tab3){var tab4=MP2.QUANT_LUT_STEP_3[tab3][sb];var qtab=MP2.QUANT_LUT_STEP4[tab4&15][this.bits.read(tab4>>4)];return qtab?MP2.QUANT_TAB[qtab-1]:0};MP2.prototype.readSamples=function(ch,sb,part){var q=this.allocation[ch][sb],sf=this.scaleFactor[ch][sb][part],sample=this.sample[ch][sb],val=0;if(!q){sample[0]=sample[1]=sample[2]=0;return}if(sf===63){sf=0}else{var shift=sf/3|0;sf=MP2.SCALEFACTOR_BASE[sf%3]+(1<<shift>>1)>>shift}var adj=q.levels;if(q.group){val=this.bits.read(q.bits);sample[0]=val%adj;val=val/adj|0;sample[1]=val%adj;sample[2]=val/adj|0}else{sample[0]=this.bits.read(q.bits);sample[1]=this.bits.read(q.bits);sample[2]=this.bits.read(q.bits)}var scale=65536/(adj+1)|0;adj=(adj+1>>1)-1;val=(adj-sample[0])*scale;sample[0]=val*(sf>>12)+(val*(sf&4095)+2048>>12)>>12;val=(adj-sample[1])*scale;sample[1]=val*(sf>>12)+(val*(sf&4095)+2048>>12)>>12;val=(adj-sample[2])*scale;sample[2]=val*(sf>>12)+(val*(sf&4095)+2048>>12)>>12};MP2.MatrixTransform=function(s,ss,d,dp){var t01,t02,t03,t04,t05,t06,t07,t08,t09,t10,t11,t12,t13,t14,t15,t16,t17,t18,t19,t20,t21,t22,t23,t24,t25,t26,t27,t28,t29,t30,t31,t32,t33;t01=s[0][ss]+s[31][ss];t02=(s[0][ss]-s[31][ss])*.500602998235;t03=s[1][ss]+s[30][ss];t04=(s[1][ss]-s[30][ss])*.505470959898;t05=s[2][ss]+s[29][ss];t06=(s[2][ss]-s[29][ss])*.515447309923;t07=s[3][ss]+s[28][ss];t08=(s[3][ss]-s[28][ss])*.53104259109;t09=s[4][ss]+s[27][ss];t10=(s[4][ss]-s[27][ss])*.553103896034;t11=s[5][ss]+s[26][ss];t12=(s[5][ss]-s[26][ss])*.582934968206;t13=s[6][ss]+s[25][ss];t14=(s[6][ss]-s[25][ss])*.622504123036;t15=s[7][ss]+s[24][ss];t16=(s[7][ss]-s[24][ss])*.674808341455;t17=s[8][ss]+s[23][ss];t18=(s[8][ss]-s[23][ss])*.744536271002;t19=s[9][ss]+s[22][ss];t20=(s[9][ss]-s[22][ss])*.839349645416;t21=s[10][ss]+s[21][ss];t22=(s[10][ss]-s[21][ss])*.972568237862;t23=s[11][ss]+s[20][ss];t24=(s[11][ss]-s[20][ss])*1.16943993343;t25=s[12][ss]+s[19][ss];t26=(s[12][ss]-s[19][ss])*1.48416461631;t27=s[13][ss]+s[18][ss];t28=(s[13][ss]-s[18][ss])*2.05778100995;t29=s[14][ss]+s[17][ss];t30=(s[14][ss]-s[17][ss])*3.40760841847;t31=s[15][ss]+s[16][ss];t32=(s[15][ss]-s[16][ss])*10.1900081235;t33=t01+t31;t31=(t01-t31)*.502419286188;t01=t03+t29;t29=(t03-t29)*.52249861494;t03=t05+t27;t27=(t05-t27)*.566944034816;t05=t07+t25;t25=(t07-t25)*.64682178336;t07=t09+t23;t23=(t09-t23)*.788154623451;t09=t11+t21;t21=(t11-t21)*1.06067768599;t11=t13+t19;t19=(t13-t19)*1.72244709824;t13=t15+t17;t17=(t15-t17)*5.10114861869;t15=t33+t13;t13=(t33-t13)*.509795579104;t33=t01+t11;t01=(t01-t11)*.601344886935;t11=t03+t09;t09=(t03-t09)*.899976223136;t03=t05+t07;t07=(t05-t07)*2.56291544774;t05=t15+t03;t15=(t15-t03)*.541196100146;t03=t33+t11;t11=(t33-t11)*1.30656296488;t33=t05+t03;t05=(t05-t03)*.707106781187;t03=t15+t11;t15=(t15-t11)*.707106781187;t03+=t15;t11=t13+t07;t13=(t13-t07)*.541196100146;t07=t01+t09;t09=(t01-t09)*1.30656296488;t01=t11+t07;t07=(t11-t07)*.707106781187;t11=t13+t09;t13=(t13-t09)*.707106781187;t11+=t13;t01+=t11;t11+=t07;t07+=t13;t09=t31+t17;t31=(t31-t17)*.509795579104;t17=t29+t19;t29=(t29-t19)*.601344886935;t19=t27+t21;t21=(t27-t21)*.899976223136;t27=t25+t23;t23=(t25-t23)*2.56291544774;t25=t09+t27;t09=(t09-t27)*.541196100146;t27=t17+t19;t19=(t17-t19)*1.30656296488;t17=t25+t27;t27=(t25-t27)*.707106781187;t25=t09+t19;t19=(t09-t19)*.707106781187;t25+=t19;t09=t31+t23;t31=(t31-t23)*.541196100146;t23=t29+t21;t21=(t29-t21)*1.30656296488;t29=t09+t23;t23=(t09-t23)*.707106781187;t09=t31+t21;t31=(t31-t21)*.707106781187;t09+=t31;t29+=t09;t09+=t23;t23+=t31;t17+=t29;t29+=t25;t25+=t09;t09+=t27;t27+=t23;t23+=t19;t19+=t31;t21=t02+t32;t02=(t02-t32)*.502419286188;t32=t04+t30;t04=(t04-t30)*.52249861494;t30=t06+t28;t28=(t06-t28)*.566944034816;t06=t08+t26;t08=(t08-t26)*.64682178336;t26=t10+t24;t10=(t10-t24)*.788154623451;t24=t12+t22;t22=(t12-t22)*1.06067768599;t12=t14+t20;t20=(t14-t20)*1.72244709824;t14=t16+t18;t16=(t16-t18)*5.10114861869;t18=t21+t14;t14=(t21-t14)*.509795579104;t21=t32+t12;t32=(t32-t12)*.601344886935;t12=t30+t24;t24=(t30-t24)*.899976223136;t30=t06+t26;t26=(t06-t26)*2.56291544774;t06=t18+t30;t18=(t18-t30)*.541196100146;t30=t21+t12;t12=(t21-t12)*1.30656296488;t21=t06+t30;t30=(t06-t30)*.707106781187;t06=t18+t12;t12=(t18-t12)*.707106781187;t06+=t12;t18=t14+t26;t26=(t14-t26)*.541196100146;t14=t32+t24;t24=(t32-t24)*1.30656296488;t32=t18+t14;t14=(t18-t14)*.707106781187;t18=t26+t24;t24=(t26-t24)*.707106781187;t18+=t24;t32+=t18;t18+=t14;t26=t14+t24;t14=t02+t16;t02=(t02-t16)*.509795579104;t16=t04+t20;t04=(t04-t20)*.601344886935;t20=t28+t22;t22=(t28-t22)*.899976223136;t28=t08+t10;t10=(t08-t10)*2.56291544774;t08=t14+t28;t14=(t14-t28)*.541196100146;t28=t16+t20;t20=(t16-t20)*1.30656296488;t16=t08+t28;t28=(t08-t28)*.707106781187;t08=t14+t20;t20=(t14-t20)*.707106781187;t08+=t20;t14=t02+t10;t02=(t02-t10)*.541196100146;t10=t04+t22;t22=(t04-t22)*1.30656296488;t04=t14+t10;t10=(t14-t10)*.707106781187;t14=t02+t22;t02=(t02-t22)*.707106781187;t14+=t02;t04+=t14;t14+=t10;t10+=t02;t16+=t04;t04+=t08;t08+=t14;t14+=t28;t28+=t10;t10+=t20;t20+=t02;t21+=t16;t16+=t32;t32+=t04;t04+=t06;t06+=t08;t08+=t18;t18+=t14;t14+=t30;t30+=t28;t28+=t26;t26+=t10;t10+=t12;t12+=t20;t20+=t24;t24+=t02;d[dp+48]=-t33;d[dp+49]=d[dp+47]=-t21;d[dp+50]=d[dp+46]=-t17;d[dp+51]=d[dp+45]=-t16;d[dp+52]=d[dp+44]=-t01;d[dp+53]=d[dp+43]=-t32;d[dp+54]=d[dp+42]=-t29;d[dp+55]=d[dp+41]=-t04;d[dp+56]=d[dp+40]=-t03;d[dp+57]=d[dp+39]=-t06;d[dp+58]=d[dp+38]=-t25;d[dp+59]=d[dp+37]=-t08;d[dp+60]=d[dp+36]=-t11;d[dp+61]=d[dp+35]=-t18;d[dp+62]=d[dp+34]=-t09;d[dp+63]=d[dp+33]=-t14;d[dp+32]=-t05;d[dp+0]=t05;d[dp+31]=-t30;d[dp+1]=t30;d[dp+30]=-t27;d[dp+2]=t27;d[dp+29]=-t28;d[dp+3]=t28;d[dp+28]=-t07;d[dp+4]=t07;d[dp+27]=-t26;d[dp+5]=t26;d[dp+26]=-t23;d[dp+6]=t23;d[dp+25]=-t10;d[dp+7]=t10;\nd[dp+24]=-t15;d[dp+8]=t15;d[dp+23]=-t12;d[dp+9]=t12;d[dp+22]=-t19;d[dp+10]=t19;d[dp+21]=-t20;d[dp+11]=t20;d[dp+20]=-t13;d[dp+12]=t13;d[dp+19]=-t24;d[dp+13]=t24;d[dp+18]=-t31;d[dp+14]=t31;d[dp+17]=-t02;d[dp+15]=t02;d[dp+16]=0};MP2.FRAME_SYNC=2047;MP2.VERSION={MPEG_2_5:0,MPEG_2:2,MPEG_1:3};MP2.LAYER={III:1,II:2,I:3};MP2.MODE={STEREO:0,JOINT_STEREO:1,DUAL_CHANNEL:2,MONO:3};MP2.SAMPLE_RATE=new Uint16Array([44100,48e3,32e3,0,22050,24e3,16e3,0]);MP2.BIT_RATE=new Uint16Array([32,48,56,64,80,96,112,128,160,192,224,256,320,384,8,16,24,32,40,48,56,64,80,96,112,128,144,160]);MP2.SCALEFACTOR_BASE=new Uint32Array([33554432,26632170,21137968]);MP2.SYNTHESIS_WINDOW=new Float32Array([0,-.5,-.5,-.5,-.5,-.5,-.5,-1,-1,-1,-1,-1.5,-1.5,-2,-2,-2.5,-2.5,-3,-3.5,-3.5,-4,-4.5,-5,-5.5,-6.5,-7,-8,-8.5,-9.5,-10.5,-12,-13,-14.5,-15.5,-17.5,-19,-20.5,-22.5,-24.5,-26.5,-29,-31.5,-34,-36.5,-39.5,-42.5,-45.5,-48.5,-52,-55.5,-58.5,-62.5,-66,-69.5,-73.5,-77,-80.5,-84.5,-88,-91.5,-95,-98,-101,-104,106.5,109,111,112.5,113.5,114,114,113.5,112,110.5,107.5,104,100,94.5,88.5,81.5,73,63.5,53,41.5,28.5,14.5,-1,-18,-36,-55.5,-76.5,-98.5,-122,-147,-173.5,-200.5,-229.5,-259.5,-290.5,-322.5,-355.5,-389.5,-424,-459.5,-495.5,-532,-568.5,-605,-641.5,-678,-714,-749,-783.5,-817,-849,-879.5,-908.5,-935,-959.5,-981,-1000.5,-1016,-1028.5,-1037.5,-1042.5,-1043.5,-1040,-1031.5,1018.5,1e3,976,946.5,911,869.5,822,767.5,707,640,565.5,485,397,302.5,201,92.5,-22.5,-144,-272.5,-407,-547.5,-694,-846,-1003,-1165,-1331.5,-1502,-1675.5,-1852.5,-2031.5,-2212.5,-2394,-2576.5,-2758.5,-2939.5,-3118.5,-3294.5,-3467.5,-3635.5,-3798.5,-3955,-4104.5,-4245.5,-4377.5,-4499,-4609.5,-4708,-4792.5,-4863.5,-4919,-4958,-4979.5,-4983,-4967.5,-4931.5,-4875,-4796,-4694.5,-4569.5,-4420,-4246,-4046,-3820,-3567,3287,2979.5,2644,2280.5,1888,1467.5,1018.5,541,35,-499,-1061,-1650,-2266.5,-2909,-3577,-4270,-4987.5,-5727.5,-6490,-7274,-8077.5,-8899.5,-9739,-10594.5,-11464.5,-12347,-13241,-14144.5,-15056,-15973.5,-16895.5,-17820,-18744.5,-19668,-20588,-21503,-22410.5,-23308.5,-24195,-25068.5,-25926.5,-26767,-27589,-28389,-29166.5,-29919,-30644.5,-31342,-32009.5,-32645,-33247,-33814.5,-34346,-34839.5,-35295,-35710,-36084.5,-36417.5,-36707.5,-36954,-37156.5,-37315,-37428,-37496,37519,37496,37428,37315,37156.5,36954,36707.5,36417.5,36084.5,35710,35295,34839.5,34346,33814.5,33247,32645,32009.5,31342,30644.5,29919,29166.5,28389,27589,26767,25926.5,25068.5,24195,23308.5,22410.5,21503,20588,19668,18744.5,17820,16895.5,15973.5,15056,14144.5,13241,12347,11464.5,10594.5,9739,8899.5,8077.5,7274,6490,5727.5,4987.5,4270,3577,2909,2266.5,1650,1061,499,-35,-541,-1018.5,-1467.5,-1888,-2280.5,-2644,-2979.5,3287,3567,3820,4046,4246,4420,4569.5,4694.5,4796,4875,4931.5,4967.5,4983,4979.5,4958,4919,4863.5,4792.5,4708,4609.5,4499,4377.5,4245.5,4104.5,3955,3798.5,3635.5,3467.5,3294.5,3118.5,2939.5,2758.5,2576.5,2394,2212.5,2031.5,1852.5,1675.5,1502,1331.5,1165,1003,846,694,547.5,407,272.5,144,22.5,-92.5,-201,-302.5,-397,-485,-565.5,-640,-707,-767.5,-822,-869.5,-911,-946.5,-976,-1e3,1018.5,1031.5,1040,1043.5,1042.5,1037.5,1028.5,1016,1000.5,981,959.5,935,908.5,879.5,849,817,783.5,749,714,678,641.5,605,568.5,532,495.5,459.5,424,389.5,355.5,322.5,290.5,259.5,229.5,200.5,173.5,147,122,98.5,76.5,55.5,36,18,1,-14.5,-28.5,-41.5,-53,-63.5,-73,-81.5,-88.5,-94.5,-100,-104,-107.5,-110.5,-112,-113.5,-114,-114,-113.5,-112.5,-111,-109,106.5,104,101,98,95,91.5,88,84.5,80.5,77,73.5,69.5,66,62.5,58.5,55.5,52,48.5,45.5,42.5,39.5,36.5,34,31.5,29,26.5,24.5,22.5,20.5,19,17.5,15.5,14.5,13,12,10.5,9.5,8.5,8,7,6.5,5.5,5,4.5,4,3.5,3.5,3,2.5,2.5,2,2,1.5,1.5,1,1,1,1,.5,.5,.5,.5,.5,.5]);MP2.QUANT_LUT_STEP_1=[[0,0,1,1,1,2,2,2,2,2,2,2,2,2],[0,0,0,0,0,0,1,1,1,2,2,2,2,2]];MP2.QUANT_TAB={A:27|64,B:30|64,C:8,D:12};MP2.QUANT_LUT_STEP_2=[[MP2.QUANT_TAB.C,MP2.QUANT_TAB.C,MP2.QUANT_TAB.D],[MP2.QUANT_TAB.A,MP2.QUANT_TAB.A,MP2.QUANT_TAB.A],[MP2.QUANT_TAB.B,MP2.QUANT_TAB.A,MP2.QUANT_TAB.B]];MP2.QUANT_LUT_STEP_3=[[68,68,52,52,52,52,52,52,52,52,52,52],[67,67,67,66,66,66,66,66,66,66,66,49,49,49,49,49,49,49,49,49,49,49,49,32,32,32,32,32,32,32],[69,69,69,69,52,52,52,52,52,52,52,36,36,36,36,36,36,36,36,36,36,36,36,36,36,36,36,36,36,36]];MP2.QUANT_LUT_STEP4=[[0,1,2,17],[0,1,2,3,4,5,6,17],[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,17],[0,1,3,5,6,7,8,9,10,11,12,13,14,15,16,17],[0,1,2,4,5,6,7,8,9,10,11,12,13,14,15,17],[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]];MP2.QUANT_TAB=[{levels:3,group:1,bits:5},{levels:5,group:1,bits:7},{levels:7,group:0,bits:3},{levels:9,group:1,bits:10},{levels:15,group:0,bits:4},{levels:31,group:0,bits:5},{levels:63,group:0,bits:6},{levels:127,group:0,bits:7},{levels:255,group:0,bits:8},{levels:511,group:0,bits:9},{levels:1023,group:0,bits:10},{levels:2047,group:0,bits:11},{levels:4095,group:0,bits:12},{levels:8191,group:0,bits:13},{levels:16383,group:0,bits:14},{levels:32767,group:0,bits:15},{levels:65535,group:0,bits:16}];return MP2}();JSMpeg.Decoder.MP2AudioWASM=function(){\"use strict\";var MP2WASM=function(options){JSMpeg.Decoder.Base.call(this,options);this.onDecodeCallback=options.onAudioDecode;this.module=options.wasmModule;this.bufferSize=options.audioBufferSize||128*1024;this.bufferMode=options.streaming?JSMpeg.BitBuffer.MODE.EVICT:JSMpeg.BitBuffer.MODE.EXPAND;this.sampleRate=0};MP2WASM.prototype=Object.create(JSMpeg.Decoder.Base.prototype);MP2WASM.prototype.constructor=MP2WASM;MP2WASM.prototype.initializeWasmDecoder=function(){if(!this.module.instance){console.warn(\"JSMpeg: WASM module not compiled yet\");return}this.instance=this.module.instance;this.functions=this.module.instance.exports;this.decoder=this.functions._mp2_decoder_create(this.bufferSize,this.bufferMode)};MP2WASM.prototype.destroy=function(){if(!this.decoder){return}this.functions._mp2_decoder_destroy(this.decoder)};MP2WASM.prototype.bufferGetIndex=function(){if(!this.decoder){return}return this.functions._mp2_decoder_get_index(this.decoder)};MP2WASM.prototype.bufferSetIndex=function(index){if(!this.decoder){return}this.functions._mp2_decoder_set_index(this.decoder,index)};MP2WASM.prototype.bufferWrite=function(buffers){if(!this.decoder){this.initializeWasmDecoder()}var totalLength=0;for(var i=0;i<buffers.length;i++){totalLength+=buffers[i].length}var ptr=this.functions._mp2_decoder_get_write_ptr(this.decoder,totalLength);for(var i=0;i<buffers.length;i++){this.instance.heapU8.set(buffers[i],ptr);ptr+=buffers[i].length}this.functions._mp2_decoder_did_write(this.decoder,totalLength);return totalLength};MP2WASM.prototype.decode=function(){var startTime=JSMpeg.Now();if(!this.decoder){return false}var decodedBytes=this.functions._mp2_decoder_decode(this.decoder);if(decodedBytes===0){return false}if(!this.sampleRate){this.sampleRate=this.functions._mp2_decoder_get_sample_rate(this.decoder)}if(this.destination){var leftPtr=this.functions._mp2_decoder_get_left_channel_ptr(this.decoder),rightPtr=this.functions._mp2_decoder_get_right_channel_ptr(this.decoder);var leftOffset=leftPtr/Float32Array.BYTES_PER_ELEMENT,rightOffset=rightPtr/Float32Array.BYTES_PER_ELEMENT;var left=this.instance.heapF32.subarray(leftOffset,leftOffset+MP2WASM.SAMPLES_PER_FRAME),right=this.instance.heapF32.subarray(rightOffset,rightOffset+MP2WASM.SAMPLES_PER_FRAME);this.destination.play(this.sampleRate,left,right)}this.advanceDecodedTime(MP2WASM.SAMPLES_PER_FRAME/this.sampleRate);var elapsedTime=JSMpeg.Now()-startTime;if(this.onDecodeCallback){this.onDecodeCallback(this,elapsedTime)}return true};MP2WASM.prototype.getCurrentTime=function(){var enqueuedTime=this.destination?this.destination.enqueuedTime:0;return this.decodedTime-enqueuedTime};MP2WASM.SAMPLES_PER_FRAME=1152;return MP2WASM}();JSMpeg.Renderer.WebGL=function(){\"use strict\";var WebGLRenderer=function(options){this.canvas=options.canvas||document.createElement(\"canvas\");this.width=this.canvas.width;this.height=this.canvas.height;this.enabled=true;this.hasTextureData={};var contextCreateOptions={preserveDrawingBuffer:!!options.preserveDrawingBuffer,alpha:false,depth:false,stencil:false,antialias:false,premultipliedAlpha:false};this.gl=this.canvas.getContext(\"webgl\",contextCreateOptions)||this.canvas.getContext(\"experimental-webgl\",contextCreateOptions);if(!this.gl){throw new Error(\"Failed to get WebGL Context\")}var gl=this.gl;var vertexAttr=null;gl.pixelStorei(gl.UNPACK_PREMULTIPLY_ALPHA_WEBGL,false);this.vertexBuffer=gl.createBuffer();var vertexCoords=new