Since their introduction in 2017, Transformers have become the de facto standard for tackling a wide range of NLP tasks in both academia and industry. In this workshop, we'll teach you the core concepts behind Transformers and how to train these models in the Hugging Face ecosystem. Read more about the workshop here https://www.mlt.ai/hands-on-transformers-workshop https://youtu.be/qvwyUiaRCD4 📚 … Continue reading WORKSHOP: NLP WITH TRANSFORMERS 🤗
MLTOKYO (study group)
ML AMA WITH LAURENCE MORONEY @GOOGLE I/O 2021
MLT welcomed Google AI Lead Laurence Moroney for an AMA session during Google I/O 2021! In the first part of the session Laurence summarized the top 10 I/O announcements for ML Developers, and afterwards we jumped right into audience questions. https://www.youtube.com/watch?v=fqK4mchNlnY Top 10 I/O announcements for ML Developers #1 Reduce your ML app size: TF … Continue reading ML AMA WITH LAURENCE MORONEY @GOOGLE I/O 2021
CLOUD NEXT – DATAOPS AND DATA ENGINEERING BEST PRACTICES
Understanding production-level machine learning and data engineering at scale has become essential for industry deployment. We're excited to welcome Director Aram Lauxtermann, Tech Leads Trevor Chinn, and Ken Tso from KPMG Ignition Tokyo for a talk on DataOps and Data Engineering Best Practices. Within a year and a half Cloud Next has been rolled out … Continue reading CLOUD NEXT – DATAOPS AND DATA ENGINEERING BEST PRACTICES
AI DIGEST #NOVEMBER20
By Alisher Abdulkhaev Issue #17: November 2020 AlphaFold: a solution to a 50-year-old grand challenge in biology The latest version of AlphaFold (AlphaFold-2) has been recognised as a solution to one of biology's grand challenges - the “protein folding problem”.It was validated at CASP14, the biennial Critical Assessment of protein Structure PredictionWe’re excited about the potential impact … Continue reading AI DIGEST #NOVEMBER20
ML MATH: MATRIX DECOMPOSITIONS
Mathematics for Machine Learning – Matrix Decompositions
CNN ARCHITECTURES: SHUFFLENET
This is part of the CNN Architectures series by Dimitris Katsios. Find all CNN Architectures online: Notebooks: MLT GitHubVideo tutorials: YouTubeSupport MLT on Patreon ShuffleNet We will use the tensorflow.keras Functional API to build ShuffleNet from the original paper: “ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices” by Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, Jian Sun. ShuffleNet: Video tutorial … Continue reading CNN ARCHITECTURES: SHUFFLENET