Basic Deep Learning Tasks from CPUs to GPUs

Basic Deep Learning Tasks from CPUs to GPUs > Part of the course [Multi-GPU Artificial Intelligence: Scaling AI with HPC](https://docs.google.com/document/d/1ztkd5I2k40QetHLwKdnOw4d6Ub_BsrR2epV2dt0wV3E/edit?tab=t.0) organized by CASTIEL2 and NCCs. [TOC] ## Something about GPU after intro to GPU architectures - [GPU programming](https://enccs.github.io/gpu-programming/) - GPU programming concepts ==???== - Introduction to GPU programming models ==???== - ==10-15 min?== ## Talk some basics of ML? - concepts - classifications - available algorithms - ~ 10 min --- --- ## ==Tentative schedule== ### First episode 14:00-15:45 - 10-15 min, GPU Programming models - 10 min, Intro to ML - 20 min, [Introduction](https://enccs.github.io/deep-learning-intro/1-introduction/) - 30 min, [Classification by a neural network using Keras (1-6)](https://enccs.github.io/deep-learning-intro/2-keras/) - 10 min, exercises - 20 min, - 30 min, [Classification by a neural network using Keras (7-10)](https://enccs.github.io/deep-learning-intro/2-keras/) ### Second episode 16:00-17:30 - 30 min, [Monitor the training process (1-6)](https://enccs.github.io/deep-learning-intro/3-monitor-the-model/) - 15 min, exercise - 20 min, [Monitor the training process (7-10)](https://enccs.github.io/deep-learning-intro/3-monitor-the-model/) - 10 min, exercise - 10 min, [advanced layer types](https://enccs.github.io/deep-learning-intro/4-advanced-layer-types/), [transfer learning](https://enccs.github.io/deep-learning-intro/5-transfer-learning/), and [outlook](https://enccs.github.io/deep-learning-intro/6-outlook/) - 5 min, Wrap-up