Float32Array([0,0,0,1,1,0,1,1]);gl.bindBuffer(gl.ARRAY_BUFFER,this.vertexBuffer);gl.bufferData(gl.ARRAY_BUFFER,vertexCoords,gl.STATIC_DRAW);this.program=this.createProgram(WebGLRenderer.SHADER.VERTEX_IDENTITY,WebGLRenderer.SHADER.FRAGMENT_YCRCB_TO_RGBA);vertexAttr=gl.getAttribLocation(this.program,\"vertex\");gl.enableVertexAttribArray(vertexAttr);gl.vertexAttribPointer(vertexAttr,2,gl.FLOAT,false,0,0);this.textureY=this.createTexture(0,\"textureY\");this.textureCb=this.createTexture(1,\"textureCb\");this.textureCr=this.createTexture(2,\"textureCr\");this.loadingProgram=this.createProgram(WebGLRenderer.SHADER.VERTEX_IDENTITY,WebGLRenderer.SHADER.FRAGMENT_LOADING);vertexAttr=gl.getAttribLocation(this.loadingProgram,\"vertex\");gl.enableVertexAttribArray(vertexAttr);gl.vertexAttribPointer(vertexAttr,2,gl.FLOAT,false,0,0);this.shouldCreateUnclampedViews=!this.allowsClampedTextureData()};WebGLRenderer.prototype.destroy=function(){var gl=this.gl;gl.deleteTexture(this.textureY);gl.deleteTexture(this.textureCb);gl.deleteTexture(this.textureCr);gl.deleteProgram(this.program);gl.deleteProgram(this.loadingProgram);gl.deleteBuffer(this.vertexBuffer);gl.getExtension(\"WEBGL_lose_context\").loseContext();this.canvas.remove()};WebGLRenderer.prototype.resize=function(width,height){this.width=width|0;this.height=height|0;this.canvas.width=this.width;this.canvas.height=this.height;this.gl.useProgram(this.program);var codedWidth=this.width+15>>4<<4;this.gl.viewport(0,0,codedWidth,this.height)};WebGLRenderer.prototype.createTexture=function(index,name){var gl=this.gl;var texture=gl.createTexture();gl.bindTexture(gl.TEXTURE_2D,texture);gl.texParameteri(gl.TEXTURE_2D,gl.TEXTURE_MAG_FILTER,gl.LINEAR);gl.texParameteri(gl.TEXTURE_2D,gl.TEXTURE_MIN_FILTER,gl.LINEAR);gl.texParameteri(gl.TEXTURE_2D,gl.TEXTURE_WRAP_S,gl.CLAMP_TO_EDGE);gl.texParameteri(gl.TEXTURE_2D,gl.TEXTURE_WRAP_T,gl.CLAMP_TO_EDGE);gl.uniform1i(gl.getUniformLocation(this.program,name),index);return texture};WebGLRenderer.prototype.createProgram=function(vsh,fsh){var gl=this.gl;var program=gl.createProgram();gl.attachShader(program,this.compileShader(gl.VERTEX_SHADER,vsh));gl.attachShader(program,this.compileShader(gl.FRAGMENT_SHADER,fsh));gl.linkProgram(program);gl.useProgram(program);return program};WebGLRenderer.prototype.compileShader=function(type,source){var gl=this.gl;var shader=gl.createShader(type);gl.shaderSource(shader,source);gl.compileShader(shader);if(!gl.getShaderParameter(shader,gl.COMPILE_STATUS)){throw new Error(gl.getShaderInfoLog(shader))}return shader};WebGLRenderer.prototype.allowsClampedTextureData=function(){var gl=this.gl;var texture=gl.createTexture();gl.bindTexture(gl.TEXTURE_2D,texture);gl.texImage2D(gl.TEXTURE_2D,0,gl.LUMINANCE,1,1,0,gl.LUMINANCE,gl.UNSIGNED_BYTE,new Uint8ClampedArray([0]));return gl.getError()===0};WebGLRenderer.prototype.renderProgress=function(progress){var gl=this.gl;gl.useProgram(this.loadingProgram);var loc=gl.getUniformLocation(this.loadingProgram,\"progress\");gl.uniform1f(loc,progress);gl.drawArrays(gl.TRIANGLE_STRIP,0,4)};WebGLRenderer.prototype.render=function(y,cb,cr,isClampedArray){if(!this.enabled){return}var gl=this.gl;var w=this.width+15>>4<<4,h=this.height,w2=w>>1,h2=h>>1;if(isClampedArray&&this.shouldCreateUnclampedViews){y=new Uint8Array(y.buffer),cb=new Uint8Array(cb.buffer),cr=new Uint8Array(cr.buffer)}gl.useProgram(this.program);this.updateTexture(gl.TEXTURE0,this.textureY,w,h,y);this.updateTexture(gl.TEXTURE1,this.textureCb,w2,h2,cb);this.updateTexture(gl.TEXTURE2,this.textureCr,w2,h2,cr);gl.drawArrays(gl.TRIANGLE_STRIP,0,4)};WebGLRenderer.prototype.updateTexture=function(unit,texture,w,h,data){var gl=this.gl;gl.activeTexture(unit);gl.bindTexture(gl.TEXTURE_2D,texture);if(this.hasTextureData[unit]){gl.texSubImage2D(gl.TEXTURE_2D,0,0,0,w,h,gl.LUMINANCE,gl.UNSIGNED_BYTE,data)}else{this.hasTextureData[unit]=true;gl.texImage2D(gl.TEXTURE_2D,0,gl.LUMINANCE,w,h,0,gl.LUMINANCE,gl.UNSIGNED_BYTE,data)}};WebGLRenderer.IsSupported=function(){try{if(!window.WebGLRenderingContext){return false}var canvas=document.createElement(\"canvas\");return!!(canvas.getContext(\"webgl\")||canvas.getContext(\"experimental-webgl\"))}catch(err){return false}};WebGLRenderer.SHADER={FRAGMENT_YCRCB_TO_RGBA:[\"precision mediump float;\",\"uniform sampler2D textureY;\",\"uniform sampler2D textureCb;\",\"uniform sampler2D textureCr;\",\"varying vec2 texCoord;\",\"mat4 rec601 = mat4(\",\"1.16438,  0.00000,  1.59603, -0.87079,\",\"1.16438, -0.39176, -0.81297,  0.52959,\",\"1.16438,  2.01723,  0.00000, -1.08139,\",\"0, 0, 0, 1\",\");\",\"void main() {\",\"float y = texture2D(textureY, texCoord).r;\",\"float cb = texture2D(textureCb, texCoord).r;\",\"float cr = texture2D(textureCr, texCoord).r;\",\"gl_FragColor = vec4(y, cr, cb, 1.0) * rec601;\",\"}\"].join(\"\\n\"),FRAGMENT_LOADING:[\"precision mediump float;\",\"uniform float progress;\",\"varying vec2 texCoord;\",\"void main() {\",\"float c = ceil(progress-(1.0-texCoord.y));\",\"gl_FragColor = vec4(c,c,c,1);\",\"}\"].join(\"\\n\"),VERTEX_IDENTITY:[\"attribute vec2 vertex;\",\"varying vec2 texCoord;\",\"void main() {\",\"texCoord = vertex;\",\"gl_Position = vec4((vertex * 2.0 - 1.0) * vec2(1, -1), 0.0, 1.0);\",\"}\"].join(\"\\n\")};return WebGLRenderer}();JSMpeg.Renderer.Canvas2D=function(){\"use strict\";var CanvasRenderer=function(options){this.canvas=options.canvas||document.createElement(\"canvas\");this.width=this.canvas.width;this.height=this.canvas.height;this.enabled=true;this.context=this.canvas.getContext(\"2d\")};CanvasRenderer.prototype.destroy=function(){};CanvasRenderer.prototype.resize=function(width,height){this.width=width|0;this.height=height|0;this.canvas.width=this.width;this.canvas.height=this.height;this.imageData=this.context.getImageData(0,0,this.width,this.height);JSMpeg.Fill(this.imageData.data,255)};CanvasRenderer.prototype.renderProgress=function(progress){var w=this.canvas.width,h=this.canvas.height,ctx=this.context;ctx.fillStyle=\"#222\";ctx.fillRect(0,0,w,h);ctx.fillStyle=\"#fff\";ctx.fillRect(0,h-h*progress,w,h*progress)};CanvasRenderer.prototype.render=function(y,cb,cr){this.YCbCrToRGBA(y,cb,cr,this.imageData.data);this.context.putImageData(this.imageData,0,0)};CanvasRenderer.prototype.YCbCrToRGBA=function(y,cb,cr,rgba){if(!this.enabled){return}var w=this.width+15>>4<<4,w2=w>>1;var yIndex1=0,yIndex2=w,yNext2Lines=w+(w-this.width);var cIndex=0,cNextLine=w2-(this.width>>1);var rgbaIndex1=0,rgbaIndex2=this.width*4,rgbaNext2Lines=this.width*4;var cols=this.width>>1,rows=this.height>>1;var ccb,ccr,r,g,b;for(var row=0;row<rows;row++){for(var col=0;col<cols;col++){ccb=cb[cIndex];ccr=cr[cIndex];cIndex++;r=ccb+(ccb*103>>8)-179;g=(ccr*88>>8)-44+(ccb*183>>8)-91;b=ccr+(ccr*198>>8)-227;var y1=y[yIndex1++];var y2=y[yIndex1++];rgba[rgbaIndex1]=y1+r;rgba[rgbaIndex1+1]=y1-g;rgba[rgbaIndex1+2]=y1+b;rgba[rgbaIndex1+4]=y2+r;rgba[rgbaIndex1+5]=y2-g;rgba[rgbaIndex1+6]=y2+b;rgbaIndex1+=8;var y3=y[yIndex2++];var y4=y[yIndex2++];rgba[rgbaIndex2]=y3+r;rgba[rgbaIndex2+1]=y3-g;rgba[rgbaIndex2+2]=y3+b;rgba[rgbaIndex2+4]=y4+r;rgba[rgbaIndex2+5]=y4-g;rgba[rgbaIndex2+6]=y4+b;rgbaIndex2+=8}yIndex1+=yNext2Lines;yIndex2+=yNext2Lines;rgbaIndex1+=rgbaNext2Lines;rgbaIndex2+=rgbaNext2Lines;cIndex+=cNextLine}};return CanvasRenderer}();JSMpeg.AudioOutput.WebAudio=function(){\"use strict\";var WebAudioOut=function(options){this.context=WebAudioOut.CachedContext=WebAudioOut.CachedContext||new(window.AudioContext||window.webkitAudioContext);this.gain=this.context.createGain();this.destination=this.gain;this.gain.connect(this.context.destination);this.context._connections=(this.context._connections||0)+1;this.startTime=0;this.buffer=null;this.wallclockStartTime=0;this.volume=1;this.enabled=true;this.unlocked=!WebAudioOut.NeedsUnlocking();Object.defineProperty(this,\"enqueuedTime\",{get:this.getEnqueuedTime})};WebAudioOut.prototype.destroy=function(){this.gain.disconnect();this.context._connections--;if(this.context._connections===0){this.context.close();WebAudioOut.CachedContext=null}};WebAudioOut.prototype.play=function(sampleRate,left,right){if(!this.enabled){return}if(!this.unlocked){var ts=JSMpeg.Now();if(this.wallclockStartTime<ts){this.wallclockStartTime=ts}this.wallclockStartTime+=left.length/sampleRate;return}this.gain.gain.value=this.volume;var buffer=this.context.createBuffer(2,left.length,sampleRate);buffer.getChannelData(0).set(left);buffer.getChannelData(1).set(right);var source=this.context.createBufferSource();source.buffer=buffer;source.connect(this.destination);var now=this.context.currentTime;var duration=buffer.duration;if(this.startTime<now){this.startTime=now;this.wallclockStartTime=JSMpeg.Now()}source.start(this.startTime);this.startTime+=duration;this.wallclockStartTime+=duration};WebAudioOut.prototype.stop=function(){this.gain.gain.value=0};WebAudioOut.prototype.getEnqueuedTime=function(){return Math.max(this.wallclockStartTime-JSMpeg.Now(),0)};WebAudioOut.prototype.resetEnqueuedTime=function(){this.startTime=this.context.currentTime;this.wallclockStartTime=JSMpeg.Now()};WebAudioOut.prototype.unlock=function(callback){if(this.unlocked){if(callback){callback()}return}this.unlockCallback=callback;var buffer=this.context.createBuffer(1,1,22050);var source=this.context.createBufferSource();source.buffer=buffer;source.connect(this.destination);source.start(0);setTimeout(this.checkIfUnlocked.bind(this,source,0),0)};WebAudioOut.prototype.checkIfUnlocked=function(source,attempt){if(source.playbackState===source.PLAYING_STATE||source.playbackState===source.FINISHED_STATE){this.unlocked=true;if(this.unlockCallback){this.unlockCallback();this.unlockCallback=null}}else if(attempt<10){setTimeout(this.checkIfUnlocked.bind(this,source,attempt+1),100)}};WebAudioOut.NeedsUnlocking=function(){return/iPhone|iPad|iPod/i.test(navigator.userAgent)};WebAudioOut.IsSupported=function(){return window.AudioContext||window.webkitAudioContext};WebAudioOut.CachedContext=null;return WebAudioOut}();JSMpeg.WASMModule=function(){\"use strict\";var WASM=function(){this.stackSize=5*1024*1024;this.pageSize=64*1024;this.onInitCallback=null};WASM.prototype.write=function(buffer){this.loadFromBuffer(buffer,this.onInitCallback)};WASM.prototype.loadFromFile=function(url,callback){this.onInitCallback=callback;var ajax=new JSMpeg.Source.Ajax(url);ajax.connect(this);ajax.start()};WASM.prototype.loadFromBuffer=function(buffer,callback){this.moduleInfo=this.readDylinkSection(buffer);if(!this.moduleInfo){this.callback&&this.callback(null);return}this.memory=new WebAssembly.Memory({initial:256});var env={memory:this.memory,memoryBase:0,__memory_base:0,table:new WebAssembly.Table({initial:this.moduleInfo.tableSize,element:\"anyfunc\"}),tableBase:0,__table_base:0,abort:this.c_abort.bind(this),___assert_fail:this.c_assertFail.bind(this),_sbrk:this.c_sbrk.bind(this)};this.brk=this.align(this.moduleInfo.memorySize+this.stackSize);WebAssembly.instantiate(buffer,{env:env}).then(function(results){this.instance=results.instance;if(this.instance.exports.__post_instantiate){this.instance.exports.__post_instantiate()}this.createHeapViews();callback&&callback(this)}.bind(this))};WASM.prototype.createHeapViews=function(){this.instance.heapU8=new Uint8Array(this.memory.buffer);this.instance.heapU32=new Uint32Array(this.memory.buffer);this.instance.heapF32=new Float32Array(this.memory.buffer)};WASM.prototype.align=function(addr){var a=Math.pow(2,this.moduleInfo.memoryAlignment);return Math.ceil(addr/a)*a};WASM.prototype.c_sbrk=function(size){var previousBrk=this.brk;this.brk+=size;if(this.brk>this.memory.buffer.byteLength){var bytesNeeded=this.brk-this.memory.buffer.byteLength;var pagesNeeded=Math.ceil(bytesNeeded/this.pageSize);this.memory.grow(pagesNeeded);this.createHeapViews()}return previousBrk};WASM.prototype.c_abort=function(size){console.warn(\"JSMPeg: WASM abort\",arguments)};WASM.prototype.c_assertFail=function(size){console.warn(\"JSMPeg: WASM ___assert_fail\",arguments)};WASM.prototype.readDylinkSection=function(buffer){var bytes=new Uint8Array(buffer);var next=0;var readVarUint=function(){var ret=0;var mul=1;while(1){var byte=bytes[next++];ret+=(byte&127)*mul;mul*=128;if(!(byte&128)){return ret}}};var matchNextBytes=function(expected){for(var i=0;i<expected.length;i++){var b=typeof expected[i]===\"string\"?expected[i].charCodeAt(0):expected[i];if(bytes[next++]!==b){return false}}return true};if(!matchNextBytes([0,\"a\",\"s\",\"m\"])){console.warn(\"JSMpeg: WASM header not found\");return null}var next=9;var sectionSize=readVarUint();if(!matchNextBytes([6,\"d\",\"y\",\"l\",\"i\",\"n\",\"k\"])){console.warn(\"JSMpeg: No dylink section found in WASM\");return null}return{memorySize:readVarUint(),memoryAlignment:readVarUint(),tableSize:readVarUint(),tableAlignment:readVarUint()}};WASM.IsSupported=function(){return!!window.WebAssembly};return WASM}();JSMpeg.WASM_BINARY_INLINED=\"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\"; \n</script>\n</body>\n</html>","output":"str","x":360,"y":260,"wires":[["8f61e3cb.035c9"]]},{"id":"8f61e3cb.035c9","type":"http response","z":"d7ec3041.ef1a7","name":"","statusCode":"","headers":{},"x":530,"y":260,"wires":[]},{"id":"16790a97.d2dd75","type":"udp out","z":"d7ec3041.ef1a7","name":"","addr":"192.168.10.1","iface":"","port":"8889","ipv":"udp4","outport":"52955","base64":false,"multicast":"false","x":410,"y":360,"wires":[]},{"id":"704ca411.060e4c","type":"inject","z":"d7ec3041.ef1a7","name":"","topic":"","payload":"command","payloadType":"str","repeat":"10","crontab":"","once":false,"onceDelay":0.1,"x":170,"y":360,"wires":[["16790a97.d2dd75"]]},{"id":"e6e800c4.8f6e2","type":"inject","z":"d7ec3041.ef1a7","name":"","topic":"","payload":"streamon","payloadType":"str","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":160,"y":440,"wires":[["16790a97.d2dd75"]]},{"id":"7bc9e69b.5afad","type":"ffmpeg-stream","z":"d7ec3041.ef1a7","devicetype":"tello","url":"stream","name":"","x":150,"y":140,"wires":[]},{"id":"d64d51c8.8a886","type":"inject","z":"d7ec3041.ef1a7","name":"","topic":"","payload":"streamoff","payloadType":"str","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":160,"y":500,"wires":[["16790a97.d2dd75"]]}]

