tensorflow.distributed-cifar10.json

{
  "jobName": "tensorflow-distributed-cifar10",
  "image": "openpai/pai.example.tensorflow",

  "dataDir": "$PAI_DEFAULT_FS_URI/examples/tensorflow/distributed-cifar-10/data",
  "outputDir": "$PAI_DEFAULT_FS_URI/examples/tensorflow/distributed-cifar-10/output",
  "codeDir": "$PAI_DEFAULT_FS_URI/examples/tensorflow/distributed-cifar-10/code",

  "taskRoles": [
    {
      "name": "ps_server",
      "taskNumber": 2,
      "cpuNumber": 2,
      "memoryMB": 8192,
      "gpuNumber": 0,
      "command": "pip --quiet install scipy absl-py && python code/benchmarks/scripts/tf_cnn_benchmarks/tf_cnn_benchmarks.py --local_parameter_device=cpu --batch_size=32 --model=resnet20 --variable_update=parameter_server --data_dir=$PAI_DATA_DIR --data_name=cifar10 --train_dir=$PAI_OUTPUT_DIR --ps_hosts=$PAI_TASK_ROLE_ps_server_HOST_LIST --worker_hosts=$PAI_TASK_ROLE_worker_HOST_LIST --job_name=ps --task_index=$PAI_CURRENT_TASK_ROLE_CURRENT_TASK_INDEX"
    },
    {
      "name": "worker",
      "taskNumber": 2,
      "cpuNumber": 2,
      "memoryMB": 16384,
      "gpuNumber": 4,
      "command": "pip --quiet install scipy absl-py && python code/benchmarks/scripts/tf_cnn_benchmarks/tf_cnn_benchmarks.py --local_parameter_device=cpu --batch_size=32 --model=resnet20 --variable_update=parameter_server --data_dir=$PAI_DATA_DIR --data_name=cifar10 --train_dir=$PAI_OUTPUT_DIR --ps_hosts=$PAI_TASK_ROLE_ps_server_HOST_LIST --worker_hosts=$PAI_TASK_ROLE_worker_HOST_LIST --job_name=worker --task_index=$PAI_CURRENT_TASK_ROLE_CURRENT_TASK_INDEX",
      "minSucceededTaskCount": 2
    }
  ],
  "retryCount": 0
}