-
- - {$t('connectMB.output.header')} -
-
-
-
-
- {$t('connectMB.outputMB.sameButton')}
-
-
- - {$t('connectMB.outputMB.same')} -
- -
- {$t('connectMB.outputMB.otherButton')}
-
- 
-
-| {v.name} | +{v.value} | +
+ {/* Move padding so we get a reasonable click target. */}
+
+ |
+
- {countdownConfigs[countdownIdx].value} -
- {:else} -- {$t('content.data.recordingDialog.recording')} -
- {/if} -- {$t('content.data.addActionWalkThrough')} -
-- {$t('content.data.addRecordingWalkThrough')} -
-- {$t('content.model.notEnoughDataInfoBody1')} -
-- {$t('content.model.notEnoughDataInfoBody2')} -
- {:else if $state.hasTrainedBefore} -- {$t('content.model.retrainModelBody')} -
- {:else} -- {$t('content.model.trainModelBody')} -
- {/if} -{$t(descriptionId)}
- {/if} -{$t('connectMB.reconnecting')}
-LIVE
-{$t('connectMB.bluetooth.cancelledConnection')}
-{$t('connectMB.usbTryAgain.selectMicrobit')}
- {:else if type === 'close tabs'} -{$t('connectMB.usbTryAgain.closeTabs1')}
-{$t('connectMB.usbTryAgain.closeTabs2')}
- {:else} -{$t('connectMB.usbTryAgain.replugMicrobit1')}
- -{$t('connectMB.usbTryAgain.replugMicrobit2')}
-{$t('connectMB.usbTryAgain.replugMicrobit3')}
{$t('connectMB.usbTryAgain.replugMicrobit4')}
- {#if $state.reconnectState.reconnectFailed} - {$t('reconnectFailed.subtitle')} - {/if} -
-- {$t(item.titleId)} -
- {#if item.subtitleId} -{$t(item.subtitleId)}
- {/if} -{$t('connectMB.pattern.subtitle')}
-- {$t('connectMB.bluetooth.invalidPattern')} -
-{$t('connectMB.connecting')}
-
- {$t('connectMB.connectBattery.subtitle')}
-
- {$t('connectMB.connectBattery.link')}
-
- {$t(subtitleId)} -
-{$t('connectMB.connecting')}
-- {$t('connectMB.usb.firmwareBroken.content1')} -
-
-
-
- {$t('connectMB.usb.firmwareBroken.content3')}
-
- {$t('connectMB.usbDownloading.subtitle')} -
-
-
- {$t('connectMB.transferHex.message')} -
-
-
| {$t('content.index.title')} | {import.meta.env.VITE_APP_VERSION} |
- {$t('content.data.choice.header')} -
-- {$t('content.data.choice.body')} -
-- {$t('content.index.recordButtonDescription')} -
-{$t('popup.appVersionRedirect.explain')}
-{$t('popup.appVersionRedirect.uk')}
-{$t('popup.compatibility.explain')}
-{$t('popup.compatibility.advice')}
-{$t('performanceWarning.content1')}
-{$t('performanceWarning.content2')}
- - {$t('performanceWarning.troubleshoot')} -{$t('sign-up.content')}
-
-
- {$t('connectMB.unsupportedMicrobit.withBluetooth')} -
- {:else} -
-
- {dimension.label} -
- {/each} -- {hoverIndex} -
- {/if} - - -- {iconText} -
- {/if} -- {titleText} -
- {/if} - {#if bodyText} -- {bodyText} -
- {/if} -- {Math.round(meterWidthPct)}% -
-- {$t('content.model.output.recognitionPoint')} -
- -- {line}% -
-{currentPin}
-{currentPin}
-{$t('content.model.output.pin.option.allTime')}
-{$t('content.model.output.pin.option.xTime')}
-{$t('content.model.output.pin.seconds')}
- -Buffer utilization = {utilization * 100}%
-- Buffer utilization should be about 20-50%. Too high means performance spikes may cause - the buffer to lack sufficient data to classify. Too low means performance is wasted -
