def test_experiment_run_and_compare():
for data_path in datasets:
accuracies = {}
learner_types = ['prior_probability', 'decision_tree']
for learner_type in learner_types:
accuracies[learner_type] = run(data_path, learner_type, 1.0)[1]
if 'candy' in data_path or 'ivy' in data_path:
assert (accuracies['decision_tree'] >
accuracies['prior_probability'])
def test_experiment_run_prior_probability():
accuracies = {}
for data_path in datasets:
learner_type = 'prior_probability'
confusion_matrix, accuracy, precision, recall, f1_measure = (run(
data_path, learner_type, 1.0))
accuracies[data_path] = accuracy
dataset = xp_dataset_name('ivy-league.csv')
assert (accuracies[dataset] > .2)
def test_experiment_run_decision_tree():
accuracies = {}
for data_path in datasets:
learner_type = 'decision_tree'
confusion_matrix, accuracy, precision, recall, f1_measure = (run(
data_path, learner_type, 1.0))
accuracies[data_path] = accuracy
dataset = [dataset for dataset in datasets
if 'ivy-league.csv' in dataset][0]
assert (accuracies[dataset] > .7)
def test_experiment_run_decision_tree():
accuracies = {}
for data_path in datasets:
learner_type = 'decision_tree'
confusion_matrix, accuracy, precision, recall, f1_measure = (run(
data_path, learner_type, 1.0))
accuracies[data_path] = accuracy
accuracy_goals = {
xp_dataset_name('ivy-league.csv'): .95,
xp_dataset_name('xor.csv'): 1.0,
xp_dataset_name('candy-data.csv'): .75,
xp_dataset_name('majority-rule.csv'): 1.0
}
for key in accuracy_goals:
assert (accuracies[key] >= accuracy_goals[key])
from code import run # Adelson Jhonata Silva de Sousa – 18/0114913 # José Fortes Neto - 16/0128331 # pip install pandas # pip install sklearn # python main.py run()
def __run_code(self):
# +------------- YOUR CODE HERE -------------+
import code
code.run(echo=self.screen.print, **self.config)
# +------------------------------------------+
self.__post_run_update()
# necessary imports
import matplotlib.pyplot as plt
import pandas as pd
import code
# read csv
data = pd.read_csv('data.csv')
# extract from csv
x = list(data[data.columns[0]])
y = list(data[data.columns[1]])
# obtains best value of m and b for line
m, b = code.run()
# plot all the points of data using scatter plot
plt.scatter(x, y)
line = [m * i + b for i in x]
# plot the linear regression line
plt.plot(x, line)
plt.title("Linear Regression using gradient descent")
plt.plot()
plt.savefig('figure.jpg')
import json
from code import run
config_file = open('config.json')
config = json.load(config_file)
config_file.close()
run(config)