To import this flow , copy the json in the GIST above to your clipboard.

On your Node-RED editor go to the hamburger menu on top right-hand side > Import > Clipboard

Paste the json that you copied from your clipboard and click Import

You should see the following flow on your editor :

Step 8: Adjust Nodes and Stream from Tello!

Double click on the FFmpeg node and make sure it is configured for Tello Drone. Also by default, the URL will be set to /stream . Note, this can be changed to whatever url endpoint you want

In this example our stream will be accessible at ws://:/stream

Using the stream
To render the video stream in the browser, we use a library called JSMpeg.

*If you changed the stream url endpoint to something else , make sure you configure this in the template node on line 58. In the default we have it set as /stream *

const url = `ws://${window.location.hostname}:${window.location.port}/<url_endpoint>

Turn Tello Wifi On
Make sure your Tello drone is charged and you have turned it on. You should be able to connect to your Tello’s wifi.

Once connected click Command and then Stream on

Go to http://127.0.0.1:1880/dashboard and you should see video streaming from your tello! If you have trained an object detection model, you should also see your tello detect objects! In my case I trained a model to distinguish between thumbs up and thumbs down

If you want to stop stream make sure to click on the stopstream inject button in the node-RED editor

The End

That’s all folks :). Thank you so much for reading! Check out this repo if you’re interested in video streaming without the object detection!

If you found this tutorial fun and helpful, it would mean a lot to me if you gave it some <3 and shared it to help others! Thank you again!

Originally published on dev.to

#node-js #machine-learning #npm #data-science

What is GEEK

Buddha Community

Sasha Lee

1650636000

Dl4clj: Clojure Wrapper for Deeplearning4j.

dl4clj

Port of deeplearning4j to clojure

Contact info

If you have any questions,

my email is will@yetanalytics.com
I'm will_hoyt in the clojurians slack
twitter is @FeLungz (don't check very often)

TODO

update examples dir
finish README
- add in examples using Transfer Learning
finish tests
- eval is missing regression tests, roc tests
- nn-test is missing regression tests
- spark tests need to be redone
- need dl4clj.core tests
revist spark for updates
write specs for user facing functions
- this is very important, match isnt strict for maps
- provides 100% certianty of the input -> output flow
- check the args as they come in, dispatch once I know its safe, test the pure output
collapse overlapping api namespaces
add to core use case flows

Features

Stable Features with tests

Neural Networks DSL
Early Stopping Training
Transfer Learning
Evaluation
Data import

Features being worked on for 0.1.0

Clustering (testing in progress)
Spark (currently being refactored)
Front End (maybe current release, maybe future release. Not sure yet)
Version of dl4j is 0.0.8 in this project. Current dl4j version is 0.0.9
Parallelism
Kafka support
Other items mentioned in TODO

Features being worked on for future releases

NLP
Computational Graphs
Reinforement Learning
Arbiter

Artifacts

NOT YET RELEASED TO CLOJARS

fork or clone to try it out

If using Maven add the following repository definition to your pom.xml:

<repository>
  <id>clojars.org</id>
  <url>http://clojars.org/repo</url>
</repository>

Latest release

With Leiningen:

n/a

With Maven:

n/a

<dependency>
  <groupId>_</groupId>
  <artifactId>_</artifactId>
  <version>_</version>
</dependency>

Usage

Things you need to know

All functions for creating dl4j objects return code by default

All of these functions have an option to return the dl4j object
- :as-code? = false
This because all builders require the code representation of dl4j objects
- this requirement is not going to change
INDarray creation fns default to objects, this is for convenience
- :as-code? is still respected

API functions return code when all args are provided as code

API functions return the value of calling the wrapped method when args are provided as a mixture of objects and code or just objects

The tests are there to help clarify behavior, if you are unsure of how to use a fn, search the tests

for questions about spark, refer to the spark section bellow

Example of obj/code duality

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]))

;; as code (the default)

(l/dense-layer-builder
 :activation-fn :relu
 :learning-rate 0.006
 :weight-init :xavier
 :layer-name "example layer"
 :n-in 10
 :n-out 1)

;; =>

(doto
 (org.deeplearning4j.nn.conf.layers.DenseLayer$Builder.)
 (.nOut 1)
 (.activation (dl4clj.constants/value-of {:activation-fn :relu}))
 (.weightInit (dl4clj.constants/value-of {:weight-init :xavier}))
 (.nIn 10)
 (.name "example layer")
 (.learningRate 0.006))

;; as an object

(l/dense-layer-builder
 :activation-fn :relu
 :learning-rate 0.006
 :weight-init :xavier
 :layer-name "example layer"
 :n-in 10
 :n-out 1
 :as-code? false)

;; =>

#object[org.deeplearning4j.nn.conf.layers.DenseLayer 0x69d7d160 "DenseLayer(super=FeedForwardLayer(super=Layer(layerName=example layer, activationFn=relu, weightInit=XAVIER, biasInit=NaN, dist=null, learningRate=0.006, biasLearningRate=NaN, learningRateSchedule=null, momentum=NaN, momentumSchedule=null, l1=NaN, l2=NaN, l1Bias=NaN, l2Bias=NaN, dropOut=NaN, updater=null, rho=NaN, epsilon=NaN, rmsDecay=NaN, adamMeanDecay=NaN, adamVarDecay=NaN, gradientNormalization=null, gradientNormalizationThreshold=NaN), nIn=10, nOut=1))"]

General usage examples

Importing data

Loading data from a file (here its a csv)


(ns my.ns
 (:require [dl4clj.datasets.input-splits :as s]
           [dl4clj.datasets.record-readers :as rr]
           [dl4clj.datasets.api.record-readers :refer :all]
           [dl4clj.datasets.iterators :as ds-iter]
           [dl4clj.datasets.api.iterators :refer :all]
           [dl4clj.helpers :refer [data-from-iter]]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; file splits (convert the data to records)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def poker-path "resources/poker-hand-training.csv")
;; this is not a complete dataset, it is just here to sever as an example

(def file-split (s/new-filesplit :path poker-path))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; record readers, (read the records created by the file split)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def csv-rr (initialize-rr! :rr (rr/new-csv-record-reader :skip-n-lines 0 :delimiter ",")
                                 :input-split file-split))

;; lets look at some data
(println (next-record! :rr csv-rr :as-code? false))
;; => #object[java.util.ArrayList 0x2473e02d [1, 10, 1, 11, 1, 13, 1, 12, 1, 1, 9]]
;; this is our first line from the csv


;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; record readers dataset iterators (turn our writables into a dataset)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def rr-ds-iter (ds-iter/new-record-reader-dataset-iterator
                 :record-reader csv-rr
                 :batch-size 1
                 :label-idx 10
                 :n-possible-labels 10))

;; we use our record reader created above
;; we want to see one example per dataset obj returned (:batch-size = 1)
;; we know our label is at the last index, so :label-idx = 10
;; there are 10 possible types of poker hands so :n-possible-labels = 10
;; you can also set :label-idx to -1 to use the last index no matter the size of the seq

(def other-rr-ds-iter (ds-iter/new-record-reader-dataset-iterator
                       :record-reader csv-rr
                       :batch-size 1
                       :label-idx -1
                       :n-possible-labels 10))

(str (next-example! :iter rr-ds-iter :as-code? false))
;; =>
;;===========INPUT===================
;;[1.00, 10.00, 1.00, 11.00, 1.00, 13.00, 1.00, 12.00, 1.00, 1.00]
;;=================OUTPUT==================
;;[0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 1.00]


;; and to show that :label-idx = -1 gives us the same output

(= (next-example! :iter rr-ds-iter :as-code? false)
   (next-example! :iter other-rr-ds-iter :as-code? false)) ;; => true

