[英]Train my model on GCP TPU
What architecture of program can run in TPU? TPU中可以运行什么程序架构?
Is it must to use tf.contrib.tpu.TPUEstimator
? 是否必须使用tf.contrib.tpu.TPUEstimator
?
Is there any tpu-program example for reference except the tensorflow official model? 除tensorflow官方模型外,是否有任何tpu程序示例可供参考?
Google Cloud TPUs uses TensorFlow [1] which supports Python, C++, Java, Go and Swift APIs [2]. Google Cloud TPU使用TensorFlow [1],它支持Python,C ++,Java,Go和Swift API [2]。 Here is the link to Google Cloud TPU System Architecture [1]. 这是Google Cloud TPU系统架构[1]的链接。 TPUEstimator handle low-level, hardware-specific details and it makes it easier to achieve maximum performance by applying performance optimizations automatically on your behalf [3]. TPUEstimator处理低级的,特定于硬件的详细信息,并且它可以代表您自动应用性能优化,从而更轻松地实现最高性能[3]。 Also, if you are using for neural network, then you should use TPUEstimator. 同样,如果您用于神经网络,则应使用TPUEstimator。 Here are some steps to convert from TensorFlow Estimator to TPUEstimator [4]. 以下是将TensorFlow估算器转换为TPUEstimator [4]的一些步骤。 Additional tutorials are located on Google Cloud TPU Tutorials page [5] and also on GitHub [6]. 其他教程位于Google Cloud TPU教程页面[5]和GitHub [6]上。
[1] https://cloud.google.com/tpu/docs/system-architecture [1] https://cloud.google.com/tpu/docs/system-architecture
[2] https://www.tensorflow.org/api_docs/ [2] https://www.tensorflow.org/api_docs/
[3] https://cloud.google.com/tpu/docs/using-estimator-api [3] https://cloud.google.com/tpu/docs/using-estimator-api
[4] https://cloud.google.com/tpu/docs/using-estimator-api#converting_from_tensorflow_estimator_to_tpuestimator [4] https://cloud.google.com/tpu/docs/using-estimator-api#converting_from_tensorflow_estimator_to_tpuestimator
[5] https://cloud.google.com/tpu/docs/tutorials [5] https://cloud.google.com/tpu/docs/tutorials
[6] https://github.com/tensorflow/tpu [6] https://github.com/tensorflow/tpu
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.