LogML Logo

User Manual

  • Quick Start
  • About LogML
  • Configuration Overview
  • Running LogML
  • Output Artifacts
  • How-to Topics

Internals

  • LogML Pipeline
  • Data Preprocessing
  • Feature Importance Analysis
  • Exploratory Data Analysis
  • Survival Analysis
  • Registry

Development

  • Dev How-to
  • Running - Advanced Topics
  • Dev Environment
  • Code Reference
    • logml
      • logml.analysis
      • logml.common
      • logml.configuration
      • logml.data
      • logml.eda
      • logml.feature_importance
      • logml.log
      • logml.main
      • logml.metrics
      • logml.model_search
        • logml.model_search.common
        • logml.model_search.hp_optimizer
        • logml.model_search.provider
        • logml.model_search.selection
        • logml.model_search.shap
      • logml.models
      • logml.report
      • logml.serialization
      • logml.stopwatch
      • logml.survival_analysis
  • Release Notes
LogML
  • »
  • Code Reference »
  • logml »
  • logml.model_search
  • View page source

logml.model_search

Classes

ModelSearchOutputStructure(*args, **kwargs)

Defines output structure for Model Search endpoint.

ModelSearchResultOutputStructure(artifacts_root)

Files specific for (model, dataset) pair.

class logml.model_search.ModelSearchResultOutputStructure(artifacts_root: pathlib.Path)

Bases: object

Files specific for (model, dataset) pair.

MODEL_EVALUATION_FILENAME: str = 'evaluation.pickle'
MODEL_INSTANCE_FILENAME: str = 'model.pickle'
MODEL_HPO_FILENAME: str = 'hpo.pickle'
property model_evaluation_file_path

Model evaluation result file path.

property model_instance_file_path

Model instance file path.

property model_hpo_result_file_path

Model hpo result file path.

class logml.model_search.ModelSearchOutputStructure(*args, **kwargs)

Bases: logml.common.OutputStructure

Defines output structure for Model Search endpoint. Structure:

experiment_root/…
  • /model_search/{problem_id}/
    • selection_result.json * data for model selection result

      (list of good models)

    • /{model_id}__{dataset_hash} * data for specific model/ds:
      • evaluation.pickle * model params and metrics

      • hpo.pickle * hyperparams data

      • model.pickle * tuned model for later use

MODEL_SEARCH: str = 'model_search'
SUMMARY_FILENAME: str = 'selection_result.json'
DELIMETER = '__'
get_summary_file_path()

Get path to summary file

get_model_folder_path(model_id: str, dataset_name: str) → pathlib.Path

Returns path to a folder for specific model/dataset data.

list_available_models(dataset_name: str) → List[str]

For a given dataset hash returns a list of models with existing evaluation results.

get_model_structure(model_id: str, dataset_name: str) → logml.model_search.ModelSearchResultOutputStructure

Returns path to a folder for specific model/dataset data.

artifacts_exist() → bool

Check whether summary artifact exists.

Modules

logml.model_search.common

logml.model_search.hp_optimizer

logml.model_search.provider

Models provider interface

logml.model_search.selection

Models training and selection

logml.model_search.shap

Previous Next

© Copyright 2022, LogML Team.

Built with Sphinx using a theme provided by Read the Docs.