INDArrays and Datasets from clojure data structures


(ns my.ns
  (:require [nd4clj.linalg.factory.nd4j :refer [vec->indarray matrix->indarray
                                                indarray-of-zeros indarray-of-ones
                                                indarray-of-rand vec-or-matrix->indarray]]
            [dl4clj.datasets.new-datasets :refer [new-ds]]
            [dl4clj.datasets.api.datasets :refer [as-list]]
            [dl4clj.datasets.iterators :refer [new-existing-dataset-iterator]]
            [dl4clj.datasets.api.iterators :refer :all]
            [dl4clj.datasets.pre-processors :as ds-pp]
            [dl4clj.datasets.api.pre-processors :refer :all]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; INDArray creation
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;;TODO: consider defaulting to code

;; can create from a vector

(vec->indarray [1 2 3 4])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x269df212 [1.00, 2.00, 3.00, 4.00]]

;; or from a matrix

(matrix->indarray [[1 2 3 4] [2 4 6 8]])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x20aa7fe1
;; [[1.00, 2.00, 3.00, 4.00], [2.00, 4.00, 6.00, 8.00]]]


;; will fill in spareness with zeros

(matrix->indarray [[1 2 3 4] [2 4 6 8] [10 12]])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x8b7796c
;;[[1.00, 2.00, 3.00, 4.00],
;; [2.00, 4.00, 6.00, 8.00],
;; [10.00, 12.00, 0.00, 0.00]]]

;; can create an indarray of all zeros with specified shape
;; defaults to :rows = 1 :columns = 1

(indarray-of-zeros :rows 3 :columns 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x6f586a7e
;;[[0.00, 0.00],
;; [0.00, 0.00],
;; [0.00, 0.00]]]

(indarray-of-zeros) ;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0xe59ffec 0.00]

;; and if only one is supplied, will get a vector of specified length

(indarray-of-zeros :rows 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x2899d974 [0.00, 0.00]]

(indarray-of-zeros :columns 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0xa5b9782 [0.00, 0.00]]

;; same considerations/defaults for indarray-of-ones and indarray-of-rand

(indarray-of-ones :rows 2 :columns 3)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x54f08662 [[1.00, 1.00, 1.00], [1.00, 1.00, 1.00]]]

(indarray-of-rand :rows 2 :columns 3)
;; all values are greater than 0 but less than 1
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x2f20293b [[0.85, 0.86, 0.13], [0.94, 0.04, 0.36]]]



;; vec-or-matrix->indarray is built into all functions which require INDArrays
;; so that you can use clojure data structures
;; but you still have the option of passing existing INDArrays

(def example-array (vec-or-matrix->indarray [1 2 3 4]))
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x5c44c71f [1.00, 2.00, 3.00, 4.00]]

(vec-or-matrix->indarray example-array)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x607b03b0 [1.00, 2.00, 3.00, 4.00]]

(vec-or-matrix->indarray (indarray-of-rand :rows 2))
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x49143b08 [0.76, 0.92]]

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; data-set creation
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def ds-with-single-example (new-ds :input [1 2 3 4]
                                    :output [0.0 1.0 0.0]))

(as-list :ds ds-with-single-example :as-code? false)
;; =>
;; #object[java.util.ArrayList 0x5d703d12
;;[===========INPUT===================
;;[1.00, 2.00, 3.00, 4.00]
;;=================OUTPUT==================
;;[0.00, 1.00, 0.00]]]

(def ds-with-multiple-examples (new-ds
                                :input [[1 2 3 4] [2 4 6 8]]
                                :output [[0.0 1.0 0.0] [0.0 0.0 1.0]]))

(as-list :ds ds-with-multiple-examples :as-code? false)
;; =>
;;#object[java.util.ArrayList 0x29c7a9e2
;;[===========INPUT===================
;;[1.00, 2.00, 3.00, 4.00]
;;=================OUTPUT==================
;;[0.00, 1.00, 0.00],
;;===========INPUT===================
;;[2.00, 4.00, 6.00, 8.00]
;;=================OUTPUT==================
;;[0.00, 0.00, 1.00]]]

;; we can create a dataset iterator from the code which creates datasets
;; and set the labels for our outputs (optional)

(def ds-with-multiple-examples
  (new-ds
   :input [[1 2 3 4] [2 4 6 8]]
   :output [[0.0 1.0 0.0] [0.0 0.0 1.0]]))

;; iterator
(def training-rr-ds-iter
  (new-existing-dataset-iterator
   :dataset ds-with-multiple-examples
   :labels ["foo" "baz" "foobaz"]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; data-set normalization
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; this gathers statistics on the dataset and normalizes the data
;; and applies the transformation to all dataset objects in the iterator
(def train-iter-normalized
  (c/normalize-iter! :iter training-rr-ds-iter
                     :normalizer (ds-pp/new-standardize-normalization-ds-preprocessor)
                     :as-code? false))

;; above returns the normalized iterator
;; to get fit normalizer

(def the-normalizer
  (get-pre-processor train-iter-normalized))

Model configuration

Creating a neural network configuration with singe and multiple layers

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.conf.distributions :as dist]
            [dl4clj.nn.conf.input-pre-processor :as pp]
            [dl4clj.nn.conf.step-fns :as s-fn]))

;; nn/builder has 3 types of args
;; 1) args which set network configuration params
;; 2) args which set default values for layers
;; 3) args which set multi layer network configuration params

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; single layer nn configuration
;; here we are setting network configuration
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(nn/builder :optimization-algo :stochastic-gradient-descent
            :seed 123
            :iterations 1
            :minimize? true
            :use-drop-connect? false
            :lr-score-based-decay-rate 0.002
            :regularization? false
            :step-fn :default-step-fn
            :layers {:dense-layer {:activation-fn :relu
                                   :updater :adam
                                   :adam-mean-decay 0.2
                                   :adam-var-decay 0.1
                                   :learning-rate 0.006
                                   :weight-init :xavier
                                   :layer-name "single layer model example"
                                   :n-in 10
                                   :n-out 20}})

;; there are several options within a nn-conf map which can be configuration maps
;; or calls to fns
;; It doesn't matter which option you choose and you don't have to stay consistent
;; the list of params which can be passed as config maps or fn calls will
;; be enumerated at a later date

(nn/builder :optimization-algo :stochastic-gradient-descent
            :seed 123
            :iterations 1
            :minimize? true
            :use-drop-connect? false
            :lr-score-based-decay-rate 0.002
            :regularization? false
            :step-fn (s-fn/new-default-step-fn)
            :build? true
            ;; dont need to specify layer order, theres only one
            :layers (l/dense-layer-builder
                    :activation-fn :relu
                    :updater :adam
                    :adam-mean-decay 0.2
                    :adam-var-decay 0.1
                    :dist (dist/new-normal-distribution :mean 0 :std 1)
                    :learning-rate 0.006
                    :weight-init :xavier
                    :layer-name "single layer model example"
                    :n-in 10
                    :n-out 20))

;; these configurations are the same

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; multi-layer configuration
;; here we are also setting layer defaults
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; defaults will apply to layers which do not specify those value in their config

(nn/builder
 :optimization-algo :stochastic-gradient-descent
 :seed 123
 :iterations 1
 :minimize? true
 :use-drop-connect? false
 :lr-score-based-decay-rate 0.002
 :regularization? false
 :default-activation-fn :sigmoid
 :default-weight-init :uniform

 ;; we need to specify the layer order
 :layers {0 (l/activation-layer-builder
             :activation-fn :relu
             :updater :adam
             :adam-mean-decay 0.2
             :adam-var-decay 0.1
             :learning-rate 0.006
             :weight-init :xavier
             :layer-name "example first layer"
             :n-in 10
             :n-out 20)
          1 {:output-layer {:n-in 20
                            :n-out 2
                            :loss-fn :mse
                            :layer-name "example output layer"}}})

;; specifying multi-layer config params

(nn/builder
 ;; network args
 :optimization-algo :stochastic-gradient-descent
 :seed 123
 :iterations 1
 :minimize? true
 :use-drop-connect? false
 :lr-score-based-decay-rate 0.002
 :regularization? false

 ;; layer defaults
 :default-activation-fn :sigmoid
 :default-weight-init :uniform

 ;; the layers
 :layers {0 (l/activation-layer-builder
             :activation-fn :relu
             :updater :adam
             :adam-mean-decay 0.2
             :adam-var-decay 0.1
             :learning-rate 0.006
             :weight-init :xavier
             :layer-name "example first layer"
             :n-in 10
             :n-out 20)
          1 {:output-layer {:n-in 20
                            :n-out 2
                            :loss-fn :mse
                            :layer-name "example output layer"}}}
 ;; multi layer network args
 :backprop? true
 :backprop-type :standard
 :pretrain? false
 :input-pre-processors {0 (pp/new-zero-mean-pre-pre-processor)
                        1 {:unit-variance-processor {}}})

Configuration to Trained models

Multi Layer models

an implementation of the dl4j mnist classification example

(ns my.ns
  (:require [dl4clj.datasets.iterators :as iter]
            [dl4clj.datasets.input-splits :as split]
            [dl4clj.datasets.record-readers :as rr]
            [dl4clj.optimize.listeners :as listener]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.multilayer.multi-layer-network :as mln]
            [dl4clj.nn.api.model :refer [init! set-listeners!]]
            [dl4clj.nn.api.multi-layer-network :refer [evaluate-classification]]
            [dl4clj.datasets.api.record-readers :refer [initialize-rr!]]
            [dl4clj.eval.api.eval :refer [get-stats get-accuracy]]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; nn-conf -> multi-layer-network
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123 :iterations 1 :regularization? true

   ;; setting layer defaults
   :default-activation-fn :relu :default-l2 7.5e-6
   :default-weight-init :xavier :default-learning-rate 0.0015
   :default-updater :nesterovs :default-momentum 0.98

   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}

   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def multi-layer-network (c/model-from-conf nn-conf))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; local cpu training with dl4j pre-built iterators
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; lets use the pre-built Mnist data set iterator

(def train-mnist-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-mnist-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

