Skip to content

tea_tasting.datasets #

Example datasets.

make_users_data(*, covariates=False, seed=None, n_users=4000, ratio=1, sessions_uplift=0.0, orders_uplift=0.1, revenue_uplift=0.1, avg_sessions=2, avg_orders_per_session=0.25, avg_revenue_per_order=10, to_ibis=False) #

Generate simulated data for A/B testing scenarios.

Data mimics what you might encounter in an A/B test for an online store, with a user-level randomization. Each row represents an individual user with information about:

  • user: User identifier.
  • variant: Variant of the test. 0 is control, 1 is treatment.
  • sessions: Number of user's sessions.
  • orders: Number of user's orders.
  • revenue: Revenue generated by the user.

Optionally, pre-experimental data can be generated as well:

  • sessions_covariate: Number of user's sessions before the experiment.
  • orders_covariate: Number of user's orders before the experiment.
  • revenue_covariate: Revenue generated by the user before the experiment.

Parameters:

Name Type Description Default
covariates bool

If True, generates pre-experimental data as the covariates in addition to default columns.

 False
seed int | Generator | SeedSequence | None

Random seed.

 None
n_users int

Number of users.

 4000
ratio float | int

Ratio of the number of users in treatment relative to control.

 1
sessions_uplift float | int

Sessions uplift in the treatment variant, relative to control.

 0.0
orders_uplift float

Orders uplift in the treatment variant, relative to control.

 0.1
revenue_uplift float

Revenue uplift in the treatment variant, relative to control.

 0.1
avg_sessions float | int

Average number of sessions per user.

 2
avg_orders_per_session float

Average number of orders per session. Should be less than 1.

 0.25
avg_revenue_per_order float | int

Average revenue per order.

 10
to_ibis bool

If set to True, returns an Ibis Table; otherwise, returns a Pandas DataFrame.

 False

Returns:

Type Description
Table | DataFrame

Simulated data for A/B testing scenarios.

Examples:

import tea_tasting as tt


data = tt.make_users_data(seed=42)
data
#>       user  variant  sessions  orders    revenue
#> 0        0        1         2       1   9.166147
#> 1        1        0         2       1   6.434079
#> 2        2        1         2       1   7.943873
#> 3        3        1         2       1  15.928675
#> 4        4        0         1       1   7.136917
#> ...    ...      ...       ...     ...        ...
#> 3995  3995        0         2       0   0.000000
#> 3996  3996        0         2       0   0.000000
#> 3997  3997        0         3       0   0.000000
#> 3998  3998        0         1       0   0.000000
#> 3999  3999        0         5       2  17.162459
#>
#> [4000 rows x 5 columns]

With covariates:

data = tt.make_users_data(seed=42, covariates=True)
data
#>       user  variant  sessions  orders    revenue  sessions_covariate  orders_covariate  revenue_covariate
#> 0        0        1         2       1   9.166147                   3                 2          19.191712
#> 1        1        0         2       1   6.434079                   4                 1           2.770749
#> 2        2        1         2       1   7.943873                   4                 2          22.568422
#> 3        3        1         2       1  15.928675                   1                 0           0.000000
#> 4        4        0         1       1   7.136917                   1                 1          13.683796
#> ...    ...      ...       ...     ...        ...                 ...               ...                ...
#> 3995  3995        0         2       0   0.000000                   1                 0           0.000000
#> 3996  3996        0         2       0   0.000000                   3                 1          13.517967
#> 3997  3997        0         3       0   0.000000                   2                 0           0.000000
#> 3998  3998        0         1       0   0.000000                   1                 0           0.000000
#> 3999  3999        0         5       2  17.162459                   5                 0           0.000000
#>
#> [4000 rows x 8 columns]
Source code in src/tea_tasting/datasets.py 
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
def make_users_data(
    *,
    covariates: bool = False,
    seed: int | np.random.Generator | np.random.SeedSequence | None = None,
    n_users: int = 4000,
    ratio: float | int = 1,
    sessions_uplift: float | int = 0.0,
    orders_uplift: float = 0.1,
    revenue_uplift: float = 0.1,
    avg_sessions: float | int = 2,
    avg_orders_per_session: float = 0.25,
    avg_revenue_per_order: float | int = 10,
    to_ibis: bool = False,
) -> ibis.expr.types.Table | pd.DataFrame:
    """Generate simulated data for A/B testing scenarios.

