Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Documentation for profiles.yaml #19

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Documentation for profiles.yaml #19

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c6963ba
Documentation for profiles.yaml
joker2411 Oct 5, 2023
555096c
adding prep configs in profiles.yaml
joker2411 Oct 5, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 9 additions & 9 deletions models/profiles.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,25 +12,25 @@ models:
file_extension: .json
file_validity: 168h # If the last trained model is older than this, then the model will be trained again,
inputs:
- packages/feature_table/models/shopify_user_features
- packages/feature_table/models/shopify_user_features # inputs to materialise the required data for model
config:
data: &model_data_input_configs
package_name: feature_table # Name of the package where the profiles feature table is defined (declared in pb_project.yaml packages)
label_column: is_churned_7_days
label_value: 1
prediction_horizon_days: 7
features_profiles_model: 'shopify_user_features'
output_profiles_ml_model: *model_name
label_column: is_churned_7_days # Name of column for which we want to find predictions
label_value: 1 # Value of label column for which prediction is generated
prediction_horizon_days: 7 # Number of days in future for which we want to predict
features_profiles_model: 'shopify_user_features' # Model name
output_profiles_ml_model: *model_name # Name of output model based on current model to dinstinguish between multiple models
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  1. Can you add eligible_users and keep that blank string, with comment explaining what it is?
  2. Also add preprocessing: ignore_features: [], again with a comment.
  3. Does the preprocessing get captured in predict if we add here? Is it needed?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ignore_features, timestamp_columns are being used in predict script as well. I, therefore, have passed the same configs to predict as well. Though in future, we can keep both of them to some other key, because these 2 are being used in both train and predict out of all the keys in preprocessing.


predict:
inputs:
- packages/feature_table/models/shopify_user_features
config:
data: *model_data_input_configs
data: *model_data_input_configs # Use the same config as training
outputs:
column_names:
percentile: &percentile_name percentile_churn_score_7_days
score: churn_score_7_days
percentile: &percentile_name percentile_churn_score_7_days # Name of column in output table having percentile score
score: churn_score_7_days # Name of column in output table having probabilistic score
feature_meta_data: &feature_meta_data
features:
- name: *percentile_name
Expand Down