Organízate con las colecciones
Guarda y clasifica el contenido según tus preferencias.
Antes de empezar con una de las guías de inicio rápido de máquinas virtuales de TPU de Cloud, consulta el artículo Introducción a las TPU de Cloud, que ofrece una descripción general sobre cómo trabajar con las TPU de Cloud.
Las guías de inicio rápido de máquinas virtuales y de slices ofrecen una breve introducción sobre cómo trabajar con máquinas virtuales de TPU de Cloud mediante JAX y PyTorch. En estos temas se explica cómo instalar un framework de aprendizaje automático (ML) y ejecutar una aplicación de ejemplo en una máquina virtual de TPU de Cloud.
[[["Es fácil de entender","easyToUnderstand","thumb-up"],["Me ofreció una solución al problema","solvedMyProblem","thumb-up"],["Otro","otherUp","thumb-up"]],[["Es difícil de entender","hardToUnderstand","thumb-down"],["La información o el código de muestra no son correctos","incorrectInformationOrSampleCode","thumb-down"],["Me faltan las muestras o la información que necesito","missingTheInformationSamplesINeed","thumb-down"],["Problema de traducción","translationIssue","thumb-down"],["Otro","otherDown","thumb-down"]],[],[],[],null,["# Get started\n\nBefore starting one of the Cloud TPU VM quickstarts, see [Introduction to\nCloud TPU](/tpu/docs/intro-to-tpu) which gives an overview of working with\nCloud TPUs.\n\nFollow the instructions in [Set up the Cloud TPU\nenvironment](/tpu/docs/setup-gcp-account) to prepare to create a Cloud TPU.\n\nThe VM and slice quickstarts provide a brief introduction to working with\nCloud TPU VMs using JAX and PyTorch. These topics explain how to install a\nmachine learning (ML) framework and run a sample application on a Cloud TPU\nVM. \n\nSet up your environment\n-----------------------\n\n- [### Set up the Cloud TPU environment\n Configure your Google Cloud account and project for using Cloud TPUs,\n by activating the Cloud TPU API, setting up permissions, and\ninstalling tools.](/tpu/docs/setup-gcp-account) \n\nReserve TPUs\n------------\n\n- [### About Cloud TPU reservations\n Learn how to reserve Cloud TPU capacity for your workloads.](/tpu/docs/about-tpu-reservations)\n- [### Request a short-term reservation\n Reserve Cloud TPU capacity for a specific start time and duration, up to 90 days, by requesting a future reservation in calendar mode.](/tpu/docs/calendar-mode-reservation)\n- [### Request a long-term reservation\n Purchase long-term Cloud TPU reservations with committed use discounts for predictable capacity and cost savings.](/tpu/docs/long-term-reservation)\n- [### Consume a reservation\nUse your reserved Cloud TPU resources by creating TPUs that consume your reservation.](/tpu/docs/consume-reservation) \n\nCloud TPU VM quickstarts\n------------------------\n\n- [### Run JAX on Cloud TPU VMs\n Run a simple calculation on Cloud TPU VM with JAX.](/tpu/docs/run-calculation-jax)\n- [### Run PyTorch on Cloud TPU VMs\nRun a simple calculation on Cloud TPU VM with PyTorch.](/tpu/docs/run-calculation-pytorch) \n\nCloud TPU Pod quickstarts\n-------------------------\n\n- [### Train on Cloud TPU Pods\n Learn about best practices and common issues for Cloud TPU Pods.](/tpu/docs/training-on-tpu-pods)\n- [### Run JAX on Cloud TPU slices\n Run a simple calculation on Cloud TPU slice with JAX.](/tpu/docs/jax-pods)\n- [### Run PyTorch on Cloud TPU slices\n Run a simple calculation on Cloud TPU slice with PyTorch.](/tpu/docs/pytorch-pods)"]]