    Data mimics what you might encounter in an A/B test for an online store,
    with a user-level randomization. Each row represents an individual user
    with information about:

    - `user`: User identifier.
    - `variant`: Variant of the test. 0 is control, 1 is treatment.
    - `sessions`: Number of user's sessions.
    - `orders`: Number of user's orders.
    - `revenue`: Revenue generated by the user.

    Optionally, pre-experimental data can be generated as well:

    - `sessions_covariate`: Number of user's sessions
        before the experiment.
    - `orders_covariate`: Number of user's orders before the experiment.
    - `revenue_covariate`: Revenue generated by the user
        before the experiment.

    Args:
        covariates: If `True`, generates pre-experimental data as the covariates
            in addition to default columns.
        seed: Random seed.
        n_users: Number of users.
        ratio: Ratio of the number of users in treatment relative to control.
        sessions_uplift: Sessions uplift in the treatment variant, relative to control.
        orders_uplift: Orders uplift in the treatment variant, relative to control.
        revenue_uplift: Revenue uplift in the treatment variant, relative to control.
        avg_sessions: Average number of sessions per user.
        avg_orders_per_session: Average number of orders per session.
            Should be less than `1`.
        avg_revenue_per_order: Average revenue per order.
        to_ibis: If set to `True`, returns an Ibis Table; otherwise,
            returns a Pandas DataFrame.

    Returns:
        Simulated data for A/B testing scenarios.

    Examples:
        ```python
        import tea_tasting as tt


        data = tt.make_users_data(seed=42)
        data
        #>       user  variant  sessions  orders    revenue
        #> 0        0        1         2       1   9.166147
        #> 1        1        0         2       1   6.434079
        #> 2        2        1         2       1   7.943873
        #> 3        3        1         2       1  15.928675
        #> 4        4        0         1       1   7.136917
        #> ...    ...      ...       ...     ...        ...
        #> 3995  3995        0         2       0   0.000000
        #> 3996  3996        0         2       0   0.000000
        #> 3997  3997        0         3       0   0.000000
        #> 3998  3998        0         1       0   0.000000
        #> 3999  3999        0         5       2  17.162459
        #>
        #> [4000 rows x 5 columns]
        ```

        With covariates:

        ```python
        data = tt.make_users_data(seed=42, covariates=True)
        data
        #>       user  variant  sessions  orders    revenue  sessions_covariate  orders_covariate  revenue_covariate
        #> 0        0        1         2       1   9.166147                   3                 2          19.191712
        #> 1        1        0         2       1   6.434079                   4                 1           2.770749
        #> 2        2        1         2       1   7.943873                   4                 2          22.568422
        #> 3        3        1         2       1  15.928675                   1                 0           0.000000
        #> 4        4        0         1       1   7.136917                   1                 1          13.683796
        #> ...    ...      ...       ...     ...        ...                 ...               ...                ...
        #> 3995  3995        0         2       0   0.000000                   1                 0           0.000000
        #> 3996  3996        0         2       0   0.000000                   3                 1          13.517967
        #> 3997  3997        0         3       0   0.000000                   2                 0           0.000000
        #> 3998  3998        0         1       0   0.000000                   1                 0           0.000000
        #> 3999  3999        0         5       2  17.162459                   5                 0           0.000000
        #>
        #> [4000 rows x 8 columns]
        ```
    """  # noqa: E501
    return _make_data(
        covariates=covariates,
        seed=seed,
        n_users=n_users,
        ratio=ratio,
        sessions_uplift=sessions_uplift,
        orders_uplift=orders_uplift,
        revenue_uplift=revenue_uplift,
        avg_sessions=avg_sessions,
        avg_orders_per_session=avg_orders_per_session,
        avg_revenue_per_order=avg_revenue_per_order,
        to_ibis=to_ibis,
        explode_sessions=False,
    )

make_sessions_data(*, covariates=False, seed=None, n_users=4000, ratio=1, sessions_uplift=0.0, orders_uplift=0.1, revenue_uplift=0.1, avg_sessions=2, avg_orders_per_session=0.25, avg_revenue_per_order=10, to_ibis=False) #

Generate simulated user data for A/B testing scenarios.

