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.

# 轮数
epochs: 10
# 设备
device: cpu
# 工作目录
workspace: "paddlerec.models.rank.dnn"

# dataset列表
dataset:
- name: dataset_train # 名字，用来区分不同的dataset
  batch_size: 2
  type: DataLoader # 或者QueueDataset 
  data_path: "{workspace}/data/sample_data/train" # 数据路径
  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 # 名字，用来区分不同的dataset
  batch_size: 2
  type: DataLoader # 或者QueueDataset
  data_path: "{workspace}/data/sample_data/test" # 数据路径
  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"

# 超参数
hyper_parameters:
  #优化器
  optimizer:
    class: Adam
    learning_rate: 0.001
    strategy: async
  # 用户自定义
  sparse_inputs_slots: 27
  sparse_feature_number: 1000001
  sparse_feature_dim: 9
  dense_input_dim: 13
  fc_sizes: [512, 256, 128, 32]

mode: runner1
# runner配置
runner:
 - name: runner1
   class: single_train
   save_checkpoint_interval: 2 # 保存模型
   save_inference_interval: 4 # 保存预测模型
   save_checkpoint_path: "increment" # 保存模型路径
   save_inference_path: "inference" # 保存预测模型路径
   #save_inference_feed_varnames: [] # 预测模型feed vars
   #save_inference_fetch_varnames: [] # 预测模型 fetch vars
   #init_model_path: "xxxx" # 加载模型
 - name: runner2
   class: single_infer
   init_model_path: "increment/0" # 加载模型

# 执行器，每轮要跑的所有阶段
phase:
 - name: phase1
   model: "{workspace}/model.py" # 模型路径
   dataset_name: dataset_train # 名字，用来区分不同的阶段
   thread_num: 1 # 线程数
# - name: phase2
#   model: "{workspace}/model.py" # 模型路径
#   dataset_name: dataset_infer # 名字，用来区分不同的阶段
#   thread_num: 1 # 线程数