Poisonous Mushrooms

by: Chris Cave

Feyn version: 2.1+

Last updated: 27/09/2021

Importing the dataset

Here we use the QLattice to predict whether a mushroom is poisonous or not. You can find this dataset and further descriptions of the features on UCI Machine Learning Repository.

This shows how use the QLattice when there are a lot of categorical variables.

import pandas as pd
import numpy as np
import feyn

from sklearn.model_selection import train_test_split

data = pd.read_csv("../data/mushrooms.csv")

# Replace the categories p and e with 0 and 1 respectively
data["class"] = data['class'].replace({"p": 0, "e": 1})

data.head()

	class	cap-shape	cap-surface	cap-color	bruises	odor	gill-attachment	gill-spacing	gill-size	gill-color	...	stalk-surface-below-ring	stalk-color-above-ring	stalk-color-below-ring	veil-type	veil-color	ring-number	ring-type	spore-print-color	population	habitat
0	0	x	s	n	t	p	f	c	n	k	...	s	w	w	p	w	o	p	k	s	u
1	1	x	s	y	t	a	f	c	b	k	...	s	w	w	p	w	o	p	n	n	g
2	1	b	s	w	t	l	f	c	b	n	...	s	w	w	p	w	o	p	n	n	m
3	0	x	y	w	t	p	f	c	n	n	...	s	w	w	p	w	o	p	k	s	u
4	1	x	s	g	f	n	f	w	b	k	...	s	w	w	p	w	o	e	n	a	g

5 rows × 23 columns

Connect to the QLattice

We declare which variables are categorical to the QLattice using the stypes (semantic types) parameter.

random_state = 42

train, test = train_test_split(data, test_size=0.7, random_state = random_state)

stypes = {}

for col in data.columns:
    if data[col].dtype == "object":
        stypes[col] = "c"

ql = feyn.connect_qlattice()
ql.reset(42)

Use auto_run to obtain fitted models

models = ql.auto_run(train, "class", kind="classification", n_epochs=10, stypes=stypes, max_complexity=5)

best = models[0]

Summary plot to evaluate model

best.plot(train, test)

Training Metrics

Test