ML.NET for developers without any AI experience

on YouTube ↗46m

" summary

Shows how a .NET team uses ML.NET plus Azure Cognitive Services to ship real features: a Chrome extension Should I Send It? that runs sentiment locally by loading an exported model ZIP (replacing an Azure call) and that raised agent sentiment reporting by ~20% after rollout; a Voice Bank flow that collects ~2,000 scripted sentences to synthesize a usable personal TTS voice; demos a profanity classifier (~72% accuracy on tiny data), a sarcasm model (~89% accuracy on 29k lines), and a taxi fare regressor that’s typically within ~$1.

AI-written summary · about

" chapters (14)

1. 0:00Intro
2. 1:16Rap video
3. 2:40Github Search Experiments
4. 6:20Lee Mallon
5. 10:08Should I Send
6. 11:41Audience Participation
7. 13:21Voicebank
8. 17:23Use Case
9. 22:47ML
10. 24:48Example
11. 30:39Quickdraw
12. 32:18Sample
13. 34:32Data Gathering
14. 38:02Addin

" related talks

• 01h 03mMachine learning for C# developers: Introducing MLJeff Prosise · DotNext 2019 Moscow
• 45mML: ML.NET: deep dive into dev AIJakub Wawrzyniak · 4Developers 2020
• 56mMachine Learning for .NET developersOlia Gavrysh · NDC Oslo 2020
• 41mAI + Machine Learning for .NET DevelopersCesar De la Torre Llorente · .NET Conf 2018
• 52mML.NET - Machine Learning made for .NETPeter Drougge · Update Conference Prague 2018
• 59mMachine Learning in .NET (ML.NET)Agatha Stena · .NET Conf 2018
• 58mExploring Machine Learning for .Net DevelopersNiloshima Srivastava · AzConf 2020
• 50mA Pragmatic Deep Dive into ML.NETDiederik Krols · Update Conference Prague 2019