tensorflow + dalex = :)

For the blog: https://medium.com/responsibleml/tensorflow-dalex-929ca520d3d9

introduction to the topic: Explanatory Model Analysis: Explore, Explain, and Examine Predictive Models

import warnings
warnings.filterwarnings('ignore')

read data

import pandas as pd
pd.__version__

'1.2.4'

data = pd.read_csv("https://raw.githubusercontent.com/pbiecek/xai-happiness/main/happiness.csv", index_col=0)
data.head()

	score	gdp_per_capita	social_support	healthy_life_expectancy	freedom_to_make_life_choices	generosity	perceptions_of_corruption
Afghanistan	3.203	0.350	0.517	0.361	0.000	0.158	0.025
Albania	4.719	0.947	0.848	0.874	0.383	0.178	0.027
Algeria	5.211	1.002	1.160	0.785	0.086	0.073	0.114
Argentina	6.086	1.092	1.432	0.881	0.471	0.066	0.050
Armenia	4.559	0.850	1.055	0.815	0.283	0.095	0.064

X, y = data.drop('score', axis=1), data.score
n, p = X.shape

create a model

import tensorflow as tf
tf.__version__

'2.6.0'

tf.random.set_seed(11)

normalizer  = tf.keras.layers.experimental.preprocessing.Normalization(input_shape=[p,])
normalizer.adapt(X.to_numpy())

model = tf.keras.Sequential([
    normalizer,
    tf.keras.Input(shape=(p,)),
    tf.keras.layers.Dense(p*2, activation='relu'),
    tf.keras.layers.Dense(p*3, activation='relu'),
    tf.keras.layers.Dense(p*2, activation='relu'),
    tf.keras.layers.Dense(p, activation='relu'),
    tf.keras.layers.Dense(1, activation='linear')
])

model.compile(
    optimizer=tf.keras.optimizers.Adam(0.001),
    loss=tf.keras.losses.mae
)

model.fit(X, y, batch_size=int(n/10), epochs=2000, verbose=False)

<keras.callbacks.History at 0x135feba79a0>

explain the model

Explainer initialization communicates useful information

import dalex as dx
dx.__version__

'1.4.0'

explainer = dx.Explainer(model, X, y, label='happiness')

Preparation of a new explainer is initiated

  -> data              : 156 rows 6 cols
  -> target variable   : Parameter 'y' was a pandas.Series. Converted to a numpy.ndarray.
  -> target variable   : 156 values
  -> model_class       : keras.engine.sequential.Sequential (default)
  -> label             : happiness
  -> predict function  : <function yhat_tf_regression at 0x00000135804639D0> will be used (default)
  -> predict function  : Accepts pandas.DataFrame and numpy.ndarray.
  -> predicted values  : min = 2.86, mean = 5.41, max = 7.96
  -> model type        : regression will be used (default)
  -> residual function : difference between y and yhat (default)
  -> residuals         : min = -0.609, mean = -0.00403, max = 1.09
  -> model_info        : package keras

A new explainer has been created!

model level explanations

firstly, assess the model fit to training data

explainer.model_performance()

	mse	rmse	r2	mae	mad
happiness	0.025378	0.159304	0.979386	0.077048	0.032524

which features are the most important?

explainer.model_parts().plot()