Skip to content

Fixed error from contribution push when model is text generation #1716

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

Merged
merged 2 commits into from
Jan 8, 2024
Merged

Fixed error from contribution push when model is text generation #1716

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2f8f298
Fixed error from contribution push when model is text generation
kevinmessiaen Jan 5, 2024
ce06977
Merge branch 'main' into feature/gsk-2321-push-feature-error-no-attri…
kevinmessiaen Jan 8, 2024
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
7 changes: 6 additions & 1 deletion giskard/push/contribution.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@
- existing_shap_values: Check if SHAP values exist.

"""
from typing import Optional

import numpy as np
import pandas as pd
from scipy.stats import zscore
Expand Down Expand Up @@ -126,7 +128,7 @@ def _create_text_contribution_push(shap_feature, sliced_ds, model, raw_predictio
)


def create_contribution_push(model: BaseModel, ds: Dataset, df: pd.DataFrame) -> ContributionPush:
def create_contribution_push(model: BaseModel, ds: Dataset, df: pd.DataFrame) -> Optional[ContributionPush]:
"""
Create contribution notification from SHAP values.

Expand All @@ -143,6 +145,9 @@ def create_contribution_push(model: BaseModel, ds: Dataset, df: pd.DataFrame) ->
Returns:
ContributionPush if outlier contribution found, else None
"""
if model.is_text_generation:
return None

# Check if there is only one feature type in the dataset, and if it's "text"
_text_is_the_only_feature = len(ds.column_types.values()) == 1 and list(ds.column_types.values())[0] == "text"

Expand Down