config_online.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.

runner:
  train_data_dir: "data/train_with_insid"
  test_data_dir: "data/test_with_insid"
  # train_reader_path: "bq_reader_train"  # importlib format
  days: "{20210803..20210804}"
  pass_per_day: 1
  train_batch_size: 8
  test_batch_size: 8
  model_save_path: "output_model_dssm"

  reader_type: "InmemoryDataset"  # DataLoader / QueueDataset / RecDataset / InmemoryDataset
  pipe_command: "python3 bq_reader_train_insid.py"

  sync_mode: "async"
  # thread_num: 1
  train_thread_num: 1
  test_thread_num: 1

  use_gpu: False
  epochs: 1
  print_interval: 1

  dataset_debug: False

  # when you need to prune net, please set need_prune to True,
  # and need to set prune_feed_vars and prune_target_var in static_model.py
  need_prune: True

  parse_ins_id: True
  parse_content: True
  
  # when you need to dump fileds and params in training, please set need_train_dump to True,
  # and need to set train_dump_fields and train_dump_params in static_model.py
  need_train_dump: True
  # train_dump_fields_dir: "afs:/xxx"
  train_dump_fields_dir: "./train_dump_data"

  # when you need to dump fileds in inference, please set need_infer_dump to True,
  # and need to set infer_dump_fields in static_model.py
  need_infer_dump: True
  # infer_dump_fields_dir: "afs:/xxx"
  infer_dump_fields_dir: "./infer_dump_data"

  fs_name: "afs://xxx"
  fs_ugi: "xxx,xxx"
  
hyper_parameters:
  optimizer:
    class: adam
    learning_rate: 0.001
    strategy: sync
  trigram_d: 2900
  neg_num: 1
  slice_end: 8
  fc_sizes: [300, 300, 128]
  fc_acts: ['relu', 'relu', 'relu']