def test_valid_url(self):
"""
Tests for valid data URL
"""
file = 'listing.csv.gz'
with self.assertRaises(ValueError):
gd.download_dataset(c.DATASET_PROPERTIES, file)
def test_input_dictionary(self):
"""
Tests input dictionary has required keys
"""
dataset_properties = {'d': '2019-04-15', 'c': 'Seattle', 's': 'WA'}
file = c.LISTINGS_DATA
with self.assertRaises(ValueError):
gd.download_dataset(dataset_properties, file)
def test_gd_output_shape(self):
"""
Tests output dataframe has both rows and columns
"""
data = gd.download_dataset(c.DATASET_PROPERTIES, c.REVIEWS_DATA)
rows, cols = list(data.shape)
self.assertTrue(rows > 0 and cols > 0)
"""
This module runs unit test for training the machine learning model.
"""
import unittest
import zillowbnb.test.submodule_path
import constants as co
import convert_to_matrix as cm
import get_data as gd
import get_cleaned_listings as gcl
import train_model as tm
DATA = gd.download_dataset(co.DATASET_PROPERTIES, co.LISTINGS_DATA)
DATAFRAME = gcl.get_listings_dataframe(DATA, co.LISTING_COLUMNS)
X_VAR, Y_VAR = cm.to_matrix(DATAFRAME, co.LISTING_COLUMNS)
class TrainModelTest(unittest.TestCase):
"""
This class runs unit tests for the train_model module.
"""
def test_input_size(self):
"""
Tests that train_model will not run if sizes of x_var and y_var
do not match.
:params self:
:returns boolean:
"""
This module runs unit tests for ZillowBnb
"""
import unittest
import zillowbnb.test.submodule_path
import constants
import get_data
import sentiment
DATA = get_data.download_dataset(constants.DATASET_PROPERTIES,
constants.REVIEWS_DATA)
class SentimentTest(unittest.TestCase):
"""
This class runs unit tests for the sentiment submodule
"""
def test_polarity_check_dtype(self):
"""
Tests that polarity will not run if datatypes are incorrect
:param self:
:return boolean:
"""
with self.assertRaises(ValueError):
sentiment.polarity(DATA, 1)
def test_polarity_check_column(self):
"""
Tests that polarity will not run if column isnt in dataframe
"""
This module runs unit tests for ZillowBnb
"""
import unittest
import zillowbnb.test.submodule_path
import constants
import get_data
import get_cleaned_listings
DATA = get_data.download_dataset(constants.DATASET_PROPERTIES,
constants.LISTINGS_DATA)
class ListingsTest(unittest.TestCase):
"""
This class runs unit tests for the get_cleaned_listings submodule
"""
def test_clean_listings_row(self):
"""
Tests cleaned_listings data for more than one row
:param self:
:return boolean:
"""
listings = get_cleaned_listings.get_listings_dataframe(DATA, constants.LISTING_COLUMNS)
self.assertTrue(listings.shape[0] >= 1)
def test_listings_col(self):
"""
Test cleaned_listings data has all 40 columns
import os
import pandas as pd
import constants as c
import get_data as gd
import price_prediction as pp
# Uncomment below if regenerating all datasets
# from submodule import convert_to_matrix as ctm, get_calendar_summary as gcs
# from submodule import get_cleaned_listings as gcl, sentiment as s, train_model as tml
# Set data folder path
DATA_FOLDER = os.path.abspath('../../data') + '/'
# Import the datafiles
CALENDAR = gd.download_dataset(c.DATASET_PROPERTIES, c.CALENDAR_DATA)
LISTINGS = gd.download_dataset(c.DATASET_PROPERTIES, c.LISTINGS_DATA)
REVIEWS = gd.download_dataset(c.DATASET_PROPERTIES, c.REVIEWS_DATA)
# Clean the calendar dataset
# Run:
# 1. CALENDAR_DF = gcs.create_calendar_price_averages(CALENDAR)
CALENDAR_DF = c.CALENDAR_CSV
# Run sentiment analysis on review datasets
# Run: (Can take a few hours)
# 1. SENTIMENT_SCORES = s.polarity(REVIEWS, 'comments')
# 2. SENTIMENT_DF = s.summarize_sentiment(SENTIMENT_SCORES, ['listing_id'], 'compound')
SENTIMENT_DF = c.SENTIMENT_CSV
# Clean the listings datasets
"""
This module runs unit tests for get_calendar_summary
"""
import unittest
import datetime
import zillowbnb.test.submodule_path
import get_data
import get_calendar_summary
import constants
DATA = get_data.download_dataset(constants.DATASET_PROPERTIES,
constants.CALENDAR_DATA)
TEST = get_calendar_summary.create_calendar_price_averages(DATA)
# used to test error throwing (can not be a string or datetime.datetime)
INCORRCT_DATA_TYPE = 9
TEST_DATE = datetime.date(2019, 4, 23)
TEST_CURRENCY_STRING = "$1,000.00"
class CalendarTest(unittest.TestCase):
"""
This class runs unit tests for ZillowBnb
"""
def test_calendar_no_na(self):
"""
Tests imported calendar data has no empty values
:param self:
:return boolean: