crawles

fc = tf.feature_column
CATEGORICAL_COLUMNS = ['sex', 'n_siblings_spouses', 'parch', 'class', 'deck', 
                       'embark_town', 'alone']
NUMERIC_COLUMNS = ['age', 'fare']
  
def one_hot_cat_column(feature_name, vocab):
  return tf.feature_column.indicator_column(
      tf.feature_column.categorical_column_with_vocabulary_list(feature_name,
                                                                vocab))
feature_columns = []

crawles

import numpy as np
import pandas as pd
import tensorflow as tf
tf.enable_eager_execution()

# Load dataset.
dftrain = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/train.csv')
dfeval = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/eval.csv')
y_train = dftrain.pop('survived')
y_eval = dfeval.pop('survived')

crawles

## TL;DR:
# Train model.
est = tf.estimator.BoostedTreesClassifier(feature_columns, n_batches_per_layer)
est.train(train_input_fn)
# Per instance model interpretability:
pred_dict = est.experimental_predict_with_explanations(pred_input_fn)
# Global gain-based feature importances:
importances = est.experimental_feature_importances()

crawles

	survived	pclass	sex	age	sibsp	parch	fare	embarked	class	who	adult_male	deck	embark_town	alive	alone
0	0	3	male	22.0	1	0	7.25	S	Third	man	True		Southampton	no	False
1	1	1	female	38.0	1	0	71.2833	C	First	woman	False	C	Cherbourg	yes	False
2	1	3	female	26.0	0	0	7.925	S	Third	woman	False		Southampton	yes	True
3	1	1	female	35.0	1	0	53.1	S	First	woman	False	C	Southampton	yes	False
4	0	3	male	35.0	0	0	8.05	S	Third	man	True		Southampton	no	True
5	0	3	male		0	0	8.4583	Q	Third	man	True		Queenstown	no	True
6	0	1	male	54.0	0	0	51.8625	S	First	man	True	E	Southampton	no	True
7	0	3	male	2.0	3	1	21.075	S	Third	child	False		Southampton	no	False
8	1	3	female	27.0	0	2	11.1333	S	Third	woman	False		Southampton	yes	False

crawles

crawles

def _numeric_column_normalized(column_name, normalizer_fn):
    return tf.feature_column.numeric_column(column_name, normalizer_fn=normalizer_fn)

def _make_zscaler(mean, std):
    def zscaler(col):
        return (col - mean)/std
    return zscaler

# Define your feature columns
def create_feature_cols(features, use_normalization):

crawles

def get_normalization_parameters(traindf, features):
    """Get the normalization parameters (E.g., mean, std) for traindf for 
    features. We will use these parameters for training, eval, and serving."""

    def _z_score_params(column):
        mean = traindf[column].mean()
        std = traindf[column].std()
        return {'mean': mean, 'std': std}

    normalization_parameters = {}

crawles

crawles

crawles

def random_null(col, keep=0.9):
    import random
    if keep < random.random():
        return None
    return col
df.applymap(lambda x: random_null(x))