У нас вы можете посмотреть бесплатно PhiFlow Tutorial: Smoke Plume Simulation или скачать в максимальном доступном качестве, видео которое было загружено на ютуб. Для загрузки выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием видео, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса ClipSaver.ru
PhiFlow is a Python package that builds upon modern deep learning frameworks like JAX, PyTorch and TensorFlow. It simplifies the simulation of PDEs like for Comptational Fluid Dynamics and enables "Differentiable Physics". Here is the code: https://github.com/Ceyron/machine-lea... "Differential Physics" is a modern paradigm that builds upon a clever combination of simulation techniques (like Finite Differences, Finite Volumes, Finite Volumes, Smoothed Particle Hydrodynamics etc.) and adjoint methods to interleave physics and deep learning. This can be used to solve inverse problems, train low-level deep learning correction models or control complicated PDEs. PhiFlow enables this paradigm by making many advanced simulation techniques easily available from a high-level language like Python. It uses the powers of modern deep learning packages and thereby allows the efficient usage of accelerators like GPUs and TPUs. In this video, we will only use the simulation aspects of PhiFlow to compute a simple scenario in fluid mechanics. Here, a species (hot smoke) is immersed in a fluid. The fluid adheres to the Navier-Stokes equations. The hot smoke follows an advection equation that is driven by the fluids velocity field. Additionally, the hot smoke exerts a force due to Buoyancy into the momentum equation of the fluid. Here is the documentation of Φ-Flow: https://tum-pbs.github.io/PhiFlow/ Check out the GitHub Repo of Φ-Flow: https://github.com/tum-pbs/PhiFlow Also check out the tutorial series over on @holl-science 's channel: • Φ-Flow Tutorials ------- 📝 : Check out the GitHub Repository of the channel, where I upload all the handwritten notes and source-code files (contributions are very welcome): https://github.com/Ceyron/machine-lea... 📢 : Follow me on LinkedIn or Twitter for updates on the channel and other cool Machine Learning & Simulation stuff: / felix-koehler and / felix_m_koehler 💸 : If you want to support my work on the channel, you can become a Patreon here: / mlsim ------- Timestamps: 00:00 Intro 00:22 What is PhiFlow? Differentiable Physics? 00:49 Governing Equations (with Buoyancy) 01:30 Scenario and Boundary Conditions 02:08 Solution Strategy 04:30 Imports 05:10 Main Function Boilerplate 05:21 Defining velocity staggered field 06:50 Defining smoke centered field 07:56 Pre-Define inflow smoke pattern 09:35 Simulation Step Function 14:24 Simulation Loop + Visualization 16:22 First Run + Small Bug Fix 17:05 Changing to JAX + JIT for faster simulation 19:30 Discussion 19:43 Outro