def handles_predicting_cases_it_has_not_seen_before(self):
user_tz = ['Eastern (US & Canada)', 'Pacific (US & Canada)']
l1_time = [9, 9]
l1_day = [0, 1]
l2_time = [9, 9]
l2_day = [3, 4]
l3_time = [None, 9]
l3_day = [None, 5]
schedule_type = [2, 3]
training_data = pd.DataFrame({ 'user_tz': user_tz,
'l1_time': l1_time,
'l1_day': l1_day,
'l2_time': l2_time,
'l2_day': l2_day,
'l3_time': l3_time,
'l3_day': l3_day,
'schedule_type': schedule_type
})
business_forecast = [{'schedule_type': 1,
'user_tz': 'Eastern (US & Canada)',
'frequency': 1 },
{'schedule_type': 4,
'user_tz': 'Pacific (US & Canada)',
'frequency': 2 }]
model = GeneralProbModel()
model.fit(training_data)
p = model.predict(business_forecast)
expect(True).to.equal(True)
def should_take_into_account_conditional_probability(self):
user_tz = ['Eastern (US & Canada)', 'Pacific (US & Canada)']
l1_time = [9, 9]
l1_day = [0, 1]
l2_time = [9, 9]
l2_day = [3, 4]
l3_time = [None, 9]
l3_day = [None, 5]
schedule_type = [2, 3]
training_data = pd.DataFrame({ 'user_tz': user_tz,
'l1_time': l1_time,
'l1_day': l1_day,
'l2_time': l2_time,
'l2_day': l2_day,
'l3_time': l3_time,
'l3_day': l3_day,
'schedule_type': schedule_type
})
business_forecast = [{'schedule_type': 2,
'user_tz': 'Eastern (US & Canada)',
'frequency': 1 },
{'schedule_type': 3,
'user_tz': 'Pacific (US & Canada)',
'frequency': 2 }]
model = GeneralProbModel()
model.fit(training_data)
p = model.predict(business_forecast)
expect(True).to.equal(True)
def handles_predicting_cases_it_has_not_seen_before(self):
user_tz = ['Eastern (US & Canada)', 'Pacific (US & Canada)']
l1_time = [9, 9]
l1_day = [0, 1]
l2_time = [9, 9]
l2_day = [3, 4]
l3_time = [None, 9]
l3_day = [None, 5]
schedule_type = [2, 3]
training_data = pd.DataFrame({
'user_tz': user_tz,
'l1_time': l1_time,
'l1_day': l1_day,
'l2_time': l2_time,
'l2_day': l2_day,
'l3_time': l3_time,
'l3_day': l3_day,
'schedule_type': schedule_type
})
business_forecast = [{
'schedule_type': 1,
'user_tz': 'Eastern (US & Canada)',
'frequency': 1
}, {
'schedule_type': 4,
'user_tz': 'Pacific (US & Canada)',
'frequency': 2
}]
model = GeneralProbModel()
model.fit(training_data)
p = model.predict(business_forecast)
expect(True).to.equal(True)
def should_take_into_account_conditional_probability(self):
user_tz = ['Eastern (US & Canada)', 'Pacific (US & Canada)']
l1_time = [9, 9]
l1_day = [0, 1]
l2_time = [9, 9]
l2_day = [3, 4]
l3_time = [None, 9]
l3_day = [None, 5]
schedule_type = [2, 3]
training_data = pd.DataFrame({
'user_tz': user_tz,
'l1_time': l1_time,
'l1_day': l1_day,
'l2_time': l2_time,
'l2_day': l2_day,
'l3_time': l3_time,
'l3_day': l3_day,
'schedule_type': schedule_type
})
business_forecast = [{
'schedule_type': 2,
'user_tz': 'Eastern (US & Canada)',
'frequency': 1
}, {
'schedule_type': 3,
'user_tz': 'Pacific (US & Canada)',
'frequency': 2
}]
model = GeneralProbModel()
model.fit(training_data)
p = model.predict(business_forecast)
expect(True).to.equal(True)