11. Create Calendar Window Aggregates from Time Series

Creating Features with Monthly Calendar Windows¶

In this step, we define by aggregating monthly records from the CREDIT_CARD_MONTHLY_BALANCE and INSTALLMENTS_PAYMENTS tables within 6-month, 12-month and 24-month calendar windows.

Calendar aggregations windows are required for Time Series while for event table they are recommended only when the event is a monthly regular event and the feature job setting is a CRON setting.

Activate catalog¶

See SDK reference for catalogs

In [1]:

import featurebyte as fb

# Set your profile to the tutorial environment
fb.use_profile("tutorial")

catalog_name = "Loan Applications Dataset SDK Tutorial"
catalog = fb.Catalog.activate(catalog_name)

import featurebyte as fb # Set your profile to the tutorial environment fb.use_profile("tutorial") catalog_name = "Loan Applications Dataset SDK Tutorial" catalog = fb.Catalog.activate(catalog_name)

14:10:28 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:10:28 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.
14:10:28 | INFO     | Using profile: tutorial
INFO    :featurebyte:Using profile: tutorial
14:10:29 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
INFO    :featurebyte:Using configuration file at: /Users/gxav/.featurebyte/config.yaml
14:10:29 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
INFO    :featurebyte:Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
14:10:29 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:10:29 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.
14:10:29 | INFO     | Catalog activated: Loan Applications Dataset SDK Tutorial
INFO    :featurebyte.api.catalog:Catalog activated: Loan Applications Dataset SDK Tutorial

Get view from table¶

See SDK reference for views

In [2]:

# Get view from CREDIT_CARD_MONTHLY_BALANCE time series table.
credit_card_monthly_balance_view = catalog.get_view("CREDIT_CARD_MONTHLY_BALANCE")

# Get view from CREDIT_CARD_MONTHLY_BALANCE time series table. credit_card_monthly_balance_view = catalog.get_view("CREDIT_CARD_MONTHLY_BALANCE")

In [3]:

# Get view from INSTALLMENTS_PAYMENTS event table.
installments_payments_view = catalog.get_view("INSTALLMENTS_PAYMENTS")

# Get view from INSTALLMENTS_PAYMENTS event table. installments_payments_view = catalog.get_view("INSTALLMENTS_PAYMENTS")

In [4]:

# Get view from PREVIOUS_APPLICATION event table.
previous_application_view = catalog.get_view("PREVIOUS_APPLICATION")

# Get view from PREVIOUS_APPLICATION event table. previous_application_view = catalog.get_view("PREVIOUS_APPLICATION")

Create ratio column¶

See SDK reference for view columns

In [5]:

previous_application_view["AMT_ANNUITY To AMT_CREDIT"] = (
    previous_application_view["AMT_ANNUITY"] / previous_application_view["AMT_CREDIT"]
)

previous_application_view["AMT_ANNUITY To AMT_CREDIT"] = ( previous_application_view["AMT_ANNUITY"] / previous_application_view["AMT_CREDIT"] )

Derive new column¶

In [6]:

credit_card_monthly_balance_view["Available Credit"] = (
    credit_card_monthly_balance_view["AMT_CREDIT_LIMIT_ACTUAL"] - credit_card_monthly_balance_view["AMT_BALANCE"]
)

credit_card_monthly_balance_view["Available Credit"] = ( credit_card_monthly_balance_view["AMT_CREDIT_LIMIT_ACTUAL"] - credit_card_monthly_balance_view["AMT_BALANCE"] )

In [7]:

installments_payments_view["Days_Difference_Actual_vs_Scheduled"] = (
    installments_payments_view["actual_installment_date"] - installments_payments_view["scheduled_installment_date"]
).dt.day

installments_payments_view["Days_Difference_Actual_vs_Scheduled"] = ( installments_payments_view["actual_installment_date"] - installments_payments_view["scheduled_installment_date"] ).dt.day

Join views¶

In [8]:

# Join PREVIOUS_APPLICATION view to INSTALLMENTS_PAYMENTS view.
installments_payments_view = installments_payments_view.join(previous_application_view, rprefix="PriorApplication_")

# Join PREVIOUS_APPLICATION view to INSTALLMENTS_PAYMENTS view. installments_payments_view = installments_payments_view.join(previous_application_view, rprefix="PriorApplication_")

Do window aggregation from CREDIT_CARD_MONTHLY_BALANCE¶

See SDK reference for features

See SDK reference to groupby a view

See SDK reference to do aggregation over time

See SDK reference for feature group

See list of aggregation operations

In [9]:

# Group CREDIT_CARD_MONTHLY_BALANCE view by Client entity (ClientID).
credit_card_monthly_balance_view_by_client = credit_card_monthly_balance_view.groupby(["ClientID"])

# Group CREDIT_CARD_MONTHLY_BALANCE view by Client entity (ClientID). credit_card_monthly_balance_view_by_client = credit_card_monthly_balance_view.groupby(["ClientID"])

In [10]:

# Get Std of CNT_DRAWINGS_CURRENT for the Client over time.
client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo = (
    credit_card_monthly_balance_view_by_client.aggregate_over(
        "CNT_DRAWINGS_ATM_CURRENT",
        method="std",
        feature_names=["CLIENT_Std_of_Credit_card_monthly_balance_records_CNT_DRAWINGS_ATM_CURRENTs_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Std_of_Credit_card_monthly_balance_records_CNT_DRAWINGS_ATM_CURRENTs_24cMo"]
)

# Get Std of CNT_DRAWINGS_CURRENT for the Client over time. client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo = ( credit_card_monthly_balance_view_by_client.aggregate_over( "CNT_DRAWINGS_ATM_CURRENT", method="std", feature_names=["CLIENT_Std_of_Credit_card_monthly_balance_records_CNT_DRAWINGS_ATM_CURRENTs_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Std_of_Credit_card_monthly_balance_records_CNT_DRAWINGS_ATM_CURRENTs_24cMo"] )

In [11]:

# Get Min of Available Credit for the Client over time.
client_min_of_credit_card_monthly_balance_records_available_credits_6cmo = (
    credit_card_monthly_balance_view_by_client.aggregate_over(
        "Available Credit",
        method="min",
        feature_names=["CLIENT_Min_of_Credit_card_monthly_balance_records_Available_Credits_6cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=6)],
    )["CLIENT_Min_of_Credit_card_monthly_balance_records_Available_Credits_6cMo"]
)

# Get Min of Available Credit for the Client over time. client_min_of_credit_card_monthly_balance_records_available_credits_6cmo = ( credit_card_monthly_balance_view_by_client.aggregate_over( "Available Credit", method="min", feature_names=["CLIENT_Min_of_Credit_card_monthly_balance_records_Available_Credits_6cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=6)], )["CLIENT_Min_of_Credit_card_monthly_balance_records_Available_Credits_6cMo"] )

Do window aggregation from INSTALLMENTS_PAYMENTS¶

See SDK reference for features

See SDK reference to groupby a view

See SDK reference to do aggregation over time

See SDK reference for feature group

See list of aggregation operations

In [12]:

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID).
installments_payments_view_by_client = installments_payments_view.groupby(["PriorApplication_ClientID"])

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID). installments_payments_view_by_client = installments_payments_view.groupby(["PriorApplication_ClientID"])

In [13]:

# Get Min of AMT_PAYMENT for the Client over time.
client_min_of_installments_amt_payments_24cmo = installments_payments_view_by_client.aggregate_over(
    "AMT_PAYMENT",
    method="min",
    feature_names=["CLIENT_Min_of_Installments_AMT_PAYMENTs_24cMo"],
    windows=[fb.CalendarWindow(unit="MONTH", size=24)],
)["CLIENT_Min_of_Installments_AMT_PAYMENTs_24cMo"]

# Get Min of AMT_PAYMENT for the Client over time. client_min_of_installments_amt_payments_24cmo = installments_payments_view_by_client.aggregate_over( "AMT_PAYMENT", method="min", feature_names=["CLIENT_Min_of_Installments_AMT_PAYMENTs_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Min_of_Installments_AMT_PAYMENTs_24cMo"]

In [14]:

# Get Max of Days_Difference_Actual_vs_Scheduled for the Client over time.
client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo = (
    installments_payments_view_by_client.aggregate_over(
        "Days_Difference_Actual_vs_Scheduled",
        method="max",
        feature_names=["CLIENT_Max_of_Installments_Days_Difference_Actual_vs_Scheduleds_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Max_of_Installments_Days_Difference_Actual_vs_Scheduleds_24cMo"]
)

# Get Max of Days_Difference_Actual_vs_Scheduled for the Client over time. client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo = ( installments_payments_view_by_client.aggregate_over( "Days_Difference_Actual_vs_Scheduled", method="max", feature_names=["CLIENT_Max_of_Installments_Days_Difference_Actual_vs_Scheduleds_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Max_of_Installments_Days_Difference_Actual_vs_Scheduleds_24cMo"] )

In [15]:

# Get Max of PriorApplication_AMT_ANNUITY To AMT_CREDIT for the Client over time.
client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo = (
    installments_payments_view_by_client.aggregate_over(
        "PriorApplication_AMT_ANNUITY To AMT_CREDIT",
        method="max",
        feature_names=["CLIENT_Max_of_Installments_PriorApplication_AMT_ANNUITY_To_AMT_CREDITs_6cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=6)],
    )["CLIENT_Max_of_Installments_PriorApplication_AMT_ANNUITY_To_AMT_CREDITs_6cMo"]
)

# Get Max of PriorApplication_AMT_ANNUITY To AMT_CREDIT for the Client over time. client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo = ( installments_payments_view_by_client.aggregate_over( "PriorApplication_AMT_ANNUITY To AMT_CREDIT", method="max", feature_names=["CLIENT_Max_of_Installments_PriorApplication_AMT_ANNUITY_To_AMT_CREDITs_6cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=6)], )["CLIENT_Max_of_Installments_PriorApplication_AMT_ANNUITY_To_AMT_CREDITs_6cMo"] )

In [16]:

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different
# INSTALLMENT_STATUSs.
installments_payments_view_by_client_across_installment_statuss = installments_payments_view.groupby(
    ["PriorApplication_ClientID"], category="INSTALLMENT_STATUS"
)

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different # INSTALLMENT_STATUSs. installments_payments_view_by_client_across_installment_statuss = installments_payments_view.groupby( ["PriorApplication_ClientID"], category="INSTALLMENT_STATUS" )

In [17]:

# Distribution of the total AMT_PAYMENTs of Installments, segmented by INSTALLMENT_STATUS for the
# Client over time.
client_installments_amt_payments_by_installment_status_6cmo = (
    installments_payments_view_by_client_across_installment_statuss.aggregate_over(
        "AMT_PAYMENT",
        method="sum",
        feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_6cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=6)],
    )["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_6cMo"]
)

# Distribution of the total AMT_PAYMENTs of Installments, segmented by INSTALLMENT_STATUS for the # Client over time. client_installments_amt_payments_by_installment_status_6cmo = ( installments_payments_view_by_client_across_installment_statuss.aggregate_over( "AMT_PAYMENT", method="sum", feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_6cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=6)], )["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_6cMo"] )

In [18]:

# Distribution of the total AMT_PAYMENTs of Installments, segmented by INSTALLMENT_STATUS for the
# Client over time.
client_installments_amt_payments_by_installment_status_12cmo = (
    installments_payments_view_by_client_across_installment_statuss.aggregate_over(
        "AMT_PAYMENT",
        method="sum",
        feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_12cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=12)],
    )["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_12cMo"]
)

# Distribution of the total AMT_PAYMENTs of Installments, segmented by INSTALLMENT_STATUS for the # Client over time. client_installments_amt_payments_by_installment_status_12cmo = ( installments_payments_view_by_client_across_installment_statuss.aggregate_over( "AMT_PAYMENT", method="sum", feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_12cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=12)], )["CLIENT_Installments_AMT_PAYMENTs_by_INSTALLMENT_STATUS_12cMo"] )

In [19]:

# Count Installments across different INSTALLMENT_STATUSs for the Client over time.
client_count_of_installments_by_installment_status_24cmo = (
    installments_payments_view_by_client_across_installment_statuss.aggregate_over(
        None,
        method="count",
        feature_names=["CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo"]
)

# Count Installments across different INSTALLMENT_STATUSs for the Client over time. client_count_of_installments_by_installment_status_24cmo = ( installments_payments_view_by_client_across_installment_statuss.aggregate_over( None, method="count", feature_names=["CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo"] )

In [20]:

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different
# PriorApplication_PAYMENT_TYPEs.
installments_payments_view_by_client_across_priorapplication_payment_types = installments_payments_view.groupby(
    ["PriorApplication_ClientID"], category="PriorApplication_PAYMENT_TYPE"
)

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different # PriorApplication_PAYMENT_TYPEs. installments_payments_view_by_client_across_priorapplication_payment_types = installments_payments_view.groupby( ["PriorApplication_ClientID"], category="PriorApplication_PAYMENT_TYPE" )

In [21]:

# Distribution of the total AMT_PAYMENTs of Installments, segmented by
# PriorApplication_PAYMENT_TYPE for the Client over time.
client_installments_amt_payments_by_priorapplication_payment_type_24cmo = (
    installments_payments_view_by_client_across_priorapplication_payment_types.aggregate_over(
        "AMT_PAYMENT",
        method="sum",
        feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_PAYMENT_TYPE_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_PAYMENT_TYPE_24cMo"]
)

# Distribution of the total AMT_PAYMENTs of Installments, segmented by # PriorApplication_PAYMENT_TYPE for the Client over time. client_installments_amt_payments_by_priorapplication_payment_type_24cmo = ( installments_payments_view_by_client_across_priorapplication_payment_types.aggregate_over( "AMT_PAYMENT", method="sum", feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_PAYMENT_TYPE_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_PAYMENT_TYPE_24cMo"] )

In [22]:

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different
# PriorApplication_CLIENT_TYPEs.
installments_payments_view_by_client_across_priorapplication_client_types = installments_payments_view.groupby(
    ["PriorApplication_ClientID"], category="PriorApplication_CLIENT_TYPE"
)

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different # PriorApplication_CLIENT_TYPEs. installments_payments_view_by_client_across_priorapplication_client_types = installments_payments_view.groupby( ["PriorApplication_ClientID"], category="PriorApplication_CLIENT_TYPE" )

In [23]:

# Distribution of the total AMT_PAYMENTs of Installments, segmented by PriorApplication_CLIENT_TYPE
# for the Client over time.
client_installments_amt_payments_by_priorapplication_client_type_24cmo = (
    installments_payments_view_by_client_across_priorapplication_client_types.aggregate_over(
        "AMT_PAYMENT",
        method="sum",
        feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_CLIENT_TYPE_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_CLIENT_TYPE_24cMo"]
)

# Distribution of the total AMT_PAYMENTs of Installments, segmented by PriorApplication_CLIENT_TYPE # for the Client over time. client_installments_amt_payments_by_priorapplication_client_type_24cmo = ( installments_payments_view_by_client_across_priorapplication_client_types.aggregate_over( "AMT_PAYMENT", method="sum", feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_CLIENT_TYPE_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_CLIENT_TYPE_24cMo"] )

In [24]:

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different
# PriorApplication_YIELD_GROUPs.
installments_payments_view_by_client_across_priorapplication_yield_groups = installments_payments_view.groupby(
    ["PriorApplication_ClientID"], category="PriorApplication_YIELD_GROUP"
)