;; and lets set a listener so we can know how training is going

(def score-listener (listener/new-score-iteration-listener :print-every-n 5))

;; and attach it to our model

;; TODO: listeners are broken, look into log4j warnning
(def mln-with-listener (set-listeners! :model multi-layer-network
                                       :listeners [score-listener]))

(def trained-mln (mln/train-mln-with-ds-iter! :mln mln-with-listener
                                              :iter train-mnist-iter
                                              :n-epochs 15
                                              :as-code? false))

;; training happens because :as-code? = false
;; if it was true, we would still just have a data structure
;; we now have a trained model that has seen the training dataset 15 times
;; time to evaluate our model

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;Create an evaluation object
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def eval-obj (evaluate-classification :mln trained-mln
                                       :iter test-mnist-iter))

;; always remember that these objects are stateful, dont use the same eval-obj
;; to eval two different networks
;; we trained the model on a training dataset.  We evaluate on a test set

(println (get-stats :evaler eval-obj))
;; this will print the stats to standard out for each feature/label pair

;;Examples labeled as 0 classified by model as 0: 968 times
;;Examples labeled as 0 classified by model as 1: 1 times
;;Examples labeled as 0 classified by model as 2: 1 times
;;Examples labeled as 0 classified by model as 3: 1 times
;;Examples labeled as 0 classified by model as 5: 1 times
;;Examples labeled as 0 classified by model as 6: 3 times
;;Examples labeled as 0 classified by model as 7: 1 times
;;Examples labeled as 0 classified by model as 8: 2 times
;;Examples labeled as 0 classified by model as 9: 2 times
;;Examples labeled as 1 classified by model as 1: 1126 times
;;Examples labeled as 1 classified by model as 2: 2 times
;;Examples labeled as 1 classified by model as 3: 1 times
;;Examples labeled as 1 classified by model as 5: 1 times
;;Examples labeled as 1 classified by model as 6: 2 times
;;Examples labeled as 1 classified by model as 7: 1 times
;;Examples labeled as 1 classified by model as 8: 2 times
;;Examples labeled as 2 classified by model as 0: 3 times
;;Examples labeled as 2 classified by model as 1: 2 times
;;Examples labeled as 2 classified by model as 2: 1006 times
;;Examples labeled as 2 classified by model as 3: 2 times
;;Examples labeled as 2 classified by model as 4: 3 times
;;Examples labeled as 2 classified by model as 6: 3 times
;;Examples labeled as 2 classified by model as 7: 7 times
;;Examples labeled as 2 classified by model as 8: 6 times
;;Examples labeled as 3 classified by model as 2: 4 times
;;Examples labeled as 3 classified by model as 3: 990 times
;;Examples labeled as 3 classified by model as 5: 3 times
;;Examples labeled as 3 classified by model as 7: 3 times
;;Examples labeled as 3 classified by model as 8: 3 times
;;Examples labeled as 3 classified by model as 9: 7 times
;;Examples labeled as 4 classified by model as 2: 2 times
;;Examples labeled as 4 classified by model as 3: 1 times
;;Examples labeled as 4 classified by model as 4: 967 times
;;Examples labeled as 4 classified by model as 6: 4 times
;;Examples labeled as 4 classified by model as 7: 1 times
;;Examples labeled as 4 classified by model as 9: 7 times
;;Examples labeled as 5 classified by model as 0: 2 times
;;Examples labeled as 5 classified by model as 3: 6 times
;;Examples labeled as 5 classified by model as 4: 1 times
;;Examples labeled as 5 classified by model as 5: 874 times
;;Examples labeled as 5 classified by model as 6: 3 times
;;Examples labeled as 5 classified by model as 7: 1 times
;;Examples labeled as 5 classified by model as 8: 3 times
;;Examples labeled as 5 classified by model as 9: 2 times
;;Examples labeled as 6 classified by model as 0: 4 times
;;Examples labeled as 6 classified by model as 1: 3 times
;;Examples labeled as 6 classified by model as 3: 2 times
;;Examples labeled as 6 classified by model as 4: 4 times
;;Examples labeled as 6 classified by model as 5: 4 times
;;Examples labeled as 6 classified by model as 6: 939 times
;;Examples labeled as 6 classified by model as 7: 1 times
;;Examples labeled as 6 classified by model as 8: 1 times
;;Examples labeled as 7 classified by model as 1: 7 times
;;Examples labeled as 7 classified by model as 2: 4 times
;;Examples labeled as 7 classified by model as 3: 3 times
;;Examples labeled as 7 classified by model as 7: 1005 times
;;Examples labeled as 7 classified by model as 8: 2 times
;;Examples labeled as 7 classified by model as 9: 7 times
;;Examples labeled as 8 classified by model as 0: 3 times
;;Examples labeled as 8 classified by model as 2: 3 times
;;Examples labeled as 8 classified by model as 3: 2 times
;;Examples labeled as 8 classified by model as 4: 4 times
;;Examples labeled as 8 classified by model as 5: 3 times
;;Examples labeled as 8 classified by model as 6: 2 times
;;Examples labeled as 8 classified by model as 7: 4 times
;;Examples labeled as 8 classified by model as 8: 947 times
;;Examples labeled as 8 classified by model as 9: 6 times
;;Examples labeled as 9 classified by model as 0: 2 times
;;Examples labeled as 9 classified by model as 1: 2 times
;;Examples labeled as 9 classified by model as 3: 4 times
;;Examples labeled as 9 classified by model as 4: 8 times
;;Examples labeled as 9 classified by model as 6: 1 times
;;Examples labeled as 9 classified by model as 7: 4 times
;;Examples labeled as 9 classified by model as 8: 2 times
;;Examples labeled as 9 classified by model as 9: 986 times

;;==========================Scores========================================
;; Accuracy:        0.9808
;; Precision:       0.9808
;; Recall:          0.9807
;; F1 Score:        0.9807
;;========================================================================

;; can get the stats that are printed via fns in the evaluation namespace
;; after running eval-model-whole-ds

(get-accuracy :evaler evaler-with-stats) ;; => 0.9808

Model Tuning

Early Stopping (controlling training)

it is recommened you start here when designing models

using dl4clj.core


(ns my.ns
  (:require [dl4clj.earlystopping.termination-conditions :refer :all]
            [dl4clj.earlystopping.model-saver :refer [new-in-memory-saver]]
            [dl4clj.nn.api.multi-layer-network :refer [evaluate-classification]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :as iter]
            [dl4clj.core :as c]))

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123
   :iterations 1
   :regularization? true

   ;; setting layer defaults
   :default-activation-fn :relu
   :default-l2 7.5e-6
   :default-weight-init :xavier
   :default-learning-rate 0.0015
   :default-updater :nesterovs
   :default-momentum 0.98

   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}

   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def train-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

(def invalid-score-condition (new-invalid-score-iteration-termination-condition))

(def max-score-condition (new-max-score-iteration-termination-condition
                          :max-score 20.0))

(def max-time-condition (new-max-time-iteration-termination-condition
                         :max-time-val 10
                         :max-time-unit :minutes))

(def score-doesnt-improve-condition (new-score-improvement-epoch-termination-condition
                                     :max-n-epoch-no-improve 5))

(def target-score-condition (new-best-score-epoch-termination-condition
                             :best-expected-score 0.009))

(def max-number-epochs-condition (new-max-epochs-termination-condition :max-n 20))

(def in-mem-saver (new-in-memory-saver))

(def trained-mln
;; defaults to returning the model
  (c/train-with-early-stopping
   :nn-conf nn-conf
   :training-iter train-mnist-iter
   :testing-iter test-mnist-iter
   :eval-every-n-epochs 1
   :iteration-termination-conditions [invalid-score-condition
                                      max-score-condition
                                      max-time-condition]
   :epoch-termination-conditions [score-doesnt-improve-condition
                                  target-score-condition
                                  max-number-epochs-condition]
   :save-last-model? true
   :model-saver in-mem-saver
   :as-code? false))

(def model-evaler
  (evaluate-classification :mln trained-mln :iter test-mnist-iter))

(println (get-stats :evaler model-evaler))

explicit, step by step way of doing this

(ns my.ns
  (:require [dl4clj.earlystopping.early-stopping-config :refer [new-early-stopping-config]]
            [dl4clj.earlystopping.termination-conditions :refer :all]
            [dl4clj.earlystopping.model-saver :refer [new-in-memory-saver new-local-file-model-saver]]
            [dl4clj.earlystopping.score-calc :refer [new-ds-loss-calculator]]
            [dl4clj.earlystopping.early-stopping-trainer :refer [new-early-stopping-trainer]]
            [dl4clj.earlystopping.api.early-stopping-trainer :refer [fit-trainer!]]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.multilayer.multi-layer-network :as mln]
            [dl4clj.utils :refer [load-model!]]
            [dl4clj.datasets.iterators :as iter]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; start with our network config
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123 :iterations 1 :regularization? true
   ;; setting layer defaults
   :default-activation-fn :relu :default-l2 7.5e-6
   :default-weight-init :xavier :default-learning-rate 0.0015
   :default-updater :nesterovs :default-momentum 0.98
   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}
   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def mln (c/model-from-conf nn-conf))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; the training/testing data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def train-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; we are going to need termination conditions
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; these allow us to control when we exit training

;; this can be based off of iterations or epochs

;; iteration termination conditions

(def invalid-score-condition (new-invalid-score-iteration-termination-condition))

(def max-score-condition (new-max-score-iteration-termination-condition
                          :max-score 20.0))

(def max-time-condition (new-max-time-iteration-termination-condition
                         :max-time-val 10
                         :max-time-unit :minutes))

;; epoch termination conditions

(def score-doesnt-improve-condition (new-score-improvement-epoch-termination-condition
                                     :max-n-epoch-no-improve 5))

