9. Create Window Aggregates from Event Table

Creating Features from the Event Table¶

In this step, we define features by aggregating events from the PREVIOUS_APPLICATIONS and BUREAU tables.

1. Statistical Aggregations¶

2. Aggregations Across categorical fields¶

3. Features Based on Latest Event Datetime fields and Time to those fields¶

Some aggregations are after filtering the views or creating new columns.

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)

14:09:51 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:09:51 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.
14:09:51 | INFO     | Using profile: tutorial
INFO    :featurebyte:Using profile: tutorial
14:09:51 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
INFO    :featurebyte:Using configuration file at: /Users/gxav/.featurebyte/config.yaml
14:09:51 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
INFO    :featurebyte:Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
14:09:51 | INFO     | SDK version: 3.0.1.dev45
INFO    :featurebyte:SDK version: 3.0.1.dev45
14:09:51 | INFO     | No catalog activated.
INFO    :featurebyte:No catalog activated.
14:09:51 | 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 BUREAU event table.
bureau_view = catalog.get_view("BUREAU")

# Get view from BUREAU event table. bureau_view = catalog.get_view("BUREAU")

In [3]:

# 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 [4]:

bureau_view["AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM"] = (
    bureau_view["AMT_CREDIT_SUM_DEBT"] / bureau_view["AMT_CREDIT_SUM"]
)

bureau_view["AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM"] = ( bureau_view["AMT_CREDIT_SUM_DEBT"] / bureau_view["AMT_CREDIT_SUM"] )

Derive new column¶

In [5]:

bureau_view["End to Update Gap"] = (bureau_view["credit_update"] - bureau_view["credit_end_date"]).dt.minute

bureau_view["End to Update Gap"] = (bureau_view["credit_update"] - bureau_view["credit_end_date"]).dt.minute

In [6]:

bureau_view["Available Credit"] = bureau_view["AMT_CREDIT_SUM"] - bureau_view["AMT_CREDIT_SUM_DEBT"]

bureau_view["Available Credit"] = bureau_view["AMT_CREDIT_SUM"] - bureau_view["AMT_CREDIT_SUM_DEBT"]

Filter View¶

Read on view subsetting

In [7]:

# Filter bureau_view with CREDIT_ACTIVE equal to Active.
cond = bureau_view["CREDIT_ACTIVE"] == "Active"
bureau_active_cr_active_view = bureau_view[cond]

# Filter bureau_view with CREDIT_ACTIVE equal to Active. cond = bureau_view["CREDIT_ACTIVE"] == "Active" bureau_active_cr_active_view = bureau_view[cond]

In [8]:

# Filter previous_application_view with CONTRACT_STATUS equal to Approved.
cond = previous_application_view["CONTRACT_STATUS"] == "Approved"
previous_application_approved_contract_status_view = previous_application_view[cond]

# Filter previous_application_view with CONTRACT_STATUS equal to Approved. cond = previous_application_view["CONTRACT_STATUS"] == "Approved" previous_application_approved_contract_status_view = previous_application_view[cond]

In [9]:

# Filter bureau_view with CREDIT_TYPE equal to Consumer credit.
cond = bureau_view["CREDIT_TYPE"] == "Consumer credit"
bureau_consumer_credit_cr_type_view = bureau_view[cond]

# Filter bureau_view with CREDIT_TYPE equal to Consumer credit. cond = bureau_view["CREDIT_TYPE"] == "Consumer credit" bureau_consumer_credit_cr_type_view = bureau_view[cond]

Do window aggregation from BUREAU¶

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 [10]:

# Group BUREAU view by Client entity (ClientID).
bureau_view_by_client = bureau_view.groupby(["ClientID"])

# Group BUREAU view by Client entity (ClientID). bureau_view_by_client = bureau_view.groupby(["ClientID"])

In [11]:

# Get Avg of Available Credit for the Client over time.
client_avg_of_bureaureportedcredits_available_credits_26w = bureau_view_by_client.aggregate_over(
    "Available Credit",
    method="avg",
    feature_names=["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_26w"],
    windows=["26w"],
)["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_26w"]

# Get Avg of Available Credit for the Client over time. client_avg_of_bureaureportedcredits_available_credits_26w = bureau_view_by_client.aggregate_over( "Available Credit", method="avg", feature_names=["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_26w"], windows=["26w"], )["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_26w"]