# Group INSTALLMENTS_PAYMENTS view by Client entity (PriorApplication_ClientID) across different # PriorApplication_YIELD_GROUPs. installments_payments_view_by_client_across_priorapplication_yield_groups = installments_payments_view.groupby( ["PriorApplication_ClientID"], category="PriorApplication_YIELD_GROUP" )

In [25]:

# Distribution of the total AMT_PAYMENTs of Installments, segmented by PriorApplication_YIELD_GROUP
# for the Client over time.
client_installments_amt_payments_by_priorapplication_yield_group_24cmo = (
    installments_payments_view_by_client_across_priorapplication_yield_groups.aggregate_over(
        "AMT_PAYMENT",
        method="sum",
        feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_YIELD_GROUP_24cMo"],
        windows=[fb.CalendarWindow(unit="MONTH", size=24)],
    )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_YIELD_GROUP_24cMo"]
)

# Distribution of the total AMT_PAYMENTs of Installments, segmented by PriorApplication_YIELD_GROUP # for the Client over time. client_installments_amt_payments_by_priorapplication_yield_group_24cmo = ( installments_payments_view_by_client_across_priorapplication_yield_groups.aggregate_over( "AMT_PAYMENT", method="sum", feature_names=["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_YIELD_GROUP_24cMo"], windows=[fb.CalendarWindow(unit="MONTH", size=24)], )["CLIENT_Installments_AMT_PAYMENTs_by_PriorApplication_YIELD_GROUP_24cMo"] )

Derive feature from aggregation across categories¶

See list of feature transforms

In [26]:

# Get Entropy of count of Installments by INSTALLMENT_STATUS from
# CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo
client_entropy_of_count_of_installments_by_installment_status_24cmo = (
    client_count_of_installments_by_installment_status_24cmo.cd.entropy()
)

# Give a name to new feature
client_entropy_of_count_of_installments_by_installment_status_24cmo.name = (
    "CLIENT_Entropy_of_count_of_Installments_by_INSTALLMENT_STATUS_24cMo"
)

# Get Entropy of count of Installments by INSTALLMENT_STATUS from # CLIENT_Count_of_Installments_by_INSTALLMENT_STATUS_24cMo client_entropy_of_count_of_installments_by_installment_status_24cmo = ( client_count_of_installments_by_installment_status_24cmo.cd.entropy() ) # Give a name to new feature client_entropy_of_count_of_installments_by_installment_status_24cmo.name = ( "CLIENT_Entropy_of_count_of_Installments_by_INSTALLMENT_STATUS_24cMo" )

Save features¶

See SDK reference to save feature

See SDK reference to save feature group

In [27]:

fb.FeatureGroup(
    [
        client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo,
        client_min_of_credit_card_monthly_balance_records_available_credits_6cmo,
        client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo,
        client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo,
        client_min_of_installments_amt_payments_24cmo,
        client_installments_amt_payments_by_priorapplication_yield_group_24cmo,
        client_installments_amt_payments_by_priorapplication_payment_type_24cmo,
        client_installments_amt_payments_by_priorapplication_client_type_24cmo,
        client_installments_amt_payments_by_installment_status_6cmo,
        client_installments_amt_payments_by_installment_status_12cmo,
        client_entropy_of_count_of_installments_by_installment_status_24cmo,
    ]
).save()

fb.FeatureGroup( [ client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo, client_min_of_credit_card_monthly_balance_records_available_credits_6cmo, client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo, client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo, client_min_of_installments_amt_payments_24cmo, client_installments_amt_payments_by_priorapplication_yield_group_24cmo, client_installments_amt_payments_by_priorapplication_payment_type_24cmo, client_installments_amt_payments_by_priorapplication_client_type_24cmo, client_installments_amt_payments_by_installment_status_6cmo, client_installments_amt_payments_by_installment_status_12cmo, client_entropy_of_count_of_installments_by_installment_status_24cmo, ] ).save()

Done! |████████████████████████████████████████| 100% in 9.1s (0.11%/s)         
Done! |████████████████████████████████████████| 100% in 6.2s (0.16%/s)         
Loading Feature(s) |████████████████████████████████████████| 11/11 [100%] in 0.

