springboard-curriculum/featuretools

Adapted exercise from here: https://github.com/Featuretools/predict-customer-churn/blob/master/churn/3.%20Feature%20Engineering.ipynb

featuretools

Adapted exercise from here: https://github.com/Featuretools/predict-customer-churn/blob/master/churn/3.%20Feature%20Engineering.ipynb

Here is a video showing how to run the tutorial in colab or in a python env on your machine.

Youtube Video

featuretools's Issues

Have an error when 'run deep feature synthesis'

Hi,

I need help with this exercise. I got an error as below.
AttributeError: Cutoff time DataFrame must contain a column with either the same name as the target entity time_index or a column named "time"

The picture shows where I got this error. And I paste the full error message below the picture. Thank you,

Built 881 features
C:\Users\prue1\anaconda3\lib\site-packages\featuretools\synthesis\dfs.py:317: UnusedPrimitiveWarning: Some specified primitives were not used during DFS:
where_primitives: ['any']
This may be caused by a using a value of max_depth that is too small, not setting interesting values, or it may indicate no compatible variable types for the primitive were found in the data.
warnings.warn(warning_msg, UnusedPrimitiveWarning)

AttributeError Traceback (most recent call last)
in
2
3 start = timer()
----> 4 feature_matrix, feature_defs = ft.dfs(entityset=es, target_entity='members',
5 cutoff_time = cutoff_times,
6 agg_primitives = agg_primitives,

~\anaconda3\lib\site-packages\featuretools\utils\entry_point.py in function_wrapper(*args, **kwargs)
38 ep.on_error(error=e,
39 runtime=runtime)
---> 40 raise e
41
42 # send return value

~\anaconda3\lib\site-packages\featuretools\utils\entry_point.py in function_wrapper(*args, **kwargs)
30 # call function
31 start = time.time()
---> 32 return_value = func(*args, **kwargs)
33 runtime = time.time() - start
34 except Exception as e:

~\anaconda3\lib\site-packages\featuretools\synthesis\dfs.py in dfs(entities, relationships, entityset, target_entity, cutoff_time, instance_ids, agg_primitives, trans_primitives, groupby_trans_primitives, allowed_paths, max_depth, ignore_entities, ignore_variables, primitive_options, seed_features, drop_contains, drop_exact, where_primitives, max_features, cutoff_time_in_index, save_progress, features_only, training_window, approximate, chunk_size, n_jobs, dask_kwargs, verbose, return_variable_types, progress_callback, include_cutoff_time)
276 return features
277
--> 278 feature_matrix = calculate_feature_matrix(features,
279 entityset=entityset,
280 cutoff_time=cutoff_time,

~\anaconda3\lib\site-packages\featuretools\computational_backends\calculate_feature_matrix.py in calculate_feature_matrix(features, entityset, cutoff_time, instance_ids, entities, relationships, cutoff_time_in_index, training_window, approximate, save_progress, verbose, chunk_size, n_jobs, dask_kwargs, progress_callback, include_cutoff_time)
158 target_entity = entityset[features[0].entity.id]
159
--> 160 cutoff_time = _validate_cutoff_time(cutoff_time, target_entity)
161
162 if isinstance(cutoff_time, pd.DataFrame):

~\anaconda3\lib\site-packages\featuretools\computational_backends\utils.py in _validate_cutoff_time(cutoff_time, target_entity)
230 if "time" not in cutoff_time.columns:
231 if target_entity.time_index and target_entity.time_index not in cutoff_time.columns:
--> 232 raise AttributeError('Cutoff time DataFrame must contain a column with either the same name'
233 ' as the target entity time_index or a column named "time"')
234 # rename to time

AttributeError: Cutoff time DataFrame must contain a column with either the same name as the target entity time_index or a column named "time"

Featuretools issue

I have some issues related to the exercise in Ch 16.2: "Automated Feature Engineering with Featuretools"
I tried to practice it on the Jupyter notebook but realized that many of the codes in this notebook are not working. I tried but got stuck in the code related to featuretools.primitives and gave up.

Codes:
from featuretools.primitives import make_agg_primitive
from featuretools.primitives import make_trans_primitive

Error messages:
cannot import name 'make_agg_primitive' from 'featuretools.primitives'
cannot import name 'make_trans_primitive' from 'featuretools.primitives'

Feature engineering example

This is a very minor issue with the wording used in the "Domain Knowledge Features" section of the Feature Engineering notebook.

I believe the sentence should indicate there is an anomaly where some memberships expire before the transactions date (instead of after). The following code appears to apply a filter to include only data where memberships expire after the transaction date.

From the notebook:
There is one slight anomaly with the transactions where some membership expire dates are after the transactions date, so we will filter those out.

#Filter anomalies
trans = trans[trans['membership_expire_date'] > trans['transaction_date']]