Hyperparameters supported by PAI-TensorFlow

You can configure hyperparameters by using a local file. The following code shows the format of the local file:

batch_size=10
learning_rate=0.01

PAI-TensorFlow SDK for Python provides the parameters required to obtain the hyperparameters. You can use tf.app.flags.FLAGS to read the required hyperparameters and pass them to the script that is running. Then, the hyperparameter definitions can be obtained from the model training file. Procedure:

1. Assume that the hyperparameter file is stored in oss://xxx.oss-cn-beijing.aliyuncs.com/tf/hyper_para.txt. Use the following sample code to read hyperparameters:

   import tensorflow as tf
   tf.app.flags.DEFINE_string("learning_rate", "", "learning_rate")
   tf.app.flags.DEFINE_string("batch_size", "", "batch size")
   FAGS = tf.app.flags.FLAGS
   print("learning rate:" + FAGS.learning_rate)
   print("batch size:" + FAGS.batch_size)

2. Use -DhyperParameters to pass the hyperparameters to the script that is running. Example:

    pai -name tensorflow1120_ext
            -Dscript='oss://xxx.oss-cn-beijing.aliyuncs.com/tf/hello_hyperpara.py'
            -Dbuckets='oss://xxx.oss-cn-beijing.aliyuncs.com/'
            -DhyperParameters='oss://xxx.oss-cn-beijing.aliyuncs.com/tf/hyper_para.txt'
            -Darn='acs:ram::111***:role/***role';

You can pass PAI-TensorFlow parameters in the form of strings by using userDefinedParameters. Example:

pai -name tensorflow1120_ext
         -Dscript='oss://xxx.oss-cn-beijing.aliyuncs.com/tf/hello_hyperpara.py'
         -Dbuckets='oss://xxx.oss-cn-beijing.aliyuncs.com/'
         -DuserDefinedParameters="--batch_size=10 --learning_rate=0.01"
         -Darn='acs:ram::111***:role/***role';