(def target-score-condition (new-best-score-epoch-termination-condition :best-expected-score 0.009))

(def max-number-epochs-condition (new-max-epochs-termination-condition :max-n 20))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; we also need a way to save our model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; can be in memory or to a local directory

(def in-mem-saver (new-in-memory-saver))

(def local-file-saver (new-local-file-model-saver :directory "resources/tmp/readme/"))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; set up your score calculator
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def score-calcer (new-ds-loss-calculator :iter test-iter
                                          :average? true))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; create an early stopping configuration
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; termination conditions
;; a way to save our model
;; a way to calculate the score of our model on the dataset

(def early-stopping-conf
  (new-early-stopping-config
   :epoch-termination-conditions [score-doesnt-improve-condition
                                  target-score-condition
                                  max-number-epochs-condition]
   :iteration-termination-conditions [invalid-score-condition
                                      max-score-condition
                                      max-time-condition]
   :eval-every-n-epochs 5
   :model-saver local-file-saver
   :save-last-model? true
   :score-calculator score-calcer))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; create an early stopping trainer from our data, model and early stopping conf
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def es-trainer (new-early-stopping-trainer :early-stopping-conf early-stopping-conf
                                            :mln mln
                                            :iter train-iter))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; fit and use our early stopping trainer
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def es-trainer-fitted (fit-trainer! es-trainer :as-code? false))

;; when the trainer terminates, you will see something like this
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO  Completed training epoch 14
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO  New best model: score = 0.005225599372851298,
;;                                                   epoch = 14 (previous: score = 0.018243224899038346, epoch = 7)
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO Hit epoch termination condition at epoch 14.
;;                                           Details: BestScoreEpochTerminationCondition(0.009)

;; and if we look at the es-trainer-fitted object we see

;;#object[org.deeplearning4j.earlystopping.EarlyStoppingResult 0x5ab74f27 EarlyStoppingResult
;;(terminationReason=EpochTerminationCondition,details=BestScoreEpochTerminationCondition(0.009),
;; bestModelEpoch=14,bestModelScore=0.005225599372851298,totalEpochs=15)]

;; and our model has been saved to /resources/tmp/readme/bestModel.bin
;; there we have our model config, model params and our updater state

;; we can then load this model to use it or continue refining it

(def loaded-model (load-model! :path "resources/tmp/readme/bestModel.bin"
                               :load-updater? true))

Transfer Learning (freezing layers)


;; TODO: need to write up examples

Spark Training

dl4j Spark usage

How it is done in dl4clj

Uses dl4clj.core
- This example uses a fn which takes care of most steps for you
  - allows you to pass args as code bc the fn accounts for the multiple spark contexts issue encountered when everything is just a data structure


(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :refer [new-iris-data-set-iterator]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.spark.masters.param-avg :as master]
            [dl4clj.spark.data.java-rdd :refer [new-java-spark-context
                                                java-rdd-from-iter]]
            [dl4clj.spark.api.dl4j-multi-layer :refer [eval-classification-spark-mln
                                                       get-spark-context]]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 1, create your model config
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def mln-conf
  (nn/builder
   :optimization-algo :stochastic-gradient-descent
   :default-learning-rate 0.006
   :layers {0 (l/dense-layer-builder :n-in 4 :n-out 2 :activation-fn :relu)
            1 {:output-layer
               {:loss-fn :negativeloglikelihood
                :n-in 2 :n-out 3
                :activation-fn :soft-max
                :weight-init :xavier}}}
   :backprop? true
   :backprop-type :standard))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 2, training master
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def training-master
  (master/new-parameter-averaging-training-master
   :build? true
   :rdd-n-examples 10
   :n-workers 4
   :averaging-freq 10
   :batch-size-per-worker 2
   :export-dir "resources/spark/master/"
   :rdd-training-approach :direct
   :repartition-data :always
   :repartition-strategy :balanced
   :seed 1234
   :save-updater? true
   :storage-level :none))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 3, spark context
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def your-spark-context
  (new-java-spark-context :app-name "example app"))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 4, training data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def iris-iter
  (new-iris-data-set-iterator
   :batch-size 1
   :n-examples 5))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, spark mln
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def fitted-spark-mln
  (c/train-with-spark :spark-context your-spark-context
                      :mln-conf mln-conf
                      :training-master training-master
                      :iter iris-iter
                      :n-epochs 1
                      :as-code? false))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, use spark context from spark-mln to create rdd
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; TODO: eliminate this step

(def our-rdd
  (let [sc (get-spark-context fitted-spark-mln :as-code? false)]
    (java-rdd-from-iter :spark-context sc
                        :iter iris-iter)))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 6, evaluation model and print stats (poor performance of model expected)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def eval-obj
  (eval-classification-spark-mln
   :spark-mln fitted-spark-mln
   :rdd our-rdd))

(println (get-stats :evaler eval-obj))

this example demonstrates the dl4j workflow
- NOTE: unlike the previous example, this one requires dl4j objects to be used
  - this is becaues spark only wants you to have one spark context at a time

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :refer [new-iris-data-set-iterator]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.spark.masters.param-avg :as master]
            [dl4clj.spark.data.java-rdd :refer [new-java-spark-context java-rdd-from-iter]]
            [dl4clj.spark.dl4j-multi-layer :as spark-mln]
            [dl4clj.spark.api.dl4j-multi-layer :refer [fit-spark-mln!
                                                       eval-classification-spark-mln]]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 1, create your model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def mln-conf
  (nn/builder
   :optimization-algo :stochastic-gradient-descent
   :default-learning-rate 0.006
   :layers {0 (l/dense-layer-builder :n-in 4 :n-out 2 :activation-fn :relu)
            1 {:output-layer
               {:loss-fn :negativeloglikelihood
                :n-in 2 :n-out 3
                :activation-fn :soft-max
                :weight-init :xavier}}}
   :backprop? true
   :as-code? false
   :backprop-type :standard))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 2, create a training master
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; not all options specified, but most are

(def training-master
  (master/new-parameter-averaging-training-master
   :build? true
   :rdd-n-examples 10
   :n-workers 4
   :averaging-freq 10
   :batch-size-per-worker 2
   :export-dir "resources/spark/master/"
   :rdd-training-approach :direct
   :repartition-data :always
   :repartition-strategy :balanced
   :seed 1234
   :as-code? false
   :save-updater? true
   :storage-level :none))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 3, create a Spark Multi Layer Network
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def your-spark-context
  (new-java-spark-context :app-name "example app" :as-code? false))

;; new-java-spark-context will turn an existing spark-configuration into a java spark context
;; or create a new java spark context with master set to "local[*]" and the app name
;; set to :app-name


(def spark-mln
  (spark-mln/new-spark-multi-layer-network
   :spark-context your-spark-context
   :mln mln-conf
   :training-master training-master
   :as-code? false))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 4, load your data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; one way is via a dataset-iterator
;; can make one directly from a dataset (iterator data-set)
;; see: nd4clj.linalg.dataset.api.data-set and nd4clj.linalg.dataset.data-set
;; we are going to use a pre-built one

(def iris-iter
  (new-iris-data-set-iterator
   :batch-size 1
   :n-examples 5
   :as-code? false))

;; now lets convert the data into a javaRDD

(def our-rdd
  (java-rdd-from-iter :spark-context your-spark-context
                      :iter iris-iter))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, fit and evaluate the model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def fitted-spark-mln
  (fit-spark-mln!
   :spark-mln spark-mln
   :rdd our-rdd
   :n-epochs 1))
;; this fn also has the option to supply :path-to-data instead of :rdd
;; that path should point to a directory containing a number of dataset objects

(def eval-obj
  (eval-classification-spark-mln
   :spark-mln fitted-spark-mln
   :rdd our-rdd))
;; we would want to have different testing and training rdd's but here we are using
;; the data we trained on

;; lets get the stats for how our model performed

(println (get-stats :evaler eval-obj))

Terminology

Coming soon

Packages to come back to:

Implement ComputationGraphs and the classes which use them

NLP

Parallelism

TSNE

https://deeplearning4j.org/doc/org/deeplearning4j/plot/package-summary.html

UI

Author: yetanalytics
Source Code: https://github.com/yetanalytics/dl4clj
License: BSD-2-Clause License

#machine-learning #deep-learning

Arvel Parker

1591611780

How to Find Ulimit For user on Linux

How can I find the correct ulimit values for a user account or process on Linux systems?

For proper operation, we must ensure that the correct ulimit values set after installing various software. The Linux system provides means of restricting the number of resources that can be used. Limits set for each Linux user account. However, system limits are applied separately to each process that is running for that user too. For example, if certain thresholds are too low, the system might not be able to server web pages using Nginx/Apache or PHP/Python app. System resource limits viewed or set with the NA command. Let us see how to use the ulimit that provides control over the resources available to the shell and processes.

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]

German Donnelly

1591993440

MEAN Stack Tutorial MongoDB ExpressJS AngularJS NodeJS

We are going to build a full stack Todo App using the MEAN (MongoDB, ExpressJS, AngularJS and NodeJS). This is the last part of three-post series tutorial.

MEAN Stack tutorial series:

AngularJS tutorial for beginners (Part I)
Creating RESTful APIs with NodeJS and MongoDB Tutorial (Part II)
MEAN Stack Tutorial: MongoDB, ExpressJS, AngularJS and NodeJS (Part III) 👈 you are here
Before completing the app, let’s cover some background about the this stack. If you rather jump to the hands-on part click here to get started.

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]

Santiago Homenick

1592595480

systemctl List All Failed Units/Services on Linux

Is there is a command to print list all failed units or services when using systemd on Linux? Can you tell me the systemctl command to list all failed services on Linux?

This quick tutorial explains how to find/list all failed systemd services/units on Linux operating systems using the systemctl command.

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]