Update feature type¶

In [28]:

# Update feature type
client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo.update_feature_type("numeric")

# Update feature type client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo.update_feature_type("numeric")

In [29]:

# Update feature type
client_min_of_credit_card_monthly_balance_records_available_credits_6cmo.update_feature_type("numeric")

# Update feature type client_min_of_credit_card_monthly_balance_records_available_credits_6cmo.update_feature_type("numeric")

In [30]:

# Update feature type
client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo.update_feature_type("numeric")

# Update feature type client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo.update_feature_type("numeric")

In [31]:

# Update feature type
client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo.update_feature_type("numeric")

# Update feature type client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo.update_feature_type("numeric")

In [32]:

# Update feature type
client_min_of_installments_amt_payments_24cmo.update_feature_type("numeric")

# Update feature type client_min_of_installments_amt_payments_24cmo.update_feature_type("numeric")

In [33]:

# Update feature type
client_installments_amt_payments_by_priorapplication_yield_group_24cmo.update_feature_type("dictionary")

# Update feature type client_installments_amt_payments_by_priorapplication_yield_group_24cmo.update_feature_type("dictionary")

In [34]:

# Update feature type
client_installments_amt_payments_by_priorapplication_payment_type_24cmo.update_feature_type("dictionary")

# Update feature type client_installments_amt_payments_by_priorapplication_payment_type_24cmo.update_feature_type("dictionary")

In [35]:

# Update feature type
client_installments_amt_payments_by_priorapplication_client_type_24cmo.update_feature_type("dictionary")

# Update feature type client_installments_amt_payments_by_priorapplication_client_type_24cmo.update_feature_type("dictionary")

In [36]:

# Update feature type
client_installments_amt_payments_by_installment_status_6cmo.update_feature_type("dictionary")

# Update feature type client_installments_amt_payments_by_installment_status_6cmo.update_feature_type("dictionary")

In [37]:

# Update feature type
client_installments_amt_payments_by_installment_status_12cmo.update_feature_type("dictionary")

# Update feature type client_installments_amt_payments_by_installment_status_12cmo.update_feature_type("dictionary")

In [38]:

# Update feature type
client_entropy_of_count_of_installments_by_installment_status_24cmo.update_feature_type("numeric")

# Update feature type client_entropy_of_count_of_installments_by_installment_status_24cmo.update_feature_type("numeric")

Add description¶

In [39]:

# Add description
client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo.update_description(
    "Std of Credit_card_monthly_balance_records CNT_DRAWINGS_ATM_CURRENTs for "
    "the Client over a 24 calendar months period."
)

# Add description client_std_of_credit_card_monthly_balance_records_cnt_drawings_atm_currents_24cmo.update_description( "Std of Credit_card_monthly_balance_records CNT_DRAWINGS_ATM_CURRENTs for " "the Client over a 24 calendar months period." )

In [40]:

# Add description
client_min_of_credit_card_monthly_balance_records_available_credits_6cmo.update_description(
    "Min of Credit_card_monthly_balance_records Available Credits for the " "Client over a 6 calendar months period."
)

# Add description client_min_of_credit_card_monthly_balance_records_available_credits_6cmo.update_description( "Min of Credit_card_monthly_balance_records Available Credits for the " "Client over a 6 calendar months period." )

In [41]:

# Add description
client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo.update_description(
    "Max of Installments PriorApplication_AMT_ANNUITY To AMT_CREDITs for the Client over a 6 calendar months period."
)

# Add description client_max_of_installments_priorapplication_amt_annuity_to_amt_credits_6cmo.update_description( "Max of Installments PriorApplication_AMT_ANNUITY To AMT_CREDITs for the Client over a 6 calendar months period." )

In [42]:

# Add description
client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo.update_description(
    "Max of Installments Days_Difference_Actual_vs_Scheduleds for the Client over a 24 calendar months period."
)

# Add description client_max_of_installments_days_difference_actual_vs_scheduleds_24cmo.update_description( "Max of Installments Days_Difference_Actual_vs_Scheduleds for the Client over a 24 calendar months period." )

In [43]:

# Add description
client_min_of_installments_amt_payments_24cmo.update_description(
    "Min of Installments AMT_PAYMENTs for the Client over a 24 calendar months period."
)

# Add description client_min_of_installments_amt_payments_24cmo.update_description( "Min of Installments AMT_PAYMENTs for the Client over a 24 calendar months period." )

In [44]:

# Add description
client_installments_amt_payments_by_priorapplication_yield_group_24cmo.update_description(
    "Distribution of the total AMT_PAYMENTs of Installments, segmented by "
    "PriorApplication_YIELD_GROUP for the Client over a 24 calendar months "
    "period. It provides a detailed breakdown in the form of a dictionary, "
    "where each key represents a unique PriorApplication_YIELD_GROUP. The "
    "corresponding value for each key is the cumulative sum of AMT_PAYMENTs"
    " within that PriorApplication_YIELD_GROUP. This distribution offers "
    "insights into the AMT_PAYMENT spread across different "
    "PriorApplication_YIELD_GROUPs for the Client."
)

# Add description client_installments_amt_payments_by_priorapplication_yield_group_24cmo.update_description( "Distribution of the total AMT_PAYMENTs of Installments, segmented by " "PriorApplication_YIELD_GROUP for the Client over a 24 calendar months " "period. It provides a detailed breakdown in the form of a dictionary, " "where each key represents a unique PriorApplication_YIELD_GROUP. The " "corresponding value for each key is the cumulative sum of AMT_PAYMENTs" " within that PriorApplication_YIELD_GROUP. This distribution offers " "insights into the AMT_PAYMENT spread across different " "PriorApplication_YIELD_GROUPs for the Client." )

In [45]:

# Add description
client_installments_amt_payments_by_priorapplication_payment_type_24cmo.update_description(
    "Distribution of the total AMT_PAYMENTs of Installments, segmented by "
    "PriorApplication_PAYMENT_TYPE for the Client over a 24 calendar months"
    " period. It provides a detailed breakdown in the form of a dictionary,"
    " where each key represents a unique PriorApplication_PAYMENT_TYPE. The"
    " corresponding value for each key is the cumulative sum of "
    "AMT_PAYMENTs within that PriorApplication_PAYMENT_TYPE. This "
    "distribution offers insights into the AMT_PAYMENT spread across "
    "different PriorApplication_PAYMENT_TYPEs for the Client."
)

# Add description client_installments_amt_payments_by_priorapplication_payment_type_24cmo.update_description( "Distribution of the total AMT_PAYMENTs of Installments, segmented by " "PriorApplication_PAYMENT_TYPE for the Client over a 24 calendar months" " period. It provides a detailed breakdown in the form of a dictionary," " where each key represents a unique PriorApplication_PAYMENT_TYPE. The" " corresponding value for each key is the cumulative sum of " "AMT_PAYMENTs within that PriorApplication_PAYMENT_TYPE. This " "distribution offers insights into the AMT_PAYMENT spread across " "different PriorApplication_PAYMENT_TYPEs for the Client." )

In [46]:

# Add description
client_installments_amt_payments_by_priorapplication_client_type_24cmo.update_description(
    "Distribution of the total AMT_PAYMENTs of Installments, segmented by "
    "PriorApplication_CLIENT_TYPE for the Client over a 24 calendar months "
    "period. It provides a detailed breakdown in the form of a dictionary, "
    "where each key represents a unique PriorApplication_CLIENT_TYPE. The "
    "corresponding value for each key is the cumulative sum of AMT_PAYMENTs"
    " within that PriorApplication_CLIENT_TYPE. This distribution offers "
    "insights into the AMT_PAYMENT spread across different "
    "PriorApplication_CLIENT_TYPEs for the Client."
)

# Add description client_installments_amt_payments_by_priorapplication_client_type_24cmo.update_description( "Distribution of the total AMT_PAYMENTs of Installments, segmented by " "PriorApplication_CLIENT_TYPE for the Client over a 24 calendar months " "period. It provides a detailed breakdown in the form of a dictionary, " "where each key represents a unique PriorApplication_CLIENT_TYPE. The " "corresponding value for each key is the cumulative sum of AMT_PAYMENTs" " within that PriorApplication_CLIENT_TYPE. This distribution offers " "insights into the AMT_PAYMENT spread across different " "PriorApplication_CLIENT_TYPEs for the Client." )

In [47]:

# Add description
client_installments_amt_payments_by_installment_status_6cmo.update_description(
    "Distribution of the total AMT_PAYMENTs of Installments, segmented by "
    "INSTALLMENT_STATUS for the Client over a 6 calendar months period. It "
    "provides a detailed breakdown in the form of a dictionary, where each "
    "key represents a unique INSTALLMENT_STATUS. The corresponding value "
    "for each key is the cumulative sum of AMT_PAYMENTs within that "
    "INSTALLMENT_STATUS. This distribution offers insights into the "
    "AMT_PAYMENT spread across different INSTALLMENT_STATUSs for the "
    "Client."
)

# Add description client_installments_amt_payments_by_installment_status_6cmo.update_description( "Distribution of the total AMT_PAYMENTs of Installments, segmented by " "INSTALLMENT_STATUS for the Client over a 6 calendar months period. It " "provides a detailed breakdown in the form of a dictionary, where each " "key represents a unique INSTALLMENT_STATUS. The corresponding value " "for each key is the cumulative sum of AMT_PAYMENTs within that " "INSTALLMENT_STATUS. This distribution offers insights into the " "AMT_PAYMENT spread across different INSTALLMENT_STATUSs for the " "Client." )

In [48]:

# Add description
client_installments_amt_payments_by_installment_status_12cmo.update_description(
    "Distribution of the total AMT_PAYMENTs of Installments, segmented by "
    "INSTALLMENT_STATUS for the Client over a 12 calendar months period. It"
    " provides a detailed breakdown in the form of a dictionary, where each"
    " key represents a unique INSTALLMENT_STATUS. The corresponding value "
    "for each key is the cumulative sum of AMT_PAYMENTs within that "
    "INSTALLMENT_STATUS. This distribution offers insights into the "
    "AMT_PAYMENT spread across different INSTALLMENT_STATUSs for the "
    "Client."
)

# Add description client_installments_amt_payments_by_installment_status_12cmo.update_description( "Distribution of the total AMT_PAYMENTs of Installments, segmented by " "INSTALLMENT_STATUS for the Client over a 12 calendar months period. It" " provides a detailed breakdown in the form of a dictionary, where each" " key represents a unique INSTALLMENT_STATUS. The corresponding value " "for each key is the cumulative sum of AMT_PAYMENTs within that " "INSTALLMENT_STATUS. This distribution offers insights into the " "AMT_PAYMENT spread across different INSTALLMENT_STATUSs for the " "Client." )

In [49]:

# Add description
client_entropy_of_count_of_installments_by_installment_status_24cmo.update_description(
    "The feature evaluates the distribution of a Client's count of "
    "Installments across various INSTALLMENT_STATUSs over a 24 calendar "
    "months period using entropy. Higher entropy signifies a more uniform "
    "distribution of count of Installments across INSTALLMENT_STATUSs, "
    "while lower entropy suggests a concentration within specific "
    "INSTALLMENT_STATUSs."
)

# Add description client_entropy_of_count_of_installments_by_installment_status_24cmo.update_description( "The feature evaluates the distribution of a Client's count of " "Installments across various INSTALLMENT_STATUSs over a 24 calendar " "months period using entropy. Higher entropy signifies a more uniform " "distribution of count of Installments across INSTALLMENT_STATUSs, " "while lower entropy suggests a concentration within specific " "INSTALLMENT_STATUSs." )