config.yaml

# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# workspace
workspace: "models/rank/dnn"

# list of dataset
dataset:
- name: dataloader_train # name of dataset to distinguish different datasets
  batch_size: 2
  type: DataLoader # or QueueDataset 
  data_path: "{workspace}/data/sample_data/train"
  sparse_slots: "click 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26"
  dense_slots: "dense_var:13"
- name: dataset_train # name of dataset to distinguish different datasets
  batch_size: 2
  type: QueueDataset # or DataLoader 
  data_path: "{workspace}/data/sample_data/train"
  sparse_slots: "click 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26"
  dense_slots: "dense_var:13"
- name: dataset_infer # name
  batch_size: 2
  type: DataLoader # or QueueDataset
  data_path: "{workspace}/data/sample_data/train"
  sparse_slots: "click 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26"
  dense_slots: "dense_var:13"

# hyper parameters of user-defined network
hyper_parameters:
  # optimizer config
  optimizer:
    class: Adam
    learning_rate: 0.001
    strategy: async
  # user-defined <key, value> pairs
  sparse_inputs_slots: 27
  sparse_feature_number: 1000001
  sparse_feature_dim: 9
  dense_input_dim: 13
  fc_sizes: [512, 256, 128, 32]
  distributed_embedding: 0

# select runner by name
mode: [single_cpu_train, single_cpu_infer]
# config of each runner.
# runner is a kind of paddle training class, which wraps the train/infer process.
runner:
- name: single_cpu_train
  class: train
  # num of epochs
  epochs: 4
  # device to run training or infer
  device: cpu
  save_checkpoint_interval: 2 # save model interval of epochs
  save_inference_interval: 4 # save inference
  save_checkpoint_path: "increment_dnn" # save checkpoint path
  save_inference_path: "inference" # save inference path
  save_inference_feed_varnames: [] # feed vars of save inference
  save_inference_fetch_varnames: [] # fetch vars of save inference
  print_interval: 10
  phases: [phase1]

- name: single_cpu_infer
  class: infer
  # num of epochs
  epochs: 1
  # device to run training or infer
  device: cpu
  init_model_path: "increment_dnn" # load model path
  phases: [phase2]

- name: ps_cluster
  class: cluster_train
  epochs: 2
  device: cpu
  fleet_mode: ps
  save_checkpoint_interval: 1 # save model interval of epochs
  save_checkpoint_path: "increment_dnn" # save checkpoint path
  init_model_path: "" # load model path
  print_interval: 1
  phases: [phase1]

- name: online_learning_cluster
  class: cluster_train
  runner_class_path: "{workspace}/online_learning_runner.py"
  epochs: 2
  device: cpu
  fleet_mode: ps
  save_checkpoint_interval: 1 # save model interval of epochs
  save_checkpoint_path: "increment_dnn" # save checkpoint path
  init_model_path: "" # load model path
  print_interval: 1
  phases: [phase1]

- name: collective_cluster
  class: cluster_train
  epochs: 2
  device: gpu
  fleet_mode: collective
  save_checkpoint_interval: 1 # save model interval of epochs
  save_checkpoint_path: "increment_dnn" # save checkpoint path
  init_model_path: "" # load model path
  print_interval: 1
  phases: [phase1]

- name: single_multi_gpu_train
  class: train
  # num of epochs
  epochs: 1
  # device to run training or infer
  device: gpu
  selected_gpus: "0,1" # 选择多卡执行训练
  save_checkpoint_interval: 1 # save model interval of epochs
  save_inference_interval: 4 # save inference
  save_step_interval: 1
  save_checkpoint_path: "increment_dnn" # save checkpoint path
  save_inference_path: "inference" # save inference path
  save_step_path: "step_save"
  save_inference_feed_varnames: [] # feed vars of save inference
  save_inference_fetch_varnames: [] # fetch vars of save inference
  print_interval: 1
  phases: [phase1]
# runner will run all the phase in each epoch
phase:
- name: phase1
  model: "{workspace}/model.py" # user-defined model
  dataset_name: dataloader_train # select dataset by name
  thread_num: 1

- name: phase2
  model: "{workspace}/model.py" # user-defined model
  dataset_name: dataset_infer # select dataset by name
  thread_num: 1