In [12]:

# Get Avg of Available Credit for the Client over time.
client_avg_of_bureaureportedcredits_available_credits_104w = bureau_view_by_client.aggregate_over(
    "Available Credit",
    method="avg",
    feature_names=["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_104w"],
    windows=["104w"],
)["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_104w"]

# Get Avg of Available Credit for the Client over time. client_avg_of_bureaureportedcredits_available_credits_104w = bureau_view_by_client.aggregate_over( "Available Credit", method="avg", feature_names=["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_104w"], windows=["104w"], )["CLIENT_Avg_of_BureauReportedCredits_Available_Credits_104w"]

In [13]:

# Get Std of Available Credit for the Client over time.
client_std_of_bureaureportedcredits_available_credits_26w = bureau_view_by_client.aggregate_over(
    "Available Credit",
    method="std",
    feature_names=["CLIENT_Std_of_BureauReportedCredits_Available_Credits_26w"],
    windows=["26w"],
)["CLIENT_Std_of_BureauReportedCredits_Available_Credits_26w"]

# Get Std of Available Credit for the Client over time. client_std_of_bureaureportedcredits_available_credits_26w = bureau_view_by_client.aggregate_over( "Available Credit", method="std", feature_names=["CLIENT_Std_of_BureauReportedCredits_Available_Credits_26w"], windows=["26w"], )["CLIENT_Std_of_BureauReportedCredits_Available_Credits_26w"]

Do window aggregation from PREVIOUS_APPLICATION¶

In [14]:

# Group PREVIOUS_APPLICATION view by Client entity (ClientID).
previous_application_view_by_client = previous_application_view.groupby(["ClientID"])

# Group PREVIOUS_APPLICATION view by Client entity (ClientID). previous_application_view_by_client = previous_application_view.groupby(["ClientID"])

In [15]:

# Get Max of CNT_PAYMENT for the Client over time.
client_max_of_priorapplications_cnt_payments_104w = previous_application_view_by_client.aggregate_over(
    "CNT_PAYMENT",
    method="max",
    feature_names=["CLIENT_Max_of_PriorApplications_CNT_PAYMENTs_104w"],
    windows=["104w"],
)["CLIENT_Max_of_PriorApplications_CNT_PAYMENTs_104w"]

# Get Max of CNT_PAYMENT for the Client over time. client_max_of_priorapplications_cnt_payments_104w = previous_application_view_by_client.aggregate_over( "CNT_PAYMENT", method="max", feature_names=["CLIENT_Max_of_PriorApplications_CNT_PAYMENTs_104w"], windows=["104w"], )["CLIENT_Max_of_PriorApplications_CNT_PAYMENTs_104w"]

In [16]:

# Group PREVIOUS_APPLICATION view by Client entity (ClientID) across different YIELD_GROUPs.
previous_application_view_by_client_across_yield_groups = previous_application_view.groupby(
    ["ClientID"], category="YIELD_GROUP"
)

# Group PREVIOUS_APPLICATION view by Client entity (ClientID) across different YIELD_GROUPs. previous_application_view_by_client_across_yield_groups = previous_application_view.groupby( ["ClientID"], category="YIELD_GROUP" )

In [17]:

# Distribution of the total AMT_CREDITs of PriorApplications, segmented by PriorApplication's
# YIELD_GROUP for the Client over time.
client_priorapplications_amt_credits_by_priorapplication_yield_group_104w = (
    previous_application_view_by_client_across_yield_groups.aggregate_over(
        "AMT_CREDIT",
        method="sum",
        feature_names=["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_YIELD_GROUP_104w"],
        windows=["104w"],
    )["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_YIELD_GROUP_104w"]
)

# Distribution of the total AMT_CREDITs of PriorApplications, segmented by PriorApplication's # YIELD_GROUP for the Client over time. client_priorapplications_amt_credits_by_priorapplication_yield_group_104w = ( previous_application_view_by_client_across_yield_groups.aggregate_over( "AMT_CREDIT", method="sum", feature_names=["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_YIELD_GROUP_104w"], windows=["104w"], )["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_YIELD_GROUP_104w"] )

In [18]:

