metadata.py

#!/usr/bin/env python

# Copyright 2017 Google Inc. 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.

# ******************************************************************************
# YOU NEED TO MODIFY THE FOLLOWING METADATA TO ADAPT THE TEMPLATE TO YOUR DATA
# ******************************************************************************

# Task type can be either 'classification' or 'regression'
# This is based on the target feature in the dataset.
TASK_TYPE = ''  # classification | regression

# List of all the columns (header) present in the input data file(s).
# Used for parsing the input data.
COLUMN_NAMES = []

# List of the columns expected during serving (which is probably different to
# the header of the training data).
SERVING_COLUMN_NAMES = []

# List of the default values of all the columns present in the input data.
# This helps decoding the data types of the columns.
DEFAULTS = []

# Dictionary of the feature names of type int or float. In the dictionary,
# the key is the feature name, and the value is another dictionary includes
# the mean and the variance of the numeric features.
# E.g. {feature_1: {mean: 0, variance:1}, feature_2: {mean: 10, variance:3}}
# The value can be set to None if you don't want to not normalize.
NUMERIC_FEATURE_NAMES_WITH_STATS = {}

# Dictionary of feature names with int values, but to be treated as
# categorical features. In the dictionary, the key is the feature name,
# and the value is the num_buckets (count of distinct values).
CATEGORICAL_FEATURE_NAMES_WITH_IDENTITY = {}

# Dictionary of categorical features with few nominal values. In the dictionary,
# the key is the feature name, and the value is the list of feature vocabulary.
CATEGORICAL_FEATURE_NAMES_WITH_VOCABULARY = {}

# Dictionary of categorical features with many values. In the dictionary,
# the key is the feature name, and the value is the number of buckets.
CATEGORICAL_FEATURE_NAMES_WITH_HASH_BUCKET = {}

# Column includes the relative weight of each record.
WEIGHT_COLUMN_NAME = None

# Target feature name (response or class variable).
TARGET_NAME = ''

# List of the class values (labels) in a classification dataset.
TARGET_LABELS = []