Data mimics what you might encounter in an A/B test for an online store, with a user-level randomization. Each row represents a user's session with information about:

  • user: User identifier.
  • variant: Variant of the test. 0 is control, 1 is treatment.
  • sessions: Number of user's sessions.
  • orders: Number of user's orders.
  • revenue: Revenue generated by the user.

Optionally, pre-experimental data can be generated as well:

  • sessions_covariate: Number of user's sessions before the experiment.
  • orders_covariate: Number of user's orders before the experiment.
  • revenue_covariate: Revenue generated by the user before the experiment.

Parameters:

Name Type Description Default
covariates bool

If True, generates pre-experimental data as the covariates in addition to default columns.

 False
seed int | Generator | SeedSequence | None

Random seed.

 None
n_users int

Number of users.

 4000
ratio float | int

Ratio of the number of users in treatment relative to control.

 1
sessions_uplift float | int

Sessions uplift in the treatment variant, relative to control.

 0.0
orders_uplift float

Orders uplift in the treatment variant, relative to control.

 0.1
revenue_uplift float

Revenue uplift in the treatment variant, relative to control.

 0.1
avg_sessions float | int

Average number of sessions per user.

 2
avg_orders_per_session float

Average number of orders per session. Should be less than 1.

 0.25
avg_revenue_per_order float | int

Average revenue per order.

 10
to_ibis bool

If set to True, returns an Ibis Table; otherwise, returns a Pandas DataFrame.

 False

Returns:

Type Description
Table | DataFrame

Simulated data for A/B testing scenarios.

Examples:

import tea_tasting as tt


data = tt.make_sessions_data(seed=42)
data
#>       user  variant  sessions  orders    revenue
#> 0        0        1         1       1   5.887178
#> 1        0        1         1       1   6.131080
#> 2        1        0         1       1   2.614675
#> 3        1        0         1       1  12.296075
#> 4        2        1         1       1  11.573409
#> ...    ...      ...       ...     ...        ...
#> 7953  3999        0         1       1  23.634941
#> 7954  3999        0         1       0   0.000000
#> 7955  3999        0         1       1   2.396078
#> 7956  3999        0         1       1  24.538111
#> 7957  3999        0         1       0   0.000000
#>
#> [7958 rows x 5 columns]

With covariates:

data = tt.make_sessions_data(seed=42, covariates=True)
data
#>       user  variant  sessions  orders    revenue  sessions_covariate  orders_covariate  revenue_covariate
#> 0        0        1         1       1   5.887178                 1.5               0.5           1.236732
#> 1        0        1         1       1   6.131080                 1.5               0.5           1.236732
#> 2        1        0         1       1   2.614675                 0.0               0.0           0.000000
#> 3        1        0         1       1  12.296075                 0.0               0.0           0.000000
#> 4        2        1         1       1  11.573409                 1.5               1.5          12.324434
#> ...    ...      ...       ...     ...        ...                 ...               ...                ...
#> 7953  3999        0         1       1  23.634941                 0.2               0.0           0.000000
#> 7954  3999        0         1       0   0.000000                 0.2               0.0           0.000000
#> 7955  3999        0         1       1   2.396078                 0.2               0.0           0.000000
#> 7956  3999        0         1       1  24.538111                 0.2               0.0           0.000000
#> 7957  3999        0         1       0   0.000000                 0.2               0.0           0.000000
#>
#> [7958 rows x 8 columns]
Source code in src/tea_tasting/datasets.py 
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
def make_sessions_data(
    *,
    covariates: bool = False,
    seed: int | np.random.Generator | np.random.SeedSequence | None = None,
    n_users: int = 4000,
    ratio: float | int = 1,
    sessions_uplift: float | int = 0.0,
    orders_uplift: float = 0.1,
    revenue_uplift: float = 0.1,
    avg_sessions: float | int = 2,
    avg_orders_per_session: float = 0.25,
    avg_revenue_per_order: float | int = 10,
    to_ibis: bool = False,
) -> ibis.expr.types.Table | pd.DataFrame:
    """Generate simulated user data for A/B testing scenarios.