diff --git a/src/components/playground/PlaygroundContext.ts b/src/components/playground/PlaygroundContext.ts deleted file mode 100644 index 3f88ac664..000000000 --- a/src/components/playground/PlaygroundContext.ts +++ /dev/null @@ -1,40 +0,0 @@ -/** - * (c) 2023, Center for Computational Thinking and Design at Aarhus University and contributors - * - * SPDX-License-Identifier: MIT - */ -import { Readable, Subscriber, Unsubscriber, Writable, writable } from 'svelte/store'; -export type PlaygroundContextData = { - messages: any[]; -}; -class PlaygroundContext implements ReadableGesture: {$gesture.name}
-- {JSON.stringify(reduceData($gesture), null, 2)} -
-- > {message.toString()} -
- {/each} -Synthesis interval ({$accelerometerSynthesizer.intervalSpeed}), lower is faster
-Live data synthesizer
-Uses sine-waves to produce LiveData
-x Speed (Frequency)
-y Speed (Frequency)
-z Speed (Frequency)
-{JSON.stringify($accelerometerSynthesizer, null, 2)}
-{numberOfRecodings}
-{$t('menu.data.examples')}
-- {$t(title)} -
-- {predictionLabel} -
-- {confidenceLabel} -
- {/if} -- {$t('menu.trainer.TrainingFinished')} -
-- To start using the micro:bit machine learning tool, you need to connect a micro:bit to - the micro:bit machine learning tool web page on your computer. -
-There are two options for connecting the micro:bit.
-The first option uses a Web Bluetooth wireless connection.
-Click on 'Start new session' and follow the instructions on screen.
-- If you can't use Bluetooth on your computer, the second option uses a micro:bit radio - connection instead. -
-- Click on 'Start new session', then 'Connect using micro:bit radio instead' and follow - the instructions on screen. -
-- The micro:bit radio option uses two micro:bits. The first micro:bit senses movement. - It sends movement data using radio to a second micro:bit connected to your computer - with a USB cable. -
-- Once the micro:bit sensing movement is connected, you'll see a happy face on its LED - display. -
-- As you move the micro:bit, you see live movement data from its accelerometer sensor in - a graph at the bottom of the screen. -
-- Each coloured line represents a different direction, or dimension, you're moving the - micro:bit in. -
-Looking at the front of the micro:bit:
-This data is how the machine learning tool understands how you move.
-- There are three steps to teaching the computer how to use this data to recognise - different movements you make. -
-- Choose at least two different movements, or actions, to train the - machine learning tool to recognise. Waving and clapping are good actions to start - with. -
-- The machine learning tool analyses your samples of data and makes a set of - mathematical rules to allow it to estimate which physical action you make. These rules - make up the machine learning model. -
-This screen shows which action the model estimates that you're making.