# Group PREVIOUS_APPLICATION view by Client entity (ClientID) across different
# NFLAG_INSURED_ON_APPROVALs.
previous_application_view_by_client_across_nflag_insured_on_approvals = previous_application_view.groupby(
    ["ClientID"], category="NFLAG_INSURED_ON_APPROVAL"
)

# Group PREVIOUS_APPLICATION view by Client entity (ClientID) across different # NFLAG_INSURED_ON_APPROVALs. previous_application_view_by_client_across_nflag_insured_on_approvals = previous_application_view.groupby( ["ClientID"], category="NFLAG_INSURED_ON_APPROVAL" )

In [19]:

# Distribution of the total AMT_CREDITs of PriorApplications, segmented by PriorApplication's
# NFLAG_INSURED_ON_APPROVAL for the Client over time.
client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w = (
    previous_application_view_by_client_across_nflag_insured_on_approvals.aggregate_over(
        "AMT_CREDIT",
        method="sum",
        feature_names=["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_NFLAG_INSURED_ON_APPROVAL_52w"],
        windows=["52w"],
    )["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_NFLAG_INSURED_ON_APPROVAL_52w"]
)

# Distribution of the total AMT_CREDITs of PriorApplications, segmented by PriorApplication's # NFLAG_INSURED_ON_APPROVAL for the Client over time. client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w = ( previous_application_view_by_client_across_nflag_insured_on_approvals.aggregate_over( "AMT_CREDIT", method="sum", feature_names=["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_NFLAG_INSURED_ON_APPROVAL_52w"], windows=["52w"], )["CLIENT_PriorApplications_AMT_CREDITs_by_PriorApplication_NFLAG_INSURED_ON_APPROVAL_52w"] )

Do window aggregation from bureau_active_cr_active_view¶

In [20]:

# Group bureau_active_cr_active_view view by Client entity (ClientID).
bureau_active_cr_active_view_by_client = bureau_active_cr_active_view.groupby(["ClientID"])

# Group bureau_active_cr_active_view view by Client entity (ClientID). bureau_active_cr_active_view_by_client = bureau_active_cr_active_view.groupby(["ClientID"])

In [21]:

# Get Max of Active Cr_active AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM for the Client over time.
client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w = (
    bureau_active_cr_active_view_by_client.aggregate_over(
        "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM",
        method="max",
        feature_names=[
            "CLIENT_Max_of_Active_Cr_active_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_104w"
        ],
        windows=["104w"],
    )["CLIENT_Max_of_Active_Cr_active_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_104w"]
)

# Get Max of Active Cr_active AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM for the Client over time. client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w = ( bureau_active_cr_active_view_by_client.aggregate_over( "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM", method="max", feature_names=[ "CLIENT_Max_of_Active_Cr_active_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_104w" ], windows=["104w"], )["CLIENT_Max_of_Active_Cr_active_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_104w"] )

Do window aggregation from previous_application_approved_contract_status_view¶

In [22]:

# Group previous_application_approved_contract_status_view view by Client entity (ClientID).
previous_application_approved_contract_status_view_by_client = (
    previous_application_approved_contract_status_view.groupby(["ClientID"])
)

# Group previous_application_approved_contract_status_view view by Client entity (ClientID). previous_application_approved_contract_status_view_by_client = ( previous_application_approved_contract_status_view.groupby(["ClientID"]) )

In [23]:

# Get Latest Approved Contract_status PriorApplication's last_due_1st_version_timestamp for the
# Client over a 104w period.
client_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w = (
    previous_application_approved_contract_status_view_by_client.aggregate_over(
        "last_due_1st_version_timestamp",
        method="latest",
        feature_names=["CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w"],
        windows=["104w"],
    )["CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w"]
)

# Get Latest Approved Contract_status PriorApplication's last_due_1st_version_timestamp for the # Client over a 104w period. client_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w = ( previous_application_approved_contract_status_view_by_client.aggregate_over( "last_due_1st_version_timestamp", method="latest", feature_names=["CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w"], windows=["104w"], )["CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w"] )

Do window aggregation from bureau_consumer_credit_cr_type_view¶

In [24]:

# Group bureau_consumer_credit_cr_type_view view by Client entity (ClientID).
bureau_consumer_credit_cr_type_view_by_client = bureau_consumer_credit_cr_type_view.groupby(["ClientID"])

# Group bureau_consumer_credit_cr_type_view view by Client entity (ClientID). bureau_consumer_credit_cr_type_view_by_client = bureau_consumer_credit_cr_type_view.groupby(["ClientID"])

In [25]:

# Get Avg of Consumer credit Cr_type AMT_CREDIT_SUM for the Client over time.
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w = (
    bureau_consumer_credit_cr_type_view_by_client.aggregate_over(
        "AMT_CREDIT_SUM",
        method="avg",
        feature_names=["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUMs_104w"],
        windows=["104w"],
    )["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUMs_104w"]
)

# Get Avg of Consumer credit Cr_type AMT_CREDIT_SUM for the Client over time. client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w = ( bureau_consumer_credit_cr_type_view_by_client.aggregate_over( "AMT_CREDIT_SUM", method="avg", feature_names=["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUMs_104w"], windows=["104w"], )["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUMs_104w"] )

In [26]:

# Get Max of Consumer credit Cr_type AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM for the Client over
# time.
client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w = (
    bureau_consumer_credit_cr_type_view_by_client.aggregate_over(
        "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM",
        method="max",
        feature_names=[
            "CLIENT_Max_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_52w"
        ],
        windows=["52w"],
    )["CLIENT_Max_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_52w"]
)

# Get Max of Consumer credit Cr_type AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM for the Client over # time. client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w = ( bureau_consumer_credit_cr_type_view_by_client.aggregate_over( "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUM", method="max", feature_names=[ "CLIENT_Max_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_52w" ], windows=["52w"], )["CLIENT_Max_of_Consumer_credit_Cr_type_BureauReportedCredits_AMT_CREDIT_SUM_DEBT_To_AMT_CREDIT_SUMs_52w"] )

In [27]:

# Get Avg of Consumer credit Cr_type End to Update Gap for the Client over time.
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w = (
    bureau_consumer_credit_cr_type_view_by_client.aggregate_over(
        "End to Update Gap",
        method="avg",
        feature_names=["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_End_to_Update_Gaps_104w"],
        windows=["104w"],
    )["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_End_to_Update_Gaps_104w"]
)

# Get Avg of Consumer credit Cr_type End to Update Gap for the Client over time. client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w = ( bureau_consumer_credit_cr_type_view_by_client.aggregate_over( "End to Update Gap", method="avg", feature_names=["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_End_to_Update_Gaps_104w"], windows=["104w"], )["CLIENT_Avg_of_Consumer_credit_Cr_type_BureauReportedCredits_End_to_Update_Gaps_104w"] )

Derive Time Since feature from latest timestamp¶

See SDK reference to derive time since features

In [28]:

# Compare
# CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w with
# point-in-time in days
client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w = (
    fb.RequestColumn.point_in_time()
    - client_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w
).dt.day

# Give a name to new feature
client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.name = (
    "CLIENT_Time_To_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w"
)

# Compare # CLIENT_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w with # point-in-time in days client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w = ( fb.RequestColumn.point_in_time() - client_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w ).dt.day # Give a name to new feature client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.name = ( "CLIENT_Time_To_Latest_Approved_Contract_status_PriorApplication_last_due_1st_version_timestamp_104w" )

Save features¶

See SDK reference to save feature

See SDK reference to save feature group

In [29]:

fb.FeatureGroup(
    [
        client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w,
        client_std_of_bureaureportedcredits_available_credits_26w,
        client_priorapplications_amt_credits_by_priorapplication_yield_group_104w,
        client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w,
        client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w,
        client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w,
        client_max_of_priorapplications_cnt_payments_104w,
        client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w,
        client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w,
        client_avg_of_bureaureportedcredits_available_credits_26w,
        client_avg_of_bureaureportedcredits_available_credits_104w,
    ]
).save()

fb.FeatureGroup( [ client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w, client_std_of_bureaureportedcredits_available_credits_26w, client_priorapplications_amt_credits_by_priorapplication_yield_group_104w, client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w, client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w, client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w, client_max_of_priorapplications_cnt_payments_104w, client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w, client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w, client_avg_of_bureaureportedcredits_available_credits_26w, client_avg_of_bureaureportedcredits_available_credits_104w, ] ).save()

Done! |████████████████████████████████████████| 100% in 9.1s (0.11%/s)         
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Loading Feature(s) |████████████████████████████████████████| 11/11 [100%] in 0.