    Data mimics what you might encounter in an A/B test for an online store,
    with a user-level randomization. Each row represents a user's session
    with information about:

    - `user`: User identifier.
    - `variant`: Variant of the test. 0 is control, 1 is treatment.
    - `sessions`: Number of user's sessions.
    - `orders`: Number of user's orders.
    - `revenue`: Revenue generated by the user.

    Optionally, pre-experimental data can be generated as well:

    - `sessions_covariate`: Number of user's sessions
        before the experiment.
    - `orders_covariate`: Number of user's orders before the experiment.
    - `revenue_covariate`: Revenue generated by the user
        before the experiment.

    Args:
        covariates: If `True`, generates pre-experimental data as the covariates
            in addition to default columns.
        seed: Random seed.
        n_users: Number of users.
        ratio: Ratio of the number of users in treatment relative to control.
        sessions_uplift: Sessions uplift in the treatment variant, relative to control.
        orders_uplift: Orders uplift in the treatment variant, relative to control.
        revenue_uplift: Revenue uplift in the treatment variant, relative to control.
        avg_sessions: Average number of sessions per user.
        avg_orders_per_session: Average number of orders per session.
            Should be less than `1`.
        avg_revenue_per_order: Average revenue per order.
        to_ibis: If set to `True`, returns an Ibis Table; otherwise,
            returns a Pandas DataFrame.

    Returns:
        Simulated data for A/B testing scenarios.

    Examples:
        ```python
        import tea_tasting as tt


        data = tt.make_sessions_data(seed=42)
        data
        #>       user  variant  sessions  orders    revenue
        #> 0        0        1         1       1   5.887178
        #> 1        0        1         1       1   6.131080
        #> 2        1        0         1       1   2.614675
        #> 3        1        0         1       1  12.296075
        #> 4        2        1         1       1  11.573409
        #> ...    ...      ...       ...     ...        ...
        #> 7953  3999        0         1       1  23.634941
        #> 7954  3999        0         1       0   0.000000
        #> 7955  3999        0         1       1   2.396078
        #> 7956  3999        0         1       1  24.538111
        #> 7957  3999        0         1       0   0.000000
        #>
        #> [7958 rows x 5 columns]
        ```

        With covariates:

        ```python
        data = tt.make_sessions_data(seed=42, covariates=True)
        data
        #>       user  variant  sessions  orders    revenue  sessions_covariate  orders_covariate  revenue_covariate
        #> 0        0        1         1       1   5.887178                 1.5               0.5           1.236732
        #> 1        0        1         1       1   6.131080                 1.5               0.5           1.236732
        #> 2        1        0         1       1   2.614675                 0.0               0.0           0.000000
        #> 3        1        0         1       1  12.296075                 0.0               0.0           0.000000
        #> 4        2        1         1       1  11.573409                 1.5               1.5          12.324434
        #> ...    ...      ...       ...     ...        ...                 ...               ...                ...
        #> 7953  3999        0         1       1  23.634941                 0.2               0.0           0.000000
        #> 7954  3999        0         1       0   0.000000                 0.2               0.0           0.000000
        #> 7955  3999        0         1       1   2.396078                 0.2               0.0           0.000000
        #> 7956  3999        0         1       1  24.538111                 0.2               0.0           0.000000
        #> 7957  3999        0         1       0   0.000000                 0.2               0.0           0.000000
        #>
        #> [7958 rows x 8 columns]
        ```
    """  # noqa: E501
    return _make_data(
        covariates=covariates,
        seed=seed,
        n_users=n_users,
        ratio=ratio,
        sessions_uplift=sessions_uplift,
        orders_uplift=orders_uplift,
        revenue_uplift=revenue_uplift,
        avg_sessions=avg_sessions,
        avg_orders_per_session=avg_orders_per_session,
        avg_revenue_per_order=avg_revenue_per_order,
        to_ibis=to_ibis,
        explode_sessions=True,
    )