-- The percentage number is how certain the model is that you're making an action. High - numbers mean the model is very sure you're making that action. Lower numbers mean it - is less sure. -
-- Make each of your actions to test the model. If you see low certainty numbers, the - model may need more data samples. If it has got the estimated action wrong, you'll also need to add more data. -
-- You can now go back to step 1 to collect more data to make your machine learning model - more reliable, or recognise other kinds of movement. -
-- Introduce students to machine learning concepts through physical movement and - data -
-- {$t(step.descriptionId)} -
-- The micro:bit machine learning tool introduces students to machine learning - (ML) concepts through physical movement and data, a topic familiar from mathematics and - science. -
-- It allows students to quickly teach a computer to tell the difference between - different physical movements, such as waving and clapping, using data captured by a - BBC micro:bit. -
-- Students collect samples of data of different movements and label them, so the - computer knows which sets of data represent which movements. -
-- The computer learns to recognise different movements by analysing the data samples, - and creating a set of rules called a model. This process of - collecting and labelling data to build a model is called training a machine - learning model. -
-- Students then test and improve the model. This enables learning about the importance - of collecting enough data from diverse sources when designing and making ML systems. -
-- The micro:bit machine learning tool is co-designed and tested in classrooms in Denmark - and the United Kingdom by Aarhus University and the Micro:bit Educational Foundation. -
-- There are no downloads, no logins. You just need the Chrome or Edge web browser and a - micro:bit - or two micro:bits if your computer doesn’t have Bluetooth enabled. -
-- The BBC micro:bit has a built-in accelerometer, a movement sensor also found in phones - and fitness trackers. It measures movement in three dimensions: x, y and z. -
-Looking at the front of the micro:bit:
-- Students hold or wear a micro:bit that sends movement data to the machine learning - tool. They see their physical movement captured as data in real time, as moving lines - on a graph. Making the connection between moving your body and seeing your movements - captured and visualised as the computer sees it, as data, is a powerful, immersive - experience. -
-- To create an ML model, students record samples of data of different types of - movements, such as waving or clapping. We call these different movements ‘actions’. - Students label each set of actions so the computer knows which sets of data represent - which kinds of movement. -
-- Students then use the data they have collected to train and test a machine learning - model running in the web browser. -
-- They’ll discover that to work well, successfully telling one action from another, the - model needs plenty of data. They’ll also find that to work for different people, who - move in different ways, the training data needs to be collected from as many different - people as possible. -
-- They can then improve the model by recording more samples of movement data, training - and testing the machine learning model again. -
-- Try it out for yourself, to see how quick and easy it is to get started. The video and - short written guide show you how. -
-- Artificial intelligence: the ability of computer software designed by - people to perform tasks that usually need human intelligence. -
-Data: information collected for use elsewhere.
-- Machine learning: the application of AI technology where computer software - is designed to perform a task quickly and reliably, having been trained by examples of - data provided by humans. This training process can be described as 'learning' and this - is why we use the term 'machine learning'. -
-- Machine learning model: a set of rules developed by a machine learning - system to categorise data. -
-- Training a machine learning model: providing samples of data categorised - and labelled by humans to help machine learning software to build a model. -
-- Testing a machine learning model: evaluating a machine learning model - against labelled or known data to see if it is correctly categorising that labelled or - known data. If it is not correct, it may need further training data. -
-Model store
-- {JSON.stringify($model, null, 2).substring( - 2, - JSON.stringify($model, null, 2).length - 1, - )} -
-Classifier store
-- {JSON.stringify($classifier, null, 2).substring( - 2, - JSON.stringify($classifier, null, 2).length - 1, - )} -
-Engine store
-- {JSON.stringify($engine, null, 2).substring( - 2, - JSON.stringify($engine, null, 2).length - 1, - )} -
-LiveData store
-- {JSON.stringify($liveData, null, 2).substring( - 2, - JSON.stringify($liveData, null, 2).length - 1, - )} -
-Gestures (without recordings/output)
- {#each $gestures as { ID }} -- {$t('content.filters.NoDataBody')} -
-{$t('content.index.newzHeading')}
-- {$t('content.index.newzBodyMakecode')} -
-- {$t('content.index.newzBodyViz')} -
-{$t('content.index.MLHeading')}
-- {$t('content.index.MLSubheading')} -
-- {$t(step.text)} -
-- {$t('content.model.output.estimatedGesture.iconTitle')} -
-- {prediction.name} -
-- {prediction.certainty} -
- {/if} -- {$t('content.model.output.popup.header')} -
-- {$t('content.model.output.popup.body')} -
-MakeCode
-- You can create a hex file on - MakeCode - -
-- {$t('content.trainer.failure.body')} -
-- {$t('content.trainer.failure.todo')} -
-- {$t('content.trainer.description')} -
-