pipeline-data-statistics-partial-column-missing.py 4.0 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107
  1. #
  2. # Copyright 2019 The FATE Authors. All Rights Reserved.
  3. #
  4. # Licensed under the Apache License, Version 2.0 (the "License");
  5. # you may not use this file except in compliance with the License.
  6. # You may obtain a copy of the License at
  7. #
  8. # http://www.apache.org/licenses/LICENSE-2.0
  9. #
  10. # Unless required by applicable law or agreed to in writing, software
  11. # distributed under the License is distributed on an "AS IS" BASIS,
  12. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13. # See the License for the specific language governing permissions and
  14. # limitations under the License.
  15. #
  16. import argparse
  17. import json
  18. from pipeline.backend.pipeline import PipeLine
  19. from pipeline.component import DataTransform
  20. from pipeline.component.evaluation import Evaluation
  21. from pipeline.component.intersection import Intersection
  22. from pipeline.component.reader import Reader
  23. from pipeline.interface.data import Data
  24. from pipeline.interface.model import Model
  25. from pipeline.component.data_statistics import DataStatistics
  26. from pipeline.utils.tools import load_job_config
  27. def prettify(response, verbose=True):
  28. if verbose:
  29. print(json.dumps(response, indent=4, ensure_ascii=False))
  30. print()
  31. return response
  32. def main(config="../../config.yaml", namespace=""):
  33. if isinstance(config, str):
  34. config = load_job_config(config)
  35. parties = config.parties
  36. guest = parties.guest[0]
  37. hosts = parties.host[0]
  38. guest_train_data = {"name": "ionosphere_scale_hetero_guest", "namespace": f"experiment{namespace}"}
  39. host_train_data = {"name": "ionosphere_scale_hetero_host", "namespace": f"experiment{namespace}"}
  40. # guest_train_data = {"name": "default_credit_hetero_guest", "namespace": f"experiment{namespace}"}
  41. # host_train_data = {"name": "default_credit_hetero_host", "namespace": f"experiment{namespace}"}
  42. # initialize pipeline
  43. pipeline = PipeLine()
  44. # set job initiator
  45. pipeline.set_initiator(role='guest', party_id=guest)
  46. # set participants information
  47. pipeline.set_roles(guest=guest, host=hosts)
  48. # define Reader components to read in data
  49. reader_0 = Reader(name="reader_0")
  50. # configure Reader for guest
  51. reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
  52. # configure Reader for host
  53. reader_0.get_party_instance(role='host', party_id=hosts).component_param(table=host_train_data)
  54. data_transform_0 = DataTransform(name="data_transform_0", output_format='dense', missing_fill=False)
  55. # get DataTransform party instance of guest
  56. data_transform_0_guest_party_instance = data_transform_0.get_party_instance(role='guest', party_id=guest)
  57. # configure DataTransform for guest
  58. data_transform_0_guest_party_instance.component_param(with_label=True, label_name="label")
  59. # get and configure DataTransform party instance of host
  60. data_transform_0.get_party_instance(role='host', party_id=hosts).component_param(with_label=False)
  61. # define Intersection components
  62. intersection_0 = Intersection(name="intersection_0")
  63. pipeline.add_component(reader_0)
  64. pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
  65. pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
  66. statistic_param = {
  67. "name": "statistic_0",
  68. "statistics": ["95%", "coefficient_of_variance", "stddev"],
  69. "column_indexes": [1, 2],
  70. "column_names": ["x3"]
  71. }
  72. statistic_0 = DataStatistics(**statistic_param)
  73. pipeline.add_component(statistic_0, data=Data(data=intersection_0.output.data))
  74. pipeline.compile()
  75. # fit model
  76. pipeline.fit()
  77. # query component summary
  78. prettify(pipeline.get_component("statistic_0").get_summary())
  79. if __name__ == "__main__":
  80. parser = argparse.ArgumentParser("PIPELINE DEMO")
  81. parser.add_argument("-config", type=str,
  82. help="config file")
  83. args = parser.parse_args()
  84. if args.config is not None:
  85. main(args.config)
  86. else:
  87. main()