Update feature type¶

In [30]:

# Update feature type
client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.update_feature_type(
    "numeric"
)

# Update feature type client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.update_feature_type( "numeric" )

In [31]:

# Update feature type
client_std_of_bureaureportedcredits_available_credits_26w.update_feature_type("numeric")

# Update feature type client_std_of_bureaureportedcredits_available_credits_26w.update_feature_type("numeric")

In [32]:

# Update feature type
client_priorapplications_amt_credits_by_priorapplication_yield_group_104w.update_feature_type("dictionary")

# Update feature type client_priorapplications_amt_credits_by_priorapplication_yield_group_104w.update_feature_type("dictionary")

In [33]:

# Update feature type
client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w.update_feature_type("dictionary")

# Update feature type client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w.update_feature_type("dictionary")

In [34]:

# Update feature type
client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w.update_feature_type(
    "numeric"
)

# Update feature type client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w.update_feature_type( "numeric" )

In [35]:

# Update feature type
client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w.update_feature_type(
    "numeric"
)

# Update feature type client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w.update_feature_type( "numeric" )

In [36]:

# Update feature type
client_max_of_priorapplications_cnt_payments_104w.update_feature_type("numeric")

# Update feature type client_max_of_priorapplications_cnt_payments_104w.update_feature_type("numeric")

In [37]:

# Update feature type
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w.update_feature_type("numeric")

# Update feature type client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w.update_feature_type("numeric")

In [38]:

# Update feature type
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w.update_feature_type("numeric")

# Update feature type client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w.update_feature_type("numeric")

In [39]:

# Update feature type
client_avg_of_bureaureportedcredits_available_credits_26w.update_feature_type("numeric")

# Update feature type client_avg_of_bureaureportedcredits_available_credits_26w.update_feature_type("numeric")

In [40]:

# Update feature type
client_avg_of_bureaureportedcredits_available_credits_104w.update_feature_type("numeric")

# Update feature type client_avg_of_bureaureportedcredits_available_credits_104w.update_feature_type("numeric")

Add description¶

In [41]:

# Add description
client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.update_description(
    "Time (in days) To Latest Approved Contract_status PriorApplication's "
    "last_due_1st_version_timestamp for the Client over a 104w period."
)

# Add description client_time_to_latest_approved_contract_status_priorapplication_last_due_1st_version_timestamp_104w.update_description( "Time (in days) To Latest Approved Contract_status PriorApplication's " "last_due_1st_version_timestamp for the Client over a 104w period." )

In [42]:

# Add description
client_std_of_bureaureportedcredits_available_credits_26w.update_description(
    "Std of BureauReportedCredits Available Credits for the Client over a " "26w period."
)

# Add description client_std_of_bureaureportedcredits_available_credits_26w.update_description( "Std of BureauReportedCredits Available Credits for the Client over a " "26w period." )

In [43]:

# Add description
client_priorapplications_amt_credits_by_priorapplication_yield_group_104w.update_description(
    "Distribution of the total AMT_CREDITs of PriorApplications, segmented "
    "by PriorApplication's YIELD_GROUP for the Client over a 104w period. "
    "It provides a detailed breakdown in the form of a dictionary, where "
    "each key represents a unique YIELD_GROUP. The corresponding value for "
    "each key is the cumulative sum of AMT_CREDITs within that YIELD_GROUP."
    " This distribution offers insights into the AMT_CREDIT spread across "
    "different YIELD_GROUPs for the Client."
)

# Add description client_priorapplications_amt_credits_by_priorapplication_yield_group_104w.update_description( "Distribution of the total AMT_CREDITs of PriorApplications, segmented " "by PriorApplication's YIELD_GROUP for the Client over a 104w period. " "It provides a detailed breakdown in the form of a dictionary, where " "each key represents a unique YIELD_GROUP. The corresponding value for " "each key is the cumulative sum of AMT_CREDITs within that YIELD_GROUP." " This distribution offers insights into the AMT_CREDIT spread across " "different YIELD_GROUPs for the Client." )

In [44]:

# Add description
client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w.update_description(
    "Distribution of the total AMT_CREDITs of PriorApplications, segmented "
    "by PriorApplication's NFLAG_INSURED_ON_APPROVAL for the Client over a "
    "52w period. It provides a detailed breakdown in the form of a "
    "dictionary, where each key represents a unique "
    "NFLAG_INSURED_ON_APPROVAL. The corresponding value for each key is the"
    " cumulative sum of AMT_CREDITs within that NFLAG_INSURED_ON_APPROVAL. "
    "This distribution offers insights into the AMT_CREDIT spread across "
    "different NFLAG_INSURED_ON_APPROVALs for the Client."
)

# Add description client_priorapplications_amt_credits_by_priorapplication_nflag_insured_on_approval_52w.update_description( "Distribution of the total AMT_CREDITs of PriorApplications, segmented " "by PriorApplication's NFLAG_INSURED_ON_APPROVAL for the Client over a " "52w period. It provides a detailed breakdown in the form of a " "dictionary, where each key represents a unique " "NFLAG_INSURED_ON_APPROVAL. The corresponding value for each key is the" " cumulative sum of AMT_CREDITs within that NFLAG_INSURED_ON_APPROVAL. " "This distribution offers insights into the AMT_CREDIT spread across " "different NFLAG_INSURED_ON_APPROVALs for the Client." )

In [45]:

# Add description
client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w.update_description(
    "Max of Consumer credit Cr_type BureauReportedCredits "
    "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUMs for the Client over a 52w period."
)

# Add description client_max_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_52w.update_description( "Max of Consumer credit Cr_type BureauReportedCredits " "AMT_CREDIT_SUM_DEBT To AMT_CREDIT_SUMs for the Client over a 52w period." )

In [46]:

# Add description
client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w.update_description(
    "Max of Active Cr_active BureauReportedCredits AMT_CREDIT_SUM_DEBT To "
    "AMT_CREDIT_SUMs for the Client over a 104w period."
)

# Add description client_max_of_active_cr_active_bureaureportedcredits_amt_credit_sum_debt_to_amt_credit_sums_104w.update_description( "Max of Active Cr_active BureauReportedCredits AMT_CREDIT_SUM_DEBT To " "AMT_CREDIT_SUMs for the Client over a 104w period." )

In [47]:

# Add description
client_max_of_priorapplications_cnt_payments_104w.update_description(
    "Max of PriorApplications CNT_PAYMENTs for the Client over a 104w period."
)

# Add description client_max_of_priorapplications_cnt_payments_104w.update_description( "Max of PriorApplications CNT_PAYMENTs for the Client over a 104w period." )

In [48]:

# Add description
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w.update_description(
    "Avg of Consumer credit Cr_type BureauReportedCredits End to Update Gaps for the Client over a 104w period."
)

# Add description client_avg_of_consumer_credit_cr_type_bureaureportedcredits_end_to_update_gaps_104w.update_description( "Avg of Consumer credit Cr_type BureauReportedCredits End to Update Gaps for the Client over a 104w period." )

In [49]:

# Add description
client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w.update_description(
    "Avg of Consumer credit Cr_type BureauReportedCredits AMT_CREDIT_SUMs for the Client over a 104w period."
)

# Add description client_avg_of_consumer_credit_cr_type_bureaureportedcredits_amt_credit_sums_104w.update_description( "Avg of Consumer credit Cr_type BureauReportedCredits AMT_CREDIT_SUMs for the Client over a 104w period." )

In [50]:

# Add description
client_avg_of_bureaureportedcredits_available_credits_26w.update_description(
    "Avg of BureauReportedCredits Available Credits for the Client over a 26w period."
)

# Add description client_avg_of_bureaureportedcredits_available_credits_26w.update_description( "Avg of BureauReportedCredits Available Credits for the Client over a 26w period." )

In [51]:

# Add description
client_avg_of_bureaureportedcredits_available_credits_104w.update_description(
    "Avg of BureauReportedCredits Available Credits for the Client over a 104w period."
)

# Add description client_avg_of_bureaureportedcredits_available_credits_104w.update_description( "Avg of BureauReportedCredits Available Credits for the Client